Ground Stone Analysis A Technological AP

GROUND STONE ANALYSIS
Ground Stone Analysis

A TECHNOLOGICAL APPROACH
Jenny L. Adams
The University of Utah Press

Salt Lake City
Published in conjunction with the

Center for Desert Archaeology
Tucson
© 2002 by The University of Utah Press
All rights reserved
07 06 05 04 03 02
5 4 3 2 1
Library of Congress Cataloging-in-Publication Data

Adams, Jenny L.
Ground stone analysis : a technological approach / Jenny L. Adams.
p. cm.
Includes bibliographical references and index.
ISBN 0-87480-716-6 (alk. paper) — ISBN 0-87480-717-4 (pbk. : alk. paper)
1. Stone implements—Analysis. 2. Archaeology—Methodology.
I. Center for Desert Archaeology (Tucson, Ariz.) II. Title.
CC79.5.S76 A33 2002
930.1'028—dc21 2002003442
Contents
Part 1. A Foundation for Research
1. The Groundwork 1
Determining Function 6
Laying a Foundation for Analysis 9
Classifying Ground Stone 11
2. Grinding Technology and Technological Analysis 17

Design and Manufacture 18
Use 21
Wear 25
Use-Wear Analysis 27
Kinetics 41
Disuse, or the Afterlife 42
Data Collection 43
Answering Research Questions 46
Analysis Strategy 57
3. Resources for Modeling Tool Use and Technological Behavior 59

Photographs 60
Ethnography 60
Experimental Replication 62
Designing Experimental Research 65
Part 2. Artifact Descriptions 73
4. Abrading, Smoothing, and Polishing Tools 77

Abraders and Smoothers 79
Polishers 91
Hide-Processing Stones 96
5. Grinding and Pulverizing Tools 98

Manos and Metates 99
Mortars 127
Pitted and Cupped Stones 136
vi C ONTENTS
Pestles 138
Handstones 142
Netherstones 143
Grinding Slabs 145
Lapstones 145
Palettes 146
6. Percussion Tools 151

Hammerstones 151
Pecking Stones 152
Choppers 153
Chisels 153
Crushers and Fergoliths 154
Pottery Anvils 155
Lithic Anvils 157
7. Hafted Percussion Tools 160

Axes 160
Mauls 173
Picks 175
Adzes 176
Mattocks 177
Hoes 177
Tchamahias 179
8. Spinning Tools 180

Fire-Drill Hearths 180
Spindle Bases 182
Whorls 183
9. Perforating, Cutting, and Scraping Tools 184

Awls 184
Reamers 185
Saws and Files 187
Planes 188
Tabular Tools 189
10. Paraphernalia 191

Atlatl Parts 191
Balls 193
Bell Stones 195
Cruciforms 196
Cylinders and Medicine Stones 198
C ONTENTS vii
Disks 200
Lightning Stones 204
Plummets and Weights 204
Pipes and Tubes 205
Shaped Stones 208
Pigments 209
Natural Stones 211
Personal Ornaments 212
Figurines 216
11. Containers and Container Closures 218

Bowls 218
Trays 221
Censers 223
Closures 223
12. Structural Stones 226

Loomblocks 227
Cooking Stones 227
Firedogs and Trivets 231
Fire-Cracked Rock 231
Appendix A: General Artifact Form 233

Appendix B: Handstone Form 243
Appendix C: Netherstone Form 249
Appendix D: Hafted Tool Form 255
Appendix E: Grooved Artifact Form 261
Appendix F: Perforated Artifact Form 265
Glossary 269
References Cited 275
List of Figures 297
List of Tables 300
Acknowledgments 301
Index 303
Part 1
A Foundation for Research

Chapter 1
The Groundwork
B efore plunging into the heart of ground stone research, it is impera-

tive to define what is considered ground stone. A broadly constructed
definition includes any stone item that is primarily manufactured through
mechanisms of abrasion, polish, or impaction, or is itself used to grind,
abrade, polish, or impact (these mechanisms are defined more completely
in Chapter 2). This should leave the rest of the stone-artifact universe to
be classified as flaked lithics. However, the categories of flaked stone and
ground stone are artificial constructs that have meaning only to archaeol-
ogists. Many artifacts are in a fuzzy set between flaked and ground stone,
including cores that were ground to prepare an area for efficient flake re-
moval, and axes that were either ground or flaked to shape or resharpen.
The mechanics of grinding and flaking were probably not as mutually
exclusive to prehistoric technologists as they are to lithic analysts. From
the perspective of a technician asked to analyze an assemblage, the cate-
gory ground stone has come to include: (1) processing tools such as manos,
metates, mortars, and pestles that are used to reduce intermediate sub-
stances (vegetal and animal products, pigments, clays, and tempers) to a
finer texture; (2) manufacturing tools such as abraders, polishing stones,
choppers, and hammerstones that are used to shape other items; (3) tools
and paraphernalia shaped by abrasion, polishing, or impaction, such as
axes, pipes, containers, figurines, personal ornaments, and architectural
pieces; (4) minerals ground for pigments and the processed pigments them-
selves; and (5) pecking stones and hammerstones, because they are used
to manufacture both ground and flaked items.
Why is it important to classify and analyze ground stone artifacts? If
this category is an artificial construct, how can it contribute to understand-
ing prehistoric behavior? And how are inferences about prehistoric behav-
ior formulated? These are the questions that prompted the creation of this
manual, which must take on the thorny issues of classification and jar-
gon confusions. What became obvious to me as I was organizing artifact
1
2 A F OUNDATION FOR R ESEARCH
descriptions is that sometimes the items pushing against the fuzzy bound-
aries and causing the most classification confusions are the most interest-
ing and informative. Furthermore, jargon and the inconsistent use of terms
have historically created classification tangles that often obfuscate the most
archaeologically interesting and pertinent data patterns. Patterns are, after
all, what help us distinguish attributes relative to specific activities, gen-
der, group affiliation, environment, or any other pertinent research do-
main. Even though the U.S. Southwest is the areal focus of this manual,
the analytical methods are applicable anywhere because they are struc-
tured around life-history factors: design and manufacture, primary and
secondary use, and disuse. These are considered here the measurable and
describable attributes of technology. By categorizing an item’s life history
and collectively summarizing the life histories of all ground stone at a
settlement, we can interpret the technological traditions and developments
of the people who lived there.
Every item goes through the same general design and manufacture
process: (1) somebody selects the stone for a specific item; (2) someone
decides if the stone needs to be shaped or otherwise altered before it is
used (the manufacturer may or may not be the same person who chose
the stone); (3) if it does need alteration, then the required attributes are
manufactured or the appropriate modifications are made (the user may or
may not have direct input into the nature of these features). After design
and manufacture, the life history of an item can continue: (4) someone
uses it (the user may or may not be the same person as the manufacturer),
and sometimes, (5) the item is secondarily used (the secondary user may
or may not be the same person as the primary user). At any time during
its life history an item can enter the archaeological record; upon its recov-
ery, the analyst must recognize at what point in its life history that hap-
pened. By structuring analysis techniques to recognize the various attrib-
utes associated with the different points in an item’s life history, it becomes
possible to separate designed attributes from those generated by use, reuse,
and disuse.
This manual is divided into two parts. Part 2 is the working part, in-
cluding definitions and descriptions of types, and emphasizing attributes
most suited to conducting a technological analysis. Whenever possible
the descriptions include the historical context for the use of terms, as well
as other terms that have been applied to the same types. The goal is to un-
ravel some of the complications that have arisen from conflicting applica-
tions of terms. Although various sources of information are compiled to
define technologically derived type definitions, definitions are not enough.
It is time to bring together all that has been learned about specific types
and present the range of possible interactions between prehistoric human
1. T HE G ROUNDWORK 3
agents and their things. In the process, specific issues in need of further
research are highlighted.
Before getting into the heart of the manual, however, Chapter 1 con-
tinues with discussions about determining function and ways to classify
items. Chapter 2 defines and describes the attributes of grinding technol-
ogy. The concepts most pertinent to use-wear analysis are introduced. Sev-
eral suggestions are presented for research that can be conducted with data
derived from technological analyses. Some of the research ideas have been
worked on and modified to compensate for previously unforeseen prob-
lems. Others are merely hypotheses that need testing with archaeological,
ethnographic, or experimental data. Chapter 3 explores some of the re-
sources useful for modeling ground stone tool use. Suggestions are made
for how to exploit ethnographic and experimental resources in develop-
ing criteria for analogues to be compared with patterns in the archaeolog-
ical record, and guidelines are presented for designing experiments that
answer archaeologically derived questions.
Many site locations are referred to throughout this manual, but because
it is impossible to create a legible map including all of them, Figure 1.1 il-
lustrates the general areas and districts where sites are located and includes
a list of individual sites within each larger geographic division. Appendixes
A–F define and code the attributes recorded for specific types and subtypes,
and include suggested forms for recording data. Also useful is the “Refer-
ences Cited” section. Although this manual compiles and summarizes pre-
vious research, and supplements it with research from current projects, in
no way does it cover everything there is to know about ground stone from
all sites in the Southwest. There is so much ongoing research and contract
work that a compilation of it would be obsolete well before publication.
The purpose of this manual is to present a flexible, yet carefully struc-
tured method for analyzing stone artifacts that can be manipulated to con-
tribute valuable answers to a broad range of research questions. The tech-
niques are classificatory in the sense that meaningful categories are created
into which items can be sorted (Adams and Adams 1991:47). The nature
of the categories reflects the research questions to be addressed with the
data. Analysis techniques record important attributes in a consistent man-
ner and allow for comparisons at any desired scale. More than with pre-
vious analysis schemes, this approach emphasizes technology as it is re-
flected by an item’s life history, and draws from the body of literature that
foregrounds the importance of individuals as social agents who make cul-
turally constituted choices concerning design, use, and disuse (see, for
example, Dobres and Hoffman 1999b). This does not mean, however, that
this approach is strictly structured by a particular theoretical orientation
such as social agency or behavioral theory.
1.1. Map of the U.S. Southwest indicating areas and districts mentioned
throughout the manual.
Four Corners Northern New Mexico Wupatki

San Juan River Drainage Bandelier New Caves
Aztec Glasscock Grand Falls
Talus Village Prieta Vista NA11,237
Mesa Verde Area Chama Valley NA11,535
Badger House Leaf Water Big Hawk Valley
Big Juniper House Po-shu-oinga
Long House Te’ewi Northeast Arizona
Mug House White Dog Cave
Spruce Tree House Northern Arizona Colorado Plateau
Duckfoot Flagstaff Area Hopi/Jeddito Area
Alkali Ridge Elden Pueblo Walpi
Prayer Rock District Ridge Ruin Awatovi
La Plata District Walnut Canyon Jeddito 264
Chaco Area Winona Village Homol’ovi Area
Pueblo Bonito Kinnikinnick Petrified Forest Area
Pueblo del Arroyo Lizard Man Village Flattop
Shabik’eshchee Pittsburg Fort Stone Axe
Colorado Juniper Terrace Puerco
Williford Deadman’s Fort
Utah Medicine Fort
Gnat Haven Nalakihu
East-central Arizona/ Starkweather Ruin University Ruin

west-central New Mexico Wet Leggett Los Morteros
Canyon Creek SU San Pedro Valley
Foote Canyon Pueblo Apache Creek Second Canyon
Turkey Foot Ridge White Water District
Central Arizona
Twin Bridges White Water Draw
Perry Mesa
Upper Little Colorado Area Double Adobe
Agua Fria
Table Rock Pueblo
Tuzigoot Casas Grandes Area
Hooper Ranch Pueblo
Tonto/Roosevelt Basins Paquimé
Mineral Creek Pueblo
Boatyard
Carter Ranch Mimbres Area
Tres Huerfanos
Showlow Galaz
Meddler Point
Pinedale Swarts
Prickly Saguaro
Bailey Wind Mountain
Granary Row
Forestdale Area
Los Tortugas Southern New Mexico
Bear Ruin
Griffin Wash Bat Cave
Tla kii
Phoenix Basin/
Bluff Central New Mexico/
Salt Gila Area
Kinishba Rio Grande Drainage
Snaketown
Zuni Area Paa-ko
Las Colinas
Village of the Great Zia
Los Muertos
Kivas Pecos
Las Acequias
Kiatuthlanna Forked Lightning
Hawikuh Southern Arizona Pindi
Point of Pines Area Babocomari Village Picuris
Crooked Ridge Village Ventana Cave Gran Quivera
Turkey Creek Pueblo Tucson Basin Las Humanas
Point of Pines Pueblo Santa Cruz Bend Cochiti
W:10:37 Stone Pipe Tonque
W:10:51A,B Las Capas Kua’ua
Pinelawn Area Los Pozos Puaray
Higgins Flat Pueblo Valencia Vieja Puye
Tularosa Cave Gibbon Springs Rito de los Frijoles
Cordova Cave Hodges Cuyamungue
The methods presented in this manual guide quantitative and quali-

tative assessments of artifacts and assemblages. Once recorded, the data
can be filtered through any pertinent theoretical framework. Thus, an ar-
chaeologist researching behavioral differences might select one set of at-
tributes, and an archaeologist researching adaptational strategies might
select another. However, we must use classification techniques that tack
back and forth between analytical and behavioral constructs. An analytical
construct uses the language of description and quantification; for example,
the handstone is 15 cm long, weighs 340 grams, has multiple unidirectional
striations, and has red pigment on the grinding surface. A behavioral con-
struct is inferential and uses the language of action and people; for example,
the stone was used to grind pigment with a reciprocal motion against a
rough netherstone. Qualitative analytical constructs (red pigment) and
behavioral constructs (pigment grinding) have discrete variables that can
be counted, allowing their frequencies in various assemblages to be com-
pared. For example, Assemblage A has 34 handstones, with 10 percent hav-
ing red pigment on their grinding surfaces, while Assemblage B has 30
handstones, with 50 percent having red pigment on their surfaces.
Analytical constructs with quantifiable variables are measurable such
that size and weight can be evaluated as continuous variables. Assemblage
A has a handstone size/weight ratio that is closer to the smaller end of the
scale than are those from Assemblage B, especially among those with pig-
ment. This is an analytical assessment that can be turned into a behavioral
one, thereby creating a basis for making interpretive statements about
specific assemblages. Assuming that larger handstones are more efficient,
there is evidence for more efficient pigment production at the settlement
from which Assemblage B was derived than at the settlement from which
Assemblage A was derived. The assumption can be experimentally tested
to add more credence to the concluding statement. This simple analysis,
however, should raise other questions that require a broader exploration
into the database. Were processed pigments recovered from either settle-
ment? Were painted items recovered? Can one or more recipes that for-
mulated the pigment be analyzed? Were different recipes used to make
pigments that colored baskets, stained hides, or slipped pots? Now we
are beginning to ask questions and conduct analyses that lead from ana-
lyzing things into the area of human agents operating within their cultural
environment.
The foundation for a technological approach to ground stone analysis
is empirical, experimental, and methodological. The recognition that these
concepts are interwoven at an analytical level moves us beyond traditional
methods of classifying ground stone, where form identifies function, and
into a dialectic perspective of technology that is operational in more than
one sociocultural sphere at a time. The system used to classify the physi-
cal aspects of technology should not be so tightly structured that archae-
ologists lose sight of the relationship between things and the people who
used them. The first step toward achieving this is assessing the relationship
of form to function.
DETERMINING FUNCTION
Perhaps the most important general lessons that can be learned from a tech-
nological approach are that form does not always define function, and
that many forms can serve the same function. The issue of determining
function based on form was called into question for me several times dur-
ing a project conducted through the Museum of Northern Arizona (MNA)
at the Hopi village of Walpi (1975–1980). The issue came up at a commu-
nity meeting where some of the recovered items were set out for people to
look at and discuss. What I thought were two mortars prompted a lot of
conversation in Hopi among several people, but no one told me what was
being said. Some time later, a village elder explained that one of the items
was an eagle watering bowl used when the birds were tethered on the
roof during the ceremonial season. I asked about the use of the other item
and was told that it was used to soften meat for old folks who no longer
had teeth for chewing. These two items had basically the same attributes,
each with a basin that was manufactured into a large rock.
Another time it was a Hopi moccasin maker who cast doubt on the
accuracy of interpreting function based on form. During analysis of food-
processing tools recovered from Walpi, I had arranged all the manos ac-
cording to size and number of used surfaces, as I had been taught to do.
The one-hand manos and two-hand manos were on opposite ends of the
table. As I was recording their dimensions and other attributes, Willie Coin,
a Hopi who worked at MNA, visited the analysis room. He looked at the
one-hand manos and said that he had not seen tools like that since he
stopped making moccasins. Intrigued, I asked him how he would have
used manos for making moccasins. He told me that these were not manos,
but were used to remove hair and soften the hides before cutting the moc-
casin pattern.
As an analyst, the questions for me became, What attributes could be
identified to distinguish a water container from a similarly shaped mor-
tar, and how could I distinguish prehistoric hide processing stones from
manos? Searching for the answers set me on a course of experimentation
and library research. I hunted for descriptions or photographs of other
leather workers using handstones to work hides. There must be use-wear
patterns, subtle design differences, or some previously unrecognized at-
tribute that could help distinguish mortars from containers, and manos
from hide-processing stones. It must be possible to define analytical cate-
gories useful for formulating inferences about the behaviors and activities
with which they were used.
The questions are answerable, and the methods for finding answers
involve recognizing and understanding the techniques and attributes as-
sociated with tool design, manufacture, and use. Only a use-wear analysis
can help distinguish manos and hide-processing stones (Adams 1988). The
mortars and water containers are distinguishable by distinctive design
features and by use-wear patterns (all described more fully in Chapter 2).
The most important fact to recognize before making inferences about tool
function is that the concepts of design and form are not synonymous. Form
is a descriptive concept, “the shape and structure of something as distin-
guished from its material,” whereas design is more a plan or process in
which something is devised “for a specific function or end” (Webster’s
New Collegiate Dictionary, 9th ed.). Knowing the shape of something is not
enough to determine its function; however, if we understand the design
process, including why specific shapes were made, then more possibilities
are opened for reconstructing the interactions between people and their
things from the beginning, which is when designs are conceptualized and
translated into physical form.
Questions about how design reflects function may best be assessed
through design theory. Design theory has its greatest applications in the
fields of architecture, engineering, and industry (Horsfall 1987:333). Horsfall
(1987) introduced the use of design theory for evaluating ground stone
through her analysis of metate manufacture and use in the Guatemalan
Highlands. As restructured for ground stone analysis, design theory as-
sumes that tools are made to solve problems deriving from functional,
economic, or other realms. The designed differences in form are sometimes
brought about by sociocultural constraints such as economy of production,
durability, and efficiency. Cost of production issues, such as distance to
material source and difficulty of manufacture, often dictate choices of de-
sign specifications. The prioritization of choices reflects the sociocultural
context of the relevant group making the choices (Horsfall 1987:334). To
varying degrees, these disparate sociocultural constructs govern such is-
sues as why metates are not always made of the most durable material,
why there are various techniques for constructing an angled grinding sur-
face, and why broken tools continue in use as food-grinding tools among
other things. Design theory allows for the uncertainty of knowing the best
possible choice, and assumes that tool design is often compromised by
making choices that create satisfactory, rather than optimum, tools for the
job at hand (Horsfall 1987:335).
Horsfall (1987:369) stresses that morphological variation should not
be viewed as the primary way to determine function. Furthermore, multi-
ple morphological solutions are possible for performing similar functions.
These solutions may be delimited by sociocultural standards that can be
met through group-specific technological traditions. An example of this
can be found in modern technology. For example, certain standards must
be met by various mail handlers around the world, but it is not hard to
distinguish American addresses from British or French addresses on the

basis of the nature and layout of the information imparted in the addresses.
This information embodies how different sociocultural groups subdivide
their space, leaving the post offices servicing each group to accommodate
these differences. The information and physical pieces of mail can cross
cultural boundaries because of common standards, but the group tradi-
tions of naming their respective spaces remain distinct.
Design theory contributes to ground stone studies in two main ways:
it brings to the foreground the assumptions that technological knowledge
is behaviorally expressed in object design and manufacture, and it accom-
modates competing constraints (Horsfall 1987:340–350; Lechtman 1999:223).
The first concept is particularly useful because knowledge of the prin-
ciples that structure technological performance is passed intergeneration-
ally, thereby creating culturally specific traditions identifiable even if their
practitioners move across the landscape. The second concept is important
because it allows us to recognize that people made choices among con-
straints that might not always be obvious. Ultimately, it is apparent that
people’s actions complicate analysis beyond the simple cause-and-effect
relationships an analyst might hope to find for quick quantification and
description purposes.
LAYING A FOUNDATION FOR ANALYSIS
Several sources of information can be used to lay the foundation for con-
ducting a technological analysis of ground stone: (1) ethnography, includ-
ing ethnoarchaeology; (2) experiments; (3) use-wear analyses that build
on experimentation and the science of tribology; (4) classification tech-
niques; and (5) descriptions derived from more than a hundred years of
excavations in various parts of the U.S. Southwest. What should remain
in the foreground of our thinking about ground stone tool analysis is that
what we know about tool use is based on some kind of analogy—if not
from one of the sources mentioned above, then from our own experiences.
There are extensive debates and comments on the uses of analogy (see,
for example, Gould and Watson 1982; Stahl 1993; Wylie 1982, 1985). Some
so-called understandings of how tools were used are purely conjectural,
especially when there is no active model for comparison. With just a little
work, however, we can do better than conjecture.
For ground stone research, ethnographic analogy is most commonly
accomplished initially through a literature search. Reports written by
late-nineteenth- and early-twentieth-century ethnographers who lived or
worked among Native American groups sometimes describe the tools

used in daily and ritual activities. Their work makes it possible to consider
various scenarios concerning how tools might have been used at different
locations in a village; how some tools might not have been needed at every
settlement type; how some might have been more commonly women’s
tools, while others belonged to men; or how certain tools might have been
used for more than one activity. Rarely, it is possible to find homologous
examples of tool use, most likely in archaeological contexts that have di-
rect historical associations with a living population (see Stahl 1993 for a
review of the direct historical approach to analogic reasoning). Occasion-
ally it is possible to talk with people who either used or watched others
use stone tools. In the southwestern United States, Puebloan groups (such
as the Hopi, Zuni, Acoma, Laguna, and Rio Grande Pueblos) and non-
Puebloan groups (such as the Navajo, Apache, Ute, Paiute, Walapai, O’od-
ham, and Maricopa and other Yuman tribes) are the most easily accessi-
ble. This discussion is not meant to imply, however, that the exact answer
can be found in the ethnographic record. Rather, the record should be
mined for a range of possibilities for each investigated artifact or behavior
(discussed further in Chapter 3).
Experiments allow us to use the tools ourselves so we can better under-
stand the kinetics involved and the creation of use-wear patterns. Some of
the experiments referred to in this manual were conducted by me with
the help of numerous University of Arizona graduate students and Earth-
watch volunteers (Adams 1989a, 1989b). Other experiments were conducted
to answer specific questions for the following projects: the Walpi Research
Project (Museum of Northern Arizona); the Homol’ovi Research Program
(Arizona State Museum), the Point of Pines Project (Arizona State Mu-
seum), the Roosevelt Community Development Project (Desert Archaeol-
ogy, Inc.), the I-10 Improvement Project (Desert Archaeology, Inc.), the
Tonto Creek Archaeological Project (Desert Archaeology, Inc.), and the U.S.
89 Archaeological Project (Desert Archaeology, Inc.).
Archaeological context sometimes provides information about items
that are long gone from the ethnographic record. Some items fell out of
use before historic records were kept and have been lost from oral tradi-
tions. However, their repeated discovery in particular contexts, or in as-
sociation with items whose functions are well understood, at least allow
us to make inferences about their uses. Archaeologists and ethnographers
have been working in the U.S. Southwest since the late 1800s. Their descrip-
tions, photographs, maps, and drawings have provided some semblance
of culture history upon which we can continue to draw. The critical factor
in contemporary research is to be mindful of the cultural and political

conditions under which research was conducted, and to make adjust-
ments for the current conditions. Such conditions are considered through-
out this manual, especially in Part 2, where terminological conflicts are
discussed. The point that there is more than one way to arrive at possible
interpretations of how people interacted with their material culture can-
not be overemphasized.
Classification strategies are valuable techniques that analysts can use
to bridge between the analytical and behavioral constructs mentioned
above. When we label items and consider them types, what are we classi-
fying? If we label one set “loaf manos” and another set “rocker manos,”
and then compare the frequencies to determine that loaf manos are more
common among Anasazi assemblages, and rocker manos are more com-
mon among those of the Mogollon—what inferences have been made?
What if loaf manos are new or barely used, and rocker manos are well
used or nearly worn out? Our comparative study has found evidence for
different amounts of wear between two assemblages that may have noth-
ing to do with group affiliation. The warning must be heeded to clearly
recognize what attributes are related to design, manufacture, and use,
and not confuse them into a classification that we want to use to recog-
nize group affiliation. The shape of an artifact today does not always re-
flect its originally designed shape.
CLASSIFYING GROUND STONE
In his classic analysis of stone implements of northeastern Arizona, Wood-

bury (1954:12) discussed how he used classification and the concept of types
to structure the data to address his specific concerns of temporal and cul-
tural variation. He concluded, as did Brew (1974:44) before him, that the
classification of assemblages into types should be done for specific pur-
poses so that different classification schemes are used for different proj-
ects. This attitude toward classification is lauded by Adams and Adams
(1991:157) and expanded upon here.
Woodbury (1954) and Rinaldo (1959; with Martin et al. 1956, 1961,
1964; with Di Peso et al. 1974) have provided the most in-depth discus-
sions of ground stone classification, and their work serves as the foundation
for the discussion presented here. Woodbury (1954:13) first organized the
ground stone assemblage from Awatovi and other Jeddito Valley settle-
ments by function, determined through ethnographic analogy. If function
could not be determined, then descriptive attributes were used to highlight
unusual characteristics (for example, a cupped stone or a notched disk).

Woodbury (1954) summarized the culture-specific attributes and noted
distributions through time and across space. For the Colorado Plateau,
Woodbury’s classification scheme remains useful.
Twenty years after Woodbury’s publication, Rinaldo (Di Peso et al.
1974) presented a classification of the stone items from Casas Grandes us-
ing a similar form/function approach. He expanded on some temporal
and cultural variations previously identified by Woodbury and through
his own research with Martin in the Mogollon region of Arizona (Martin
et al. 1956, 1961, 1964). The identification process for Casas Grandes items
was standardized, with charts and tables used to show structural variation
and computer codes used to keep track of attributes (Di Peso et al. 1974:2–
16). The contribution of both Woodbury and Rinaldo is that their classifi-
cation schemes are clearly defined, making it possible for others to use
them.
In the beginning of this chapter, it was necessary to define the bound-
edness of the set of ground stone. There is obvious vagueness and confu-
sion even at the most inclusive level of classification. Before I present spe-
cific type descriptions, it is important to review the act of classification
and to present the parameters of the techniques used here. The intent is
not to rehash the debate on archaeological classification and typology that
was pervasive in the literature of the 1970s and 1980s (see, for example,
Dunnell 1971, 1986; Hill and Evans 1972; Whallon and Brown 1982). Rather,
the intent is to refocus somewhat the possibilities for classification build-
ing on what has already been learned.
Classifying artifacts is the process of creating categories and sorting
items into them. Analyzing an artifact, as considered here, is the act of ex-
amining a complex item and, on the basis of the relationship of its indi-
vidual elements, deciding to which category or categories it belongs (Adams
and Adams 1991:47). The process of classification should begin with an
assessment of the range of items to be analyzed and the research issues to
be addressed by the classification. Decisions should be made about which
data to include (Dunnell 1986:151; Zimmerman 1996:243). The most im-
portant properties that define what is included in each category must be
identified. Types are the categories into which entities are sorted. An item
can be analyzed using more than one typology and therefore can have more
than one typological label. Because the goal of this technologically based
analysis is to create an understanding of human behavior via a particular
class of stone tools, it is appropriate to create categories to supplement or,
in some cases, replace the traditional form/function-based typology.
The concepts compared here are drawn from very distinct classifica-
tion methods: classic theory, set theory, and fuzzy set theory. The classic
method is Aristotelian and is perhaps the first way we are taught to con-
ceptualize the act of classification (see Lakoff 1987 for cognitive approaches
to classification). It is so ingrained in our thinking that it is not part of our
conscious thought process until someone brings to light how one thing or
another does not fit into the existing categories. Classic classification tech-
niques conceptualize the boundaries between types as clear, and member-
ship as restricted to one, and only one, type (Adams and Adams 1991:47).
Such is the traditional approach to ground stone classification. Prob-
lems arise because of poorly defined categories used inconsistently. For
example, some classification schemes exclude from the mano category small
tools, any tool used against a basin metate, and those not altered from
their natural rock shape; these are instead referred to by the more generic
term handstones (see, for example, Haury 1976:281). When classified this
way, manos are implicitly associated with the processing of maize, and
handstones with more-generic food- and nonfood-processing activities,
but this strategy attaches more meaning to the categories than can be em-
pirically determined. Without other collaborating evidence, we cannot
possibly know if manos of a particular shape were used to grind maize
and those of other shapes were not (Adams 1999). Other mano classifica-
tion schemes create descriptive subtypes such as one-hand, two-hand,
rocker, or cobble, or assemblage specific categories such as Type I, Type
Ia, Type II, Type IIa, and so on, even if Type I is the equivalent of one-
hand manos and Type II the equivalent of two-hand manos. In some clas-
sifications, subtypes and variations such as number of used surfaces or
tool shape are considered more than descriptive and are elevated to typo-
logical distinctions (see, for example, Di Peso et al. 1974:173–202; Haury
1976:281–282; Lancaster 1984:247–248; Plog 1974:139–141; Rinaldo 1959:
229–236). The differential use of such classification schemes makes it im-
possible to make meaningful comparisons among assemblages.
An alternative approach to classification, derived from set theory, is
suggested here as a more productive way to conceptualize the creation of
categories into which ground stone can be sorted. A variation of set theory
was introduced in 1965 by Zadeh, who proposed that inexact concepts can
be realistically represented by “fuzzy sets” (Kandel 1982:22). Fuzzy set
theory is cognitive and relates more to the natural sorting processes that
occur in the human mind. It can be formulated mathematically and has
been simulated in computer models such as those used for medical diag-
nosis (Zimmerman 1996:242). At a certain level, human minds are capable
of recognizing and sorting patterns in nondichotomous, nonprobabilistic

ways. Sorting through large databases requires a structuring process that
reduces large amounts of data into relatively few categories of similar ob-
jects (Kandel 1982:24; Zimmerman 1996:242). Such structuring can be ac-
complished with set theory, which is not too dissimilar from classical clas-
sification methods. Classic, or crisp, set theory places crisply defined objects
into crisply defined classes, creating a finite collection of entities or ob-
jects that can be cumulative and counted (Zimmerman 1996:242). Classic
sets are described by listing the entities that belong to each set, analytically
describing the sets by stating conditions for membership, and defining
the member entities by using a characteristic function: 1 ⫽ member, 0 ⫽
nonmember.
Fuzzy set theory allows for the placement of crisp objects into fuzzy
classes, fuzzy objects into crisp classes, or fuzzy objects into fuzzy classes
(Zimmerman 1996:242–243). Fuzzy data analysis combines many classical
dichotomous methods of pattern recognition, such as clustering and re-
gression techniques, but modified to include situations with fuzzy elements.
For fuzzy sets the characteristic function allows for degree of membership
for each entity in a given set (Kandel 1982:23–24; Zimmerman 1996:11).
Grades of membership reflect ordering with concepts such as generality,
ambiguity, and vagueness, and notions of nearness or continuity in the
universe (Kandel 1982:25).
Fuzziness is frequent in human judgment, evaluation, and decision
making, and is somewhat inherent in our natural languages. Meanings of
words are sometimes vague, especially the meaning of words used to la-
bel sets (Zimmerman 1996:3–4). There are two kinds of fuzziness: intrinsic
fuzziness, which depends on context, and informational fuzziness, which
requires a more complex, descriptive definition of variables that are not
immediately observable and may require multiple measurements (Zim-
merman 1996:3–4). In terms related to ground stone classification, examples
of the two kinds of fuzziness are large manos (intrinsic) and efficient manos
(informational). A large mano is defined relative to other manos in the set
on the basis of the formal property of size. To identify efficient manos, we
would have to define a scale of measurement (perhaps a size/weight ra-
tio) and some measure of product output to distinguish efficient from in-
efficient manos.
Data analysis techniques group items by particular attributes, and the
values assigned to the attributes become the data to be analyzed (Zim-
merman 1996:242). The goal is to find in the data some structure that has
relevance to archaeological and behavioral concerns. The specific purpose
of the typology presented here is to emphasize the technologically impor-

tant features of design, manufacture, and use, but also to build on the func-
tional typologies employed by others. In a sense, there are several typolo-
gies defined in this manual. For example, a typology of design sorts items
into the categories of strategic or expedient design (defined in Chapter 2).
This crosscuts the activity or action-based typology that sorts items into
categories of grinding tools, polishing tools, paraphernalia, containers,
and so on. The typological scheme chosen here for describing ground
stone types is activity based. Within the broad activity sets are subsets of
items that are designed for similar functions. The type definitions essen-
tially describe the typical or most common representative of the set, and
when appropriate, those that are close to the fuzzy boundaries are described
as examples of the range of variation within each set. Sometimes those
items closest to the fuzzy boundaries are the most interesting and may
represent technological innovations, experimentations, or adaptations from
foreign traditions.
The primary assumption behind the research presented here is that
most ground stone tools in the U.S. Southwest were made by the individ-
ual users, and only occasionally was there production specialization. If
the person making the tool was not the user, then the user at least had di-
rect influence on tool design. This assumption has a foundation in ethno-
graphic descriptions of material procurement trips and manufacturing
techniques for some very specific tools, such as Hopi pikistones, and Gua-
temalan metates (Horsfall 1987; O’Kane 1950:40–46). Specialization in
ground stone manufacture may have existed at times, and at varying levels,
at least in some places in the U.S. Southwest (see, for example, Bostwick
and Burton 1993; Doyel 1985). Metates, ornaments, palettes, and other items
that may require certain skill to make are the most likely candidates for
specialized production. This is an area of research that must receive more
attention, comparable to that given to specialization in pottery manufacture.
This introductory chapter has touched on a series of diverse topics rang-

ing from defining ground stone as an analytical category, to explaining
why function is better defined by design than by form, to recognizing our
ability to make analogous and homologous comparisons, and beyond to
the theoretical bounds of classification techniques. All of these topics are
explored throughout this manual. What should now be clear is that this is
not a simple topic. The items grouped under the heading ground stone
were intimately interwoven into the daily activities of prehistoric people.
We thus have the opportunity to learn about more than just the things
people used. We can learn about how people planned to use their tools,
and then compare that to how the tools actually were used. We can learn
about how tools were discarded or destroyed. If the data are appropri-
ately organized, we can select specific attributes that will let us evaluate
the motions employed with specific tools. This may be one of the best
ways to learn about the intimate details of prehistoric life. The next chapter
defines in more detail the specifics of a technological approach to ground
stone analysis.
Chapter 2
Grinding Technology and

Technological Analysis
G rinding technology, as a subset of the larger realm of technology, is

subsumed under the broader definitions that relate to socio-material
issues. Building on the works of Lechtman, Lemonnier, and others, Dobres
and Hoffman (1999a:2) broadly define technology as “a pervasive and
powerful complex of mutually re-enforcing socio-material practices struc-
tured by self- and group-interests, expressions of agency, identity and af-
filiation, cultural ways of comprehending and acting on the world, practi-
cal and esoteric knowledge, symbolic representations, and skills.” Within
this framework, the study of grinding technology focuses on the combi-
nation of knowledge, ideas, behavior, and equipment that solves the prob-
lems of altering surfaces or reducing substances. By viewing ground stone
as part of a larger technological system, we can discuss how prehistoric
people solved specific problems: (1) how they selected material and man-
ufactured items for specific functions; (2) how they used them both pri-
marily and secondarily; (3) the motions or kinetics used with specific tools;
and (4) how they wore out, destroyed, or discarded their belongings. The
theoretical definition of grinding technology needs to be structured at a
methodological level for the purposes of analysis. How can we get from
measuring, describing, and classifying to finding patterns in the data that
can be used to discuss the behavioral constructs expressed in the socio-
material practices of groups and individuals?
If we can learn to recognize baseline technological traditions, then we
can make comparisons, thereby addressing questions central to archaeo-
logical research about temporal and cultural developments. The concept
of a baseline technological tradition builds on the contention that certain
cultural principles structure the manner in which people manufacture
and manipulate their possessions and tools (see, for example, Dobres and
Hoffman 1999a:5–10; Lechtman 1999:224). Thus, distinct cultural groups
17
express their technological traditions in recognizably distinct designs and

techniques. Technological developments may accrue to meet changing
needs or spike as a result of contact with other people who had different
technological solutions to the same problem.
Specific research questions can be tailored to fit particular projects.
For example, projects at sites with considerable time depth are well suited
to evaluating technological developments through time. Those with sites
scattered across a broad geographical area provide an opportunity to dis-
cuss technological differences in environmental adaptations. The key to a
successful analysis is to be aware of the possibilities and limitations of the
data for answering project research questions. This manual should be used
to inspire analyses rather than to restrict them. It is a work in progress
and a tool for acquiring knowledge. As more knowledge is acquired, our
understandings change, and we should be flexible enough in the way we
classify material culture to adapt to the changes. Thus, this is an open-ended
analysis method designed to accommodate growth and advance our un-
derstanding of the past. Model building through ethnographically and
experimentally derived analogies (as discussed further in Chapter 3) is con-
sidered a useful method for advancing knowledge and our understand-
ing of prehistoric grinding technology.
By defining jargon that categorizes life-history concepts such as tool
design, tool use, use-wear patterns, and kinetics, it becomes possible to
standardize observations made about artifacts. This allows us to commu-
nicate better about which attributes might be the most important for an-
swering specific research questions. Some of this jargon has existed in the
archaeological literature for as long as descriptions have been written. Un-
fortunately, artifact and attribute labels have not always been consistently
applied. Other terms are new to archaeology, having developed in the field
of tribology, which is the study of friction, lubrication, and wear (Blau
1989; Czichos 1978; Dowson 1979; Kragelsky et al. 1982; Quinn 1971; Szeri
1980; Teer and Arnell 1975). Certain terms borrowed from tribology are
particularly useful for understanding the use-wear process as described
below. The following sections define attributes specific to the different
life-history stages in technologically derived terms.
DESIGN AND MANUFACTURE
Understanding technological processes begins with considering how an

item was designed and manufactured, why the material was selected, and
what features were made to fit the chosen material with the planned func-
tion of the object. These are essentially the first, post-conceptual processes
2. G RINDING T ECHNOLOGY 19
in the life-cycle of an object. Decisions made at the design stage generally

begin with choosing lithic material for appropriate size and texture (Horsfall
1987:340). Choice is a behavioral construct that we can interpret on the basis
of quantifiable variables relating to raw material and manufactured fea-
tures. The variables of raw material include granularity and rock size. Man-
ufactured features include specific shapes, decorations, handles, grooves,
or other features that make the tool comfortable to hold (collectively referred
to as comfort features). These quantifiable variables are the analytical con-
structs that make it possible to compare assemblages at the design stage.
Rocks may be chosen from among those that are loose on the land-
scape, in streambeds, or quarried from bedrock formations. Rock size and
weight are more important selection criteria for some tools than for oth-
ers. A metate can be no larger than the largest rocks available, and a mano
no larger than the width of the metate. A hammerstone should be heavy
enough to have the required impact on the contact surface. Size choices
may be restricted by available sources, however (see, for example, Stone
1994).
Lithic material has an exploitable natural granularity (a measurable
analytical construct: fine-grain, coarse-grain, and so on) that is sensed as
texture (a relational construct: smooth, rough, and so on). Texture is an im-
portant attribute for manufacturing tools. For example, a course-grain
rock may be chosen for an abrader because it is rough enough to remove
material from the surface of another item. The abrader’s surface texture
changes and becomes smoother through use: this is wear. The desired rough
texture can be restored through wear management techniques (a behav-
ioral construct). When the surface becomes worn and is too smooth to
abrade, it must be re-roughened with a pecking stone if the tool user wants
to continue to use it as an abrader. The impact of a pecking stone fractures
and removes material from the abrader surface, thereby making it rougher.
Alternatively, the tool could continue in use as a smoother or polisher. If
fine-grain rock is the only material available for making an abrader, the
surface can be roughened during manufacture to make it function as if it
were made from coarser-grain material. Thus, natural granularity may
not tell us as much about the use of a ground stone tool as does its altered
and re-altered surface texture.
As a quantifiable analytical construct, material granularity can be stan-
dardized by recording grain size. A suggested standardization is presented
in Appendix A under the attribute coding for texture. There are also more
formal geologic or sediment tables and charts that classify individual grain
sizes. Some textures are a combination of grain sizes, and some are smooth
enough to categorize as having no texture. Vesicular material can be cate-
gorized by large, small, or a combination of vesicle sizes (vesicles are cavi-

ties in volcanic rock left by bubbles of air or gas that escaped as the molten
rock hardened).
Durability, which is defined as a material’s ability to withstand wear,
is an important attribute for processing tools that alter the texture of an
intermediate substance through grinding. Especially when grinding food,
it may be more important that the material is durable enough to not erode
into the food than it is for it to be coarse- or fine-grain (Adams 1999). For
example, well-cemented granular materials were interpreted as a better
choice than loosely cemented materials for food-grinding tools at Ho-
mol’ovi III in northeastern Arizona (Fratt and Biancaniello 1993). Durabil-
ity is also an important characteristic for certain manufacturing tools. The
raw material chosen for a polishing stone must be smooth enough to pol-
ish rather than abrade, and the rock grains need to be well cemented and
hard enough to create a surface that polishes without disintegrating into
an abrasive powder. Thus, both a suitable smooth texture and high dura-
bility are important selection criteria for polishing stones. Durability is
also discussed below in the context of use-wear analysis.
Ethnographic and archaeological references to material acquisition and
manufacturing activities permit us to create diverse models. For example,
there are at least three descriptions of the methods used for quarrying pik-
istone material (Cushing 1920:326–327; O’Kane 1950:40–46; Stevenson 1904:
361). Some of the descriptions include rituals that surround the removal
of material from the earth, or stress the care required in manufacturing.
Quarrying and manufacturing tasks vary as the work of men or women,
but the final finishing is the responsibility of the women who use the pik-
istone. Such descriptions provide a baseline for modeling prehistoric be-
havior and serve as our underlying assumptions.
Quarry sites contain a variety of technological evidence concerning
material acquisition. For example, at quarries in southwestern Arizona
and southeastern California large and small hammerstones have been
found along with unfinished tools covered with flake scars and impact
fractures. These hammerstones and blanks provide evidence about the
mechanics involved in different stages of manufacture (Schneider 1996:
306). Together, the ethnographic and archaeological records allow us to
better understand the complex interactions among people and their things
as they go about acquiring the materials they need to manufacture tools
for everyday living.
Archaeologically, there are documented examples of artifact distribu-
tions that are proposed to have been the result of an organized social
infrastructure. For example, metates recovered from some Phoenix Basin
settlements came from quarries considerable distances away (Bostwick

and Burton 1993; Doyel 1985). If finished metates were circulating to the
settlements from the quarries, then an individual grinder might not have
had direct input into specific design attributes, but this may have been an
acceptable compromise for securing a better-quality material than other-
wise available. If only the raw material was circulating, the user could
have had direct input into tool design.
The behavioral constructs of design can be assessed in terms of com-
plexity (Adams 1995:45). If the natural shape of the rock was altered only
through use, the item is considered to have an expedient design. Modifica-
tions that make the item easier to hold or to achieve a specific shape indi-
cate a strategic design. Analyzing design allows us to determine whether
strategically designed items were used or treated differently from those of
expedient design. Were they stored in more protected locations, and were
they subjected to more or less primary or secondary use than tools of ex-
pedient design? Some tools went through the design stage but were never
used. Those that never made it past the initial stages of manufacture are
blanks that could have been turned into anything. Unused items are those
that were manufactured with all the necessary attributes to be specific
tools but were never used.
USE
The way an item was used can be evaluated in terms of primary and sec-
ondary uses (Adams 1995, 1994a; see Schiffer 1987:27–46 for a slightly dif-
ferent slant to use categories). Primary use is that for which the item was
originally designed. Ground stone tool designs most commonly accom-
modate a single function. Any secondary use is usually a later addition to
that for which it was originally designed. There are two types of second-
ary use: concomitant and sequential. Items that can function in two or
more activities are of concomitant secondary use, whereas those whose sec-
ondary use precludes their ability to function in their primary use are of
sequential secondary use. Concomitant secondary use broadens the range of
accomplishable activities without increasing the number of tools, and also
conserves raw material. This is a behavioral construct that has relevance
in contrast to sequential secondary. Because sequential secondary use re-
moves items from their primary use, it is a more restrictive and probably
more serial behavior than that of concomitant secondary use. These con-
cepts have potential utility for evaluating the nature of activities that oc-
curred at particular settlement types (discussed further below). We can
evaluate questions about why one particular site has a high percentage of
concomitant secondary-use tools compared to another that has a high per-

centage of sequential secondary-use tools. Is concomitant secondary use
more common at long-term habitation sites, and sequential secondary use
more common at seasonally reoccupied sites? The nature of secondary
use may provide evidence for formulating answers.
Use categories are labeled and defined as follows: single use, reused,
redesigned, multiple use, and recycled (Adams 1995, 1994a). A single-use
item is employed only in the activity for which it is designed and seem-
ingly has only one function. Examples include manos and metates that
function solely as food-processing tools, and polishing stones such as those
used to polish pottery. There is neither use wear nor manufactured evi-
dence of secondary use, and thus nothing to further evaluate unless the
recovery context is more consistent with an interpretation of secondary
use, as in the case of manos, polishing stones or other ground stone tools
left with burials, and metates that became building stones.
A reused item is designed for a specific primary use, but becomes em-
ployed in a second activity without altering the design. Sometimes dis-
tinctive attributes are discernible only through a use-wear analysis or an
analysis of residues on the tool surface. Examples include food-grinding
manos or metates reused to grind pigment. The same surface is used in
both the primary food-processing activity and in the secondary pigment
processing. Reuse may partially or completely destroy evidence of pri-
mary use. For example, a pottery polishing stone may have been reused
to polish an axe. The use wear of stone polishing may completely over-
write that of pottery polishing except on the lowest portions of the stone
not in contact with the axe surface. (This relationship is discussed more
fully below in the section on use-wear analysis.) In some cases we are ca-
pable only of evaluating the last use of the tool; if the last use destroys all
evidence of earlier uses, then realistically there is no way of recognizing
the reuse. This is a conundrum that eventually may be resolvable through
residue analyses.
A multiple-use tool is designed for a specific primary activity, but an-
other area or surface is also used in a second activity. Use in one activity
does not inhibit use in the other, even if the tool’s configuration is slightly
altered. For example, the non-working surface of a pottery polishing stone
might be modified as a shaft straightener (Figure 2.1). The tool can still
function in the activity for which it was originally designed, and some-
times it may be difficult to assess which was the primary design. In such
cases, the design most heavily used is considered primary. These examples
of reused and multiple-use items are considered to have had concomitant
secondary use.
2.1. Multiple-use tool. Polishing stone with grooved abrader on opposite side,
an example of concomitant secondary use. (Adams 1994: Figure 5.7, 1996:
Figure 1.)
Redesigned tools are designed for a primary activity and either are re-
manufactured for, or altered through use in, a second activity to the extent
that the item no longer functions in the first. Such redesign might involve
placing a groove across the working surface of a polishing stone (Figure
2.2) or using a mano as a pestle to the point that it is no longer usable in a
metate. Attributes that define redesign include use-wear patterns inter-
rupted by features that were not part of the original design, such as a
groove across the surface of a polishing stone. If the polishing stone had
continued in service, the use-wear patterns would extend across the mar-
gins of the groove. If the end of a trough mano has been so badly dam-
aged through use as a pestle that it no longer fits in the metate trough, it is
considered redesigned. If use-wear analysis determines that abrasive dam-
age from continued mano use covers the impact fractures from pestle use,
it has concomitant multiple-use. This use-wear analysis technique helps
distinguish the actions of a two-step pounding and grinding process from
those that are truly related to different activities, one pounding and the
other grinding.
Recycled items are designed and used in one activity, but ultimately
employed in a completely different context that may or may not have phys-
ically altered the item. This is somewhat different than Schiffer’s (1987:29–
30) definition of recycling, which requires physical alteration of the items.
Manos and metates used as building stones or as roasting rocks are exam-
ples of recycled tools. The roasting activities physically alter the items,
whereas their recycling as building stones does not require alteration.
2.2. Multiple-use tool. Polishing stone with grooved abrader across polishing
surface, an example of sequential secondary use. (Adams 1994: Figure 5.8,
1996: Figure 2.)
Whole manos and metates are sometimes found within prehistoric walls.
Another example of a recycled item that may or may not be physically al-
tered is a mundane object that took on ritual significance as a mortuary
item. These items have been recycled out of one context into another. Some-
times the only way of identifying a recycled yet unaltered item is through
an assessment of its context. Metates found among wall-fall rocks, and
manos found in graves were recycled out of their original food-processing
activities. Redesigned and recycled tools are considered to have had se-
quential secondary uses.
The discussion of use as defined in this section tacks back and forth
between designed or intended use and actual use. A mano that has finger
grips for easy handling and a surface the size and configuration of a trough
metate was designed for a specific function in food-processing activities.
If there are remnants of pigment on the mano surface, it was also reused
in processing nonfood substances. The ability to code and statistically an-
alyze attributes related to both uses broadens the range of data available
for evaluating prehistoric behavior. These are mostly qualitative assess-
ments that can be analytically evaluated along with other data about de-
sign and use wear.
WEAR
Wear is the progressive loss of substance from the surface of a stone item
as a result of the relative motion between it and another contact surface
(Adams 1993b:63, 1988:310; Czichos 1978:98; Szeri 1980:35; Teer and Arnell
1975:94). As an analytic construct, the amount of damage created through
wear is classifiable using qualitative, noncontinuous variables. Light wear
leaves so little evidence that it can barely be seen with the unaided eye.
Moderate wear is enough to leave obvious damage but not alter the basic
shape of the tool. Heavy wear changes the natural or manufactured shape
of the tool. Some tools have been used so much that they are difficult to
hold for continued use, or the usable surface or edge is almost gone; these
are nearly worn out. Worn out items are no longer usable in the activity for
which they were designed. Unused items may have damage to their sur-
faces if, as part of the manufacture process, pecking or grinding was em-
ployed to create the surface, but there is no damage from use.
Wear can be assessed on the item as a whole as well as on individual
surfaces. Each separate surface can be evaluated with the categories de-
fined above. If a tool is used in more than one activity, it might be moder-
ately worn on more than one surface and be considered a heavily worn
tool. Combined with this assessment of wear is an evaluation of whether
a surface has been resharpened or reroughened. By recording the presence
of resharpening separately from the amount of wear, it is possible to as-
sess wear management.
Wear management is a strategy of tool maintenance. For example, a
maintenance strategy for trough manos and metates is to replace worn
manos with larger ones, thereby increasing tool efficiency. However, the
trough width may also need enlarging to accommodate the larger mano.
This is a wear-management strategy that can be recognized on the trough
wall by a ridge. Various wear-management strategies also can be recog-
nized for manos. One includes rotation so that the proximal edge becomes
the distal edge. Such rotation distributes the wear more evenly front to
back, thereby keeping the profile width uniform and the surface flat with
a maximum area of contact between the mano and metate (Adams 1993a:
334–336; Bartlett 1933:15). Manos not maintained with such a strategy de-
velop wedge-shaped profiles. Another wear-management strategy is to
create more than one usable surface on the mano. Two opposing surfaces
allow the grinder to keep grinding until both surfaces become inefficiently
smooth and require replacement or reroughening. Two adjacent surfaces

allow the grinder to hold a thin mano so that her fingers are above the
metate surface and out of harm’s way. How these wear-management strate-
gies relate to mano profiles is discussed further in Chapter 5.
Eventually, any tool that requires a particular configuration or surface
texture for optimal performance will need maintenance if it is to continue
in its designed use. For axes, tool maintenance might require a new han-
dle, or a wedge to tighten the bind on an existing handle, or movement of
the groove to adjust the balance of the head as it becomes worn. An axe
wear-management strategy incorporates chipping or grinding to resharpen
a blunted bit edge.
The ability to code and statistically analyze attributes related to wear
and wear management broadens the range of data available for evaluat-
ing behaviors. We can evaluate questions about why one particular site
has a high percentage of heavily worn tools compared to another that has
a high percentage of lightly worn tools. Is heavy wear more common at
long-term settlements, and light wear more common at seasonally reoc-
cupied settlements? Is there more evidence for wear management at long-
term settlements than at seasonally reoccupied settlements? If we also bring
in attributes from other technological aspects, such as designed comfort
features, we can begin to discuss the concepts of intensive and extensive
tool use, and tool efficiency.
The concept of use intensity is best illustrated by comparing two heav-
ily worn tools. One became that way through intensive use, and the other
through extensive use. The intensively used tool was worked during tasks
of long duration; for example, 5 hours a day for 50 days, resulting in 250
hours of wear. The extensively used tool was worked during tasks of short
duration; for example, one hour a day for 250 days, also resulting in 250
hours of wear. From an analytical perspective, the wear amount appears
the same: heavy. How can we distinguish intensive from extensive use?
One way may be through an assessment of design. If a tool was manufac-
tured with handles or other comfort features, it seems reasonable to infer
that the tool was designed to be comfortable to hold during tasks of long
duration. A heavily worn tool that does not have comfort features was
probably not designed for intensive use, and a case can be made for infer-
ring that the wear accumulated because of extensive use. Even though
these distinctions are not verifiable, they may add data to accumulated
evidence from other sources about the organization of a settlement’s work-
force or its occupation strategy. Settlements with large percentages of in-
tensively used tools may have had more specialists using the tools to work
at more enduring tasks than settlements with fewer intensively used tools.
It is, of course, possible that tools with comfort features were manufac-
tured by skilled technicians who took pride in making comfortable tools,
and that their form had nothing to do with anticipated use intensity. These
are assumptions that can be made for research questions addressing the
relationship between specialists and specialization in tool manufacture,
and settlement economy. This illustrates a perfect situation where the same
data set can be used to evaluate different research questions.
The concept of efficiency is best evaluated in quantifiable terms as the
amount of product that can be processed or the number of hours spent at
completing a particular task. Efficiency is a relative concept: as long as a
tool gets the job done, it is efficient. A different tool design, however, may
be more or less efficient. Relative efficiency can be measured and compared
by continuous variables of dimension. For grinding stones, relative effi-
ciencies might be measured by the size of the grinding surfaces and the
weight of the tools. Heavier tools with larger surfaces are more efficient
grinding tools than lighter ones with smaller surfaces (see Chapter 5,
“Manos and Metates,” for more examples of this concept). Efficiency is a
behavioral construct, and the full potential of comparing tool efficiency
has not been completely developed, making it an area ripe for experimen-
tation. What attributes might make one axe more efficient than another?
Are there recognizable attributes that might make one polishing stone
more efficient than another? Together, the concepts of and evidence for in-
tensity of use, efficiency of use, and wear-management strategies can be
used to view the prehistoric technological process of designing and main-
taining efficient tools.
USE-WEAR ANALYSIS
Use-wear analysis is the examination of an item for macroscopic and mi-

croscopic evidence that allows us to understand how it was altered through
use (Adams 1988, 1989a, 1989b, 1993b). For research on ground stone, four
mechanisms are helpful in describing and understanding the formation of
specific damage patterns: adhesive wear, abrasive wear, fatigue wear, and
tribochemical wear (a combination of mechanical and chemical interac-
tion). These mechanisms are not mutually exclusive in how they change
the surface, nor is each the result of a single, independent event. The four
mechanisms interact, and one becomes dominant over the others depend-
ing on the characteristics of the contacting surfaces and the nature of any
intermediate substances (Adams 1988, 1989a, 1989b, 1993b, 1994a). These
concepts have been developed for application to ground stone by build-
ing upon the research of tribologists who study friction, lubrication, and
wear (Blau 1989; Czichos 1978; Dowson 1979; Kragelsky et al. 1982; Quinn
1971; Szeri 1980; Teer and Arnell 1975). These are important concepts for
ground stone use-wear analyses because they provide a means for evalu-
ating wear patterns without having to create an experimental example of
every possible use situation. Use wear on specific items should always be
evaluated against an area on the stone that is unused or broken so that the
natural condition of the stone is known.
The surfaces of ground stone tools were described in terms of texture
in the Chapter 2 section titled “Design and Manufacture.” For analytical
purposes, ground stone surfaces can be described in terms of asperity and
topography. Asperity is a combination of material granularity and surface
texture, and is influenced by material durability. The surfaces of tools made
from coarse-grain material naturally have more asperity, or are more as-
perite, than the surfaces of tools made from fine-grain material. The sur-
face of a fine-grain tool can be made more asperite by pecking it to re-
sharpen the surface texture. Through use, the asperity of both fine-grain
and coarse-grain surfaces can be reduced to equally smooth textures. The
asperite surface of a tool made from durable material (some metamor-
phic and volcanic rocks) may not cause as much abrasive damage to a
contact surface as an equally asperite surface of a tool made from weakly
cemented material (some sedimentary rocks). Because the grains are eas-
ily dislodged from the weak material, the surface maintains its asperity
longer, but it also wears out faster than a tool of more durable material.
The surface of the durable material wears smooth because the grains are
not dislodged. If abrasion is an important function of the tool, the surface
texture must be reroughened.
An asperity is a single grain or a single projection from a surface. The
spaces between grains are interstices. Vesicular material varies in asperity
depending on the closeness of the vesicles and the roughness of the margins
(the edges between vesicles). Asperity is an important concept for under-
standing how use-wear patterns are created on ground stone surfaces
(Adams 1993b). Interstices and vesicles are important to evaluate for evi-
dence about the nature of the contacting surfaces. How far the wear ex-
tends into them is evidence of the relative rigidity of the opposite surface.
Another important concept for discussing use-wear patterns is that of
topography. A sandstone netherstone is a good example of a surface with
topographic relief if differential weathering causes some stone layers to
remain higher than others. A surface with no topographic relief has no el-
evational difference: it is flat. This is not meant to imply that it is smooth,
for the surface of a tool made from coarse-grain material might have no
relief to its topography and still not be smooth because of its high asperity.
2.3. Schematic of unaltered stone surface. Grains are the minute minerals and
rocks that make up the stone. These are sometimes referred to as asperities.
Interstices are the spaces or depressions between grains. Matrix is the material
that holds the grains in place. (Drafted by Ron Beckwith.)
In this sense the stone surface would have topography on a microscopic

level, and the term microtopography distinguishes the topographic varia-
tion visible under magnification from the topography visible macroscop-
ically. Surface topography is important when two opposing surfaces come
into contact. Between hard or rigid surfaces, only the higher elevations
make initial contact, and it is here that use-wear patterns are first created.
It is important to realize that no surface is perfectly flat. There is always
some topographic variation, no matter how minute (Figure 2.3; see also
Figure 2.13).
When two surfaces come into contact, even if there is no movement,
there are molecular interactions. These interactions create bonds that are
broken when there is movement of one surface across or away from the
other surface (Kragelsky et al. 1982:6; Czichos 1978:119–123). Movement
and the subsequent breaking of bonds release energy in the form of fric-
tional heat and loosen rock grains from one or both surfaces: this is adhe-
sive wear. The loosened rock grains either remain between the surfaces
or become attached to the opposite surface or another location on the
original surface. In the early stages of wear, damage may not be visible ex-
cept at very high power magnification. However, as wear progresses, the
damage builds up and interacts with the other mechanisms. Adhesive wear
on ground stone surfaces is probably best seen where they are handled,
2.4. Schematic of a surface damaged through fatigue wear. The grains damaged
by crushing and the step fractures have a frosted appearance. Note the loose par-
ticles that become active in the abrasive mechanism. (Drafted by Ron Beckwith.)
because skin oils adhere to the stone surfaces even if there is no active
rubbing.
As pressure, or the alternating stress of movement, is applied to con-
tacting surfaces, the highest elevations bear the weight and mass of the
load. If the load is more than is bearable, there is collapse and crushing of
the elevations (Czichos 1978:105; Teer and Arnell 1975:95). This crushing
is the result of fatigue wear (Figure 2.4). Damage is visible, both macro-
scopically and at low-power magnification, as cracks, step fractures, and
pits. The effect is similar to that seen on frosted glass. Fatigue wear might
destroy damage patterns created by adhesive wear, but it also opens up
fresh surface area upon which new adhesive bonds can be created. These
areas of fatigue are called impact fractures in this manual, but have been
called pecking in other contexts. They are easily seen on tools that have
been battered with pecking stones.
Particles that are loosened through adhesive and fatigue wear remain
between surfaces, becoming abrasive agents in the wear process. These
abrasive agents create scratches and gouges across the stone’s surface.
Material gouged out by the agents also becomes involved in the abrasive
wear process. Abrasive wear is also caused by the movement of a more
durable asperite surface across a less asperite surface (Figure 2.5). The
harder, rougher grains of the durable surface dig into the smoother mate-
rial of the other surface. Movement displaces the softer material, creating
scratches in the direction of the movement (Czichos 1978:126; Teer and
2.5. Schematic of a surface damaged through abrasive wear. Striations and

gouges loosen particles that become involved in the wear process. (Drafted by
Ron Beckwith.)
Arnell 1975:106). These scratches are descriptively referred to as striations,

which also include deeper scratches sometimes called gouges.
As surfaces move against each other, the alternating stresses of move-
ment and pressure instigate the mechanisms of adhesive wear, abrasive
wear, and fatigue wear. These mechanisms create surficial cracks on both
contacting surfaces. Once a crack has formed, crack propagation results in
the release of energy in the form of frictional heat (Czichos 1978:105–112).
The release of heat is only one of the factors important in the microenvi-
ronment surrounding the contacting surfaces. Other factors include any
intermediate substances between the contacting surfaces. In the industrial
world, intermediate substances might be lubricants. In terms of traditional
societies using ground stone tools, intermediate substances might be grain,
meat, clay, or anything processed between two stones; they might also be
properties of one of the surfaces, such as the oils in hides or bone, or the
silicates in vegetal remains. Other truly environmental factors are also im-
portant, including whether the surfaces are contacting in a wet or dry at-
mosphere, or in a clean or dirty context.
In essence what happens is an interactive process. Adhesive wear,
abrasive wear, and fatigue wear create an environment for the chemical
interactions of the tribochemical wear mechanism (Figure 2.6). These chem-
ical interactions produce reaction products, which are the films and oxides
that build up on surfaces (Czichos 1978:123). These reaction products are
visible on stone surfaces as sheen, sometimes referred to as polish by tech-
nologists studying flaked stone tools. Tribochemical interactions are con-
2.6. Schematic of a surface damaged through tribochemical wear. The sheen on

worn surfaces is the buildup of residues and reaction products created by chemi-
cal interactions. (Drafted by Ron Beckwith.)
stantly occurring and are enhanced by frictional energy and mechanical

activation. However, unless the reaction products are allowed to accumu-
late, they cannot be seen. While the other three mechanisms constantly
expose fresh surfaces upon which interactions can occur, they are con-
comitantly removing any buildup of reaction products. Reaction products
continue to be removed until the higher elevations of the contacting sur-
faces are crushed to the point that fatigue wear is no longer a factor, and
the asperities of the two surfaces are no longer gouging each other. Thus,
reductions in surface topography and surface asperity allow reaction prod-
ucts to build up enough to be macroscopically visible.
It is easy to see that the mechanisms of adhesive wear, abrasive wear,
and fatigue wear are reductive processes, each with distinctive damage
patterns. Tribochemical wear, however, is additive. The two most impor-
tant facts to remember are that visible use wear is from the mechanism
most recently in operation on the surface (Adams 1988, 1989a, 1989b, 1993b),
and that the best way to evaluate use wear is to compare it either to an
unused area on the tool or to a piece of raw material of the same type. It is
important to know what the natural material looks like to distinguish nat-
ural process from human processes (see Figures 2.3 and 2.13).
Ground stone tools are used against a variety of surfaces. Manos and
metates, mortars and pestles, grinding slabs and handstones have stone-
against-stone contact with an intermediate substance adding to the con-
tact environment. Abraders most often have contact of stone-against-wood
or stone-against-bone. These contacts are more pliable than stone and cre-
ate damage patterns very different than those of stone-against-stone con-

tact. Stone surfaces used against soft surfaces, such as handstones used to
process hides, have yet another very distinctive damage pattern.
Contact surfaces can be described in terms that refer to their relative
resiliency. Soft contact surfaces can engage more of the stone’s working
surface by pushing into the interstices and topographic lows that are un-
reachable by harder contact surfaces. Fresh hides and human hands are
examples of soft contact surfaces. Pliable surfaces are more rigid than soft
ones and do not extend as deeply into the lows and interstices; examples
include green wood and bone, and dried hides. Resilient surfaces have
some give and may extend into the topographic lows, but they are not
soft enough to extend deep into the interstices; examples include so-called
green plaster and clay surfaces, and fibrous leaves. Hard surfaces engage
only the highest grains and topographic elevations. Hardened plaster and
clay, and other stones are examples of hard surfaces.
In an attempt to understand how distinctive use-wear patterns are
formed and what role intermediate substances play, a series of experiments
is continuing using replicated tools of sandstone and basalt (Adams 1989b,
1993b). Replicated grooved (Figure 2.7) and flat abraders (Figure 2.8) have
been used to work sheep medapodials into awls, smooth arrow shafts,
shape digging sticks, shape gaming pieces, and make shell beads (Adams
1989b, 1993b). Through these experiments we have learned that a func-
tional awl can be made in less than half an hour, a digging stick can be
made in less than one hour, and one olivella shell bead can be readied for
stringing in 10 minutes (Adams 1989a, 1993b).
Experimental tools have also been used to grind dried feed corn (Fig-
ure 2.9), sunflower seeds, amaranth, clay, and sherds. These results are re-
ferred to in Chapter 3 and published in full elsewhere (Adams 1999). All of
these experiments have been in progress intermittently since 1987 (Adams
1988, 1989a, 1989b, 1993b, 1999). During this time, however, only a small
baseline of data has accumulated. With this caveat in mind, it is possible
to discuss damage patterns in general, yet be aware that additional re-
search will be required to replicate and refine these preliminary results.
Stone-against-Stone Contact
Using experimental manos and metates as a baseline for description, it is

possible to identify the damage caused when two stone surfaces are worked
against each other. The first points of contact are on the highest elevations
of the stone. At the macroscopic level these are uneven bumps or ridges
on the rock. At the microscopic level these are the tops of grains or high
2.7. Experimental abrader with V-shaped grooves used to sharpen sheep bones.
(Adams 1989b: Figure 6.)
2.8. Experimental flat abrader used to smooth a digging stick. (Adams 1989b:
Figure 5.)
2.9. Experimental mano and metate used to grind dried feed corn. (Adams
1989b: Figure 1.)
points in the stone’s crystalline structure. The rigid stone does not push
into the topographic lows, the interstices between the grains, or inside
the vesicles. The moment the two stones contact, adhesive and fatigue
wear mechanisms are in operation. If the load is unbearable, the higher el-
evations and weaker grains are crushed or fractured (Figure 2.10). Move-
ment of the surfaces breaks the adhesive bonds and starts abrasion. Fric-
tional heat released by breaking bonds and crack propagation creates an
environmental change in which tribochemical reactions begin. The reac-
tion products build up enough to be macroscopically visible only if they are
not worn away by continued adhesive, fatigue, and abrasive mechanisms.
The experiments, using new manos and metates, were specifically de-
signed to determine whether or not it is possible to differentiate damage
caused by different intermediate substances between stones. The main
question is, do maize, sunflower seeds, amaranth seeds, clay, and sherds
create different use-wear patterns when worked between two grinding
stone surfaces? After 6 to 15 hours of grinding each substance, it is possi-
ble to make some very preliminary observations about the different wear
patterns.
The effects of abrasive and fatigue wear mechanisms are the easiest
and quickest to observe at a macroscopic or low-power microscopic level.
Crushed grains and striations happen immediately. As grinding proceeds,
matrix and grains worn level frosted appearance
2.10. Photomicrograph of use wear caused by grinding dried feed corn

(magnified 40X). Most of the grains have been worn flat, some level with the
matrix. Surface fatigue has caused some fracturing that gives the frosted appear-
ance. Compare to Figure 2.11. (Adams 1988: Figure 8.)
however, fine powder or meal fills up the interstices or vesicles of the

lithic material, and possibly slows the crushing effects of fatigue wear.
Essentially this makes the surface less asperite and reduces the stone’s
grinding efficiency. Once the surface is cleaned, roughness is restored,
and efficiency improves.
A comparison of the experimentally used surfaces suggests that the
most distinctive wear pattern is a result of tribochemical mechanisms. No
matter what material was ground, the stones’ surfaces all had at least a
few striations and impact fractures from abrasive and fatigue wear mech-
anisms. The difference lies mainly in a sheen created when oily substances
are ground (Figure 2.11). The sheen is most likely the result of tribochem-
ical interactions between the oily substances and the mineral properties of
the stone, enhanced through frictional heat. Whether or not the sheen can
withstand post-depositional conditions is a question yet to be analyzed.
The tribochemical sheen is most obviously visible with the grinding
of sunflower seeds. The sheen on surfaces used to grind sunflower seeds
accumulates on the highest and lowest topographic elevations on the
flattened grains
sheen flattened grains
2.11. Photomicrograph of use-wear caused by grinding sunflower seeds

(magnified 40X). Note the sheen and the flattened grains. Compare to Figure
2.10. (Adams 1989b: Figure 3.)
stone’s surfaces. It does not, however, extend into the microtopographic

interstices or vesicles. The sheen on surfaces used to grind amaranth or
corn becomes visible only when the reaction products are no longer worn
away by abrasive or fatigue wear. In fact, on lightly used surfaces it is not
possible to distinguish those used to grind corn or amaranth from those
used to grind clay or sherds. The most appropriate conclusion is that the
concomitant operation of wear mechanisms makes it difficult to see dis-
tinctive sheen until the tribochemical wear mechanism is more dominant
in the wear process than either the abrasive or fatigue wear mechanisms
(Adams 1993b).
Stone-against-Wood or Bone Contact
The experimental tools used to make arrow shafts, digging sticks, and
bone awls provide a baseline for describing differences in use-wear pat-
terns created between stone and a resilient or a pliable surface (Adams
rounded grains rounded grains
angularities in interstices
2.12. Photomicrograph of a stone surface used against a pliable surface

(magnified 40X). The tops of the angular grains became rounded, but wood,
bone, and other pliable surfaces are not soft enough to reach very deeply into
the interstices.
1989a, 1989b, 1993b). Because they are more pliable, wood and bone push
deeper into the topographic lows, vesicles, and interstices than would an-
other stone surface (Figure 2.12). Furthermore, wood and bone are not as
asperite as stone, so the only opportunity for gouging and scratching comes
from grains loosened from the stone tool. The movement of softer mate-
rial against the grains rounds off their sharp edges, which is an abrasive
action. For this reason it is important to be familiar with the appearance
of unused material; naturally round grains should not be confused with
grains rounded through use wear (Figure 2.13; also see Figure 2.3). All ex-
posed edges of the grains are much more involved in the wear process
than in stone-against-stone wear.
As green wood is worked against an abrader, it leaves a sticky residue
over the stone’s grains, interstices, and vesicles. Further rubbing removes
the residue from the grains, but the interstices and vesicles are harder to
clean. Dry wood leaves drier dust on the stone surface, but this dust does
not interfere with the abrasion process as much as the residue from green
2.13. Photomicrograph of an unused surface (magnified 40X). Note the more-

angular grains compared to Figures 2.11 and 2.12.
wood. The resultant wear patterns from green and dry wood are at this
point indistinguishable. It must be pointed out, however, that there has
not yet been enough experimentation to determine whether or not it is
possible to differentiate these wear patterns.
Green and dry bone, on the other hand, do leave distinguishable wear
patterns. The difference is probably a result of the oils that remain in the
green bone. There is noticeably more sheen on the surface of the tool used
to abrade green bone. The sheen is similar to that created by grinding
sunflower seeds, although the green bone does not abrade or level the
grains as happens with sunflower seed grinding. What remains unknown
at this point is whether any sheen created through tribochemical processes
survives burial in archaeological contexts. It is somewhat discouraging to
find very little difference in wear patterns produced by abrading dry bone
or any kind of wood. This conclusion is based on fewer than 10 hours of
use, however, and it might be possible to recognize more-distinctive use-
wear patterns on a more extensively used tool.
Stone-against-Hide Contact
An experimental stone used to process a dried deer hide provides the

baseline for describing the damage created on a stone used against a soft
rounded grains in interstices rounded grains
2.14. Photomicrograph of use wear caused by working a hide (magnified 40X).

The tops and sides of the grains become rounded as the soft hide pushes into the
deepest interstices. (Adams 1988: Figure 9.)
surface (Adams 1988). When water is worked into the hide with a stone,
the hide pushes up against the stone surface, completely filling the topo-
graphic lows and interstices or vesicles. There is nothing obviously abra-
sive in the hide. Abrasive material is added if a stray grain dislodged from
the stone’s surface or a particle from the environment is trapped between
the hide and the stone.
Pressure, triggering fatigue wear, becomes a factor as the user pushes
down on the stone while moving it across a hide placed on a hard surface.
Adhesive wear and tribological wear are the most dominant mechanisms.
Frictional heat, although not as obvious as with stone-against-stone con-
tact, helps to change the contact environment. The resulting residues build
up and become visible as a sheen on the tops of grains, down the sides of
the grains, and deep into interstices and vesicles (Figure 2.14).
How easily the handstone moves across the hide depends on the
amount of connective tissue and moisture on the hide. Unless the connec-
tive tissues have been removed and the hide worked until dry, the hand-
stone texture becomes clogged and not very abrasive. Washing removes
the clogging material and restores abrasiveness, but slows the overall
process. An abrasive stone can easily work a nap into a dry hide, creating
a soft, suede-like surface. A handstone seems much more useful for work-
ing a dry hide than a fresh one. The use wear created by rubbing a hide is
unlike anything on the surfaces of tools used to grind food. The wear that
results from handling is closest to that of hide working, but is generally
distinguishable by its location on the tool.
Each description of a particular contact situation makes it clear that
many things can be discerned about use-wear patterns by simply looking
at the surface of a ground stone tool without magnification. Striations,
crushed grains, impact fractures, leveled areas, and sheen are all macro-
scopically visible indications of surface wear. The locations of damage
patterns are important to note for assessing contact situations. For exam-
ple, if surface topography is diverse and use-wear patterns are visible
only on the highest elevations, the stone was probably in contact with an-
other stone or other very rigid surface. If the damage extends into the
lower elevations as well, the contact surface must have been pliable enough
to reach into these depths (Adams 1989a, 1989b, 1993b). Macroscopic ob-
servations also help recognize the kinetics of the tool manufacturer or ma-
nipulator. The direction of abrasive striations indicates the direction of
use. The location of damage patterns also indicates if the tool was rocked
or rotated during manufacture or use, as described further below.
KINETICS
Kinetics are the motions and forces related to tool operation. The configu-
ration and location of striations, impact fractures, chips, and facets result
from specific strokes used with each tool and are evidence of basic motions.
Facet location indicates where pressure was applied during the stroke.
Abrasive scratch direction indicates whether the stroke was a reciprocal,
back-and-forth movement or a circular stroke. A circular stoke causes stria-
tions to form in multiple directions across the stone’s surface. A reciprocal
stroke moves the tool back and forth across the surface, resulting in unidi-
rectional striations. If the edges are not lifted, and the surfaces remain in
contact at all times, then wear is confined to the surface: this is a flat stroke.
If the edges are lifted it is a rocking stroke. A circular, rocking stroke exerts
pressure on the stone’s surface at different spots as it moves across the op-
posing contact surface. The same basic strokes are also recognizable for
manos, polishing stones, abraders, and any other handstone. These strokes,
used with different types of manos and metates, are discussed in Chapter
5, “Manos and Metates.” Refer to Morris 1990 and Haury 1950:313–315
for their impressions of the kinetics used with small manos on flat metates.
Impact fractures and chips are caused when one surface is brought
into forceful contact with another, as when a pestle is used with a mortar,
or a pecking stone is used to resharpen a metate. The more forceful the

stroke, the larger the impact fractures and chips. When the weight of the
stone alone is used to crush an intermediate substance, the impact frac-
tures are uniformly shallow and uniformly distributed across the stone’s
surface. Rotating or rocking of one surface against the other further crushes
the intermediate substance and adds striations to the impact damage. When
more force is brought to the contact, either from the stone being lifted
higher or brought down with more muscle power, the impact fractures
are deep and concentrated along the area of contact between the surfaces.
Repeated forceful strokes alter the surface or edge shape through the re-
moval of chips and loose grains. For descriptive purposes, crushing de-
scribes a stroke using the stone’s weight to reduce the intermediate sub-
stance, and pounding describes a more forceful stroke.
Other stroke categories, such as pecking and chopping, describe force-
ful downward strokes used with a tool that does not have a flat working
surface. For example, pecking describes a stroke used with a tool that has a
rounded or convex surface, such as a hammerstone. It is a short-distance
stroke used with less force than a pounding stroke and is more like crush-
ing. Pecking alters the contact surface’s configuration by creating small
impact fractures that may be difficult to discern from impact fractures
made by a crushing stroke. Such a stoke would be useful for resharpening
the surfaces of manos and metates. Chopping describes the stroke used
with tools that have sharp edges designed to break away pieces of the
contacting surface. Such a stroke would be used with an axe for chopping
trees, or with a tabular tool for chopping agave leaves.
Kinetic assessment provides an opportunity to understand individual
variation. Kinetics reflect the actions of individuals in the sense that a
right-handed person moves tools differently than a left-handed person. At-
tention to the details of kinetics is important so that individual variations
are not confused with those of design and technological development.
DISUSE OR THE AFTERLIFE
What brings a ground stone item to the end of its use life? How does it
enter the archaeological record? Discard, loss, caching, and abandonment
processes have all been identified as mechanisms by which items enter
the archaeological record (Schiffer 1987:47–98). Item depositions sometimes
have meaningful cultural constructs, such as those associated with mor-
tuary features, foundation deposits, closing deposits, or other ritual offer-
ings (see, for example, Hoffman 1999:119; Walker 1995). An analysis of ar-
chaeological context helps us understand whether items were found where
they were used or stored by the prehistoric occupants, whether they were
discarded as trash, or whether they were left in a ritualistic fashion. Ex-

amples of important contexts are architecturally related structure fill (such
as roof fall, wall fall, or floor fill), floor contact, interior pit, exterior pit,
and trash mound. Information provided by excavators is important for
determining archaeological context and whether features contained trash
or possible de facto deposits. De facto deposits, as used here, were defined
by Schiffer (1987:89) as “tools . . . that although still usable (or reusable),
are left behind when an activity area is abandoned.” Scavenging and nat-
ural processes might move or remove items from where they were aban-
doned, discarded, stored, or cached. Site formation processes and the pro-
cesses of human agents must be considered when inferring how an item’s
archaeological context is relevant to its systemic context. See, for example,
Schiffer’s (1987) treatise on site formation processes and Schlanger’s (1991)
article on the archaeological context of manos and metates.
A ground stone tool may become worn out and reach the end of its
designed use life, but still remain useful through redesign or recycling. A
worn out mano might still be useful as an abrader, be redesigned into a
hoe, or become recycled in a roasting pit. Manos and metates that do not
appear worn out have been recycled as building stones in Puebloan walls.
We can recognize the intentional destruction of some items. Metates with
holes in their grinding surfaces have been described as “killed” by Haury
(1985b:231, 244). Metates at Higgins Flat Pueblo were intentionally broken
(Martin et al. 1956:72). The holes knocked through the metate bottoms are
surrounded by impact fractures and chips from where they were struck
with hammerstones. Impact fractures at the edges of the breaks were also
caused by hammerstones. Palettes found with cremated remains at Julian
Wash in the Tucson Basin are broken into three pieces, with hinge fractures
indicating the direction from which they were snapped. The tines from cru-
ciforms recovered from Las Capas, also in the Tucson Basin, have impact
fractures from where they were intentionally broken. These individual
destructive acts all had meaning within their respective cultural contexts.
The afterlife of an item continues with recovery and analysis. It is no
longer acceptable to leave ground stone unanalyzed—excavated but not
collected—because questions answerable with ground stone data are be-
ing expanded. The key is to properly frame the questions and properly se-
lect the most appropriate data to answer them. Remaining aware of ar-
chaeological context is paramount for selecting appropriate data.
DATA COLLECTION
Ground stone can be analyzed at two levels of effort. The first is an inven-
tory in which every item is counted, classified according to a specific
typology, assessed for completeness, and perhaps identified by material

type. Such an accounting is useful for preliminary reports that describe
what was found. The second, more intensive level of analysis incorporates
more detail, including measurements and assessments of attributes that
identify design and manufacture, primary and secondary uses, use wear,
the activity for which the item was designed, the activity in which it was
actually used, the archaeological context from which it was recovered, and
other variables deemed important for answering project-specific questions.
Time constraints usually dictate that choices be made among variables to
be recorded. Generally, a project research design narrows the choices.
Ground stone can be cumbersome to analyze. The artifacts vary greatly
in size, shape, and weight, and some are not easy to move around. Large
open tables are best for arranging the items for analysis where they can be
sorted and resorted by function, design, use wear, provenience, or any
other pertinent factor. This allows the researcher to visualize gross simi-
larities and differences in configurations. Some ground stone items are
too large to fit under a microscope unless the scope is mounted on an ad-
justable boom stand. A stereo microscope that has scrolling capabilities of
18 to 40 power is ideal for most use-wear analysis, and one that scrolls up
to 100 power is a bonus. An attached photograph tower and camera mount
are important if use-wear patterns are to be documented. A light source
independent of the microscope is most useful for adjusting light intensity
and direction. Sometimes adjusting the light position makes it easier to
see certain wear patterns. A heavy-duty, 150-watt, fiber light system with
dual optic focusing lenses has proven to be the most useful and versatile.
Scales and tools should be flexible enough to weigh and measure items
ranging in size from less than 1 cm to as much as 1 m, and less than 1 g to
as much as 50 kg. A good analysis laboratory should also have space for
experimentation and storage of comparative collections, which are useful
for both use-wear analysis and material identification.
Not all ground stone needs to be analyzed under a microscope, but it
is amazing what is missed by not doing so. For example, a mano second-
arily used to process pigments may go unrecognized if the only remnants
of the pigment are deep in the stone’s interstices. Some pieces, particu-
larly those that have been fire-cracked, are probably not good candidates
for microscopic analysis. Our current understanding of thermal alterations
cannot yet recognize the impacts of heat on use-wear patterns—another
research topic that needs experimentation and identification of distinguish-
ing attributes. Time constraints, more than anything else, probably will
dictate whether a microscopic analysis of each item in an assemblage is
warranted.
Use-wear analysis can be conducted at both a macroscopic and a mi-

croscopic level. Macroscopically it is possible to distinguish the larger
damage patterns created by wear from those created by manufacturing
techniques. Direction of use and type of stroke are almost always deter-
minable without magnification. As discussed in the section on use-wear
analysis, the nature of the contact surface is best recognized with magni-
fication. In an ideal research world an analyst would be able to experi-
ment with replicated tools made from the same materials as the prehis-
toric tools. Such experiments can distinguish important attributes related
to kinetics and use wear. Given that opportunities for experimentation
are rare, an understanding of wear mechanisms as defined in this chapter
should help the analyst build bridging arguments for tool use.
Another useful strategy is to identify compatible tools. There may be
manos and metates that were used together, compatible handstones, lap-
stones, or netherstones, compatible mortars and pestles, or as illustrated
in Figure 2.15, netherstones compatible with axe bits that were shaped on
them. Compatibility can be assessed in terms of how well the tools fit to-
gether and how well their use-wear patterns match. It is not possible to
prove that two tools were companion tools, but it is possible to eliminate
from consideration those that could not have functioned together. The
probability of tools with compatible configurations and matching use-wear
patterns being companion tools can be assessed through a consideration
of their archaeological context. Those recovered from the same archaeolog-
ical context, such as on house floors or within pits have a higher probabil-
ity of having been used together than those from spatially distant con-
texts. Nevertheless, should compatible tools be recognized from differing
locations this evidence might be used in conjunction with other clues to
assess contemporaneity or succession of features.
As mentioned in Chapter 1, one goal of this manual is to standardize
terms used to describe and record items, thereby making it easier to com-
pare assemblages analyzed by different researchers. Toward this end,
measurements should be taken in the same manner on each item. When
possible, a tool should be oriented according to how it was held, used, or
presented for viewing. The edge closest to the user is the proximal edge,
and the one most distant is the distal edge. For tools that are worked against
another surface, the side held in the hand is the dorsal surface and the
working surface is the ventral surface. Tools with multiple surfaces are
harder to label, and surfaces should be numbered from the largest or most
used to the smallest or least used. The illustrations accompanying the de-
scriptions in Part 2 show standardized locations for taking measurements,
and the terms used to describe specific attributes. Examples of coding
2.15. Axes and associated netherstones that were probably used in their manu-
facture. These netherstones would have been used either to hone the bit or put
the finishing polish on the bit, depending on netherstone texture. The axes and
one netherstone came from the same feature at Los Hermanos in the Tonto Basin.
The other netherstone came from a different feature. (Adams 2000c: Figure 10.3.)
sheets, attribute explanations, and recording forms are in the appendices.

The numbers assigned to the attributes on the coding sheets are arbitrary
and can be changed; however, using the same codes allows comparisons
to be made across assemblages from different projects. This recording sys-
tem was devised so that new attributes can be added and assigned unique
numbers. Once the attributes for each item have been recorded, it is pos-
sible to create data sets that will help address the selected research ques-
tions. Because it is impossible to anticipate every possible variation on an
attribute, analysts should expand on these examples as necessary.
ANSWERING RESEARCH QUESTIONS
Research questions can be chosen from a hierarchy of possibilities. At the

site level, all types of ground stone can be evaluated to answer questions
about occupation strategy. Were tools used or worn differently at settle-

ments intended for long-term occupation than they were at short-term
occupation settlements? What was the range of activities at the settlement?
Were different activities represented at year-round settlements than at sea-
sonal settlements? How were food grinding or other activities organized?
Were individual tasks repeated within each household, or was there a supra-
household level of organization?
Questions can also be asked to get at more individualized activities.
How were specific tools designed? Was effort expended to design tools that
are comfortable to hold? Was there management of wear? Were some tools
designed to be more efficient than others? Some of these questions cannot
be answered just by looking at the items. Knowing where they were found
is critical, especially for answering questions about organization. Further-
more, some answers might look very different when data from other ma-
terial categories are added. With that caveat in mind, the rest of this sec-
tion explores just a few of the possibilities. Note that the artifact types
referred to are all defined in Part 2.
Occupation Strategy
Occupation strategy is a topic of interest to many archaeologists. We want

to know how long settlements were occupied, and whether they were
continuously occupied or reoccupied. Ground stone tools are rarely used
to help answer such questions, but carefully designed hypotheses can be
used to add data from ground stone assemblages as another line of evi-
dence for assessing settlement continuity, duration, and intensity. Such
an approach was developed for a project along the Santa Cruz River in
the Tucson Basin (Adams 2001b; Gregory and Mabry 1998). For this proj-
ect, intensity was defined as “the range of activities carried out within a
settlement and the number of times per unit time those activities were re-
peated” (Gregory and Mabry 1998:15). Duration is the length of time a set-
tlement was occupied, and continuity is the degree to which the occupation
was continuous for that length of time (Gregory and Mabry 1998:15). The
ground stone variables selected as most useful for addressing occupation
strategy are design, use, wear, and activity.
The assumptions made for assessing the importance of design and
evaluating occupation strategy are: (1) if a tool manufacturer takes the
time and effort to make a tool comfortable to hold, the tool was more
likely intended for intensive than for extensive use; and (2) the tool users
were the manufacturers, or (3) if the manufacturer was not the user, the
user at least had direct input into tool design and manufacture. Different
assumptions, of course, would be required for research into complex soci-
eties with organizations of craft specialists above the household level. The
attributes recorded for all artifacts can be partitioned to evaluate the fol-
lowing hypotheses: (1) settlements where revisitation was expected have
higher percentages of strategically designed tools than settlements where
revisitation was not expected; (2) settlements where long-term occupa-
tion was anticipated have higher percentages of strategically designed
tools than those where long-term occupation was not anticipated. These
hypotheses have some precedent in research by Nelson and Lippmeier
(1993:302), who were evaluating reoccupied sites in southwest New Mexico.
Assumptions about tool use relate primarily to who used them in what
activities: (1) items were designed for use in specific activities, some of
which are more generic than others; (2) tools of concomitant secondary
use were most likely contemporaneously used in both activities; (3) The
purpose of concomitant secondary tool use was to broaden the range of
possible activities without increasing the number of tools, thereby con-
serving raw material or maintaining low numbers of stored objects; (4)
tools of sequential secondary use were most likely used at different times
such that the secondary user did not wish to maintain the tool’s primary
use. With these assumptions in mind the following hypotheses can be
tested and evaluated: (1) single-use items occur in higher percentages at
settlements with short-term or limited occupations than at sites with long-
term occupations or reoccupations; (2) conversely, secondary-use items
occur in higher percentages at settlements with long-term occupations or
at those repeatedly reoccupied; (3) long-term, continuously occupied set-
tlements have equally high percentages of concomitant- and sequential-
secondary use tools; (4) short-term continuously occupied settlements
have higher percentages of concomitant- than sequential-secondary use
tools; (5) reoccupied settlements have higher percentages of sequential-
than concomitant-secondary use tools.
The only assumption about wear put forth here is that the amount of
wear on a tool is the direct result of the amount of use. The most useful
hypotheses related to the question of occupation strategy are: (1) short-term
occupation settlements have higher percentages of lightly worn tools than
do long-term occupations; (2) long-term occupations have higher percent-
ages of unused tools than do short-term occupations; (3) long-term occu-
pations have higher percentages of moderate to heavily worn tools than
do short-term occupations.
An assemblage from any project can be partitioned into categories
that reflect activities such as the processing of food, pigment, hide, or fiber,
and the manufacture of pottery. Items that cannot be classified according
to specific activities can be grouped into more-generic functional activi-
ties such as grinding, polishing, abrading, scraping, or percussing. The
type descriptions presented in Part 2 are organized generally by activities,

a sorting technique that allows for behavioral- rather than thing-oriented
discussions.
The hypotheses that can be tested concerning activities and settlement
occupation strategy are: (1) settlements with long-term occupations have
ground stone associated with a greater variety of activities than do settle-
ments with short-term occupations; (2) continuously occupied settlements
have ground stone associated with a greater variety of activities than do
repeatedly occupied settlements (depending on duration and seasonality
of reoccupations); and (3) locations used by multiple groups have more
variation within each activity category than do locations used only by
one group.
Charts of relative percentages (Figure 2.16) are useful exploratory tools
for recognizing patterns in data sets. At a glance it is possible to see which
settlements had higher percentages of a particular attribute. Multiple charts
can be configured to highlight the co-occurrence of specific attributes that
can then be used to strengthen arguments for a particular type of settle-
ment strategy at a particular site. For example, a settlement with higher
percentages of strategically designed-, secondarily- and well-used tools,
and with a greater number of represented activities, was probably occupied
by a larger group, or by more groups, for a longer, more continuous dura-
tion than a settlement with a higher percentage of expediently designed,
single-use and lightly used tools, and with fewer represented activities.
Recognizing settlements at either end of this settlement type contin-
uum is, of course, easier than recognizing those in the middle. For exam-
ple, it might be difficult to distinguish settlements of long-term, repeated
reoccupation from those of short-term continuous occupation. If the ex-
ploratory methods do highlight patterns in the data set, statistical testing
or modeling tailored to evaluate archaeological diversity may be the next
step in the analysis. This is a previously unexplored area in ground stone
research, and the literature on the pleasures and pitfalls of diversity analy-
ses is large (see, for example, Cowgill 1990; Kintigh 1984; McCartney and
Glass 1990; Plog and Hegmon 1993; Rhode 1988). Nonetheless, testing and
modeling sites of all types may eventually help us recognize patterns in
relative frequencies of ground stone attributes that are most common to
each settlement type. The following discussion of food processing also has
suggestions for evaluating settlement strategies using data from manos.
Food-Processing Activities
To answer questions about how grinding technology was used to obtain

and process food, data on manos and metates, mortars and pestles, cook-
2.16. Bar chart illustrating the percentages of activities represented by ground

stone artifacts at three sites. This type of chart presents a good visual for compar-
ing assemblages.
ing slabs, pikistones, griddles, firedogs, fire-cracked rocks, hoes, and per-
haps axes and tchamahias (believed to have been used as hoes) can be eval-
uated in a number of ways. The presence of hoes, axes, and tchamahias
with use wear from working soil is evidence of farming activities. Partic-
ularly with axes and tchamahias, however, a use-wear analysis is required
to distinguish working in the dirt from other activities (Mills 1993). As-
semblages that have both manos and metates, and food-processing mor-
tars and pestles may represent the processing of a variety of food resources.
Some ethnographic accounts suggest that both tool sets were used to
process mesquite pods, with the mortar and pestle used in the first stages
of crushing the pods and separating the seeds, and the mano and metate
used in the final stages of grinding the pods and sometimes the roasted
seeds (Castetter and Bell 1951:179; Spier 1933:51). The caveat here is that
the types of tools involved in processing food items does not adequately
reflect the variety of food resources processed. More than one tool type
might have been involved in processing the same food, or more than one
food may have been processed with the same tool. Pollen samples taken
from prehistoric metates make this point quite clearly (Greenwald 1993:
348–349; Lancaster 1984:257). Tool morphology cannot be used to deter-
mine what was ground (Adams 1999; Horsfall 1987:369; Wright 1994).
An analysis of the types of manos and metates in an assemblage will

determine the most common tool design. Similarities and differences be-
tween the predominate mano and metate designs can help determine if
items are missing from an archaeological assemblage. For example, a re-
cent analysis of manos and metates from Santa Cruz Bend determined
that the most common metate type recovered was basin; however, most of
the manos had been used on flat/concave metates, and only one or two in
basin metates (Adams 1996). It was concluded that most of the flat/con-
cave metates either had been removed from the site at the time of aban-
donment, or perhaps later scavenged. The basin metates left behind may
represent obsolete tools or ritual caches, as evidenced by one metate found
with a severed human hand (Adams 1996:411). In this case, the metate as-
semblage alone provides little information about food-processing activi-
ties at Santa Cruz Bend. This is a caveat to those who consider metates
and other ground stone as permanent fixtures (see Schiffer 1987:93 for a
discussion of abandonment deposits).
An assessment of certain attributes may provide information on tool
efficiency and intensity of use. Data on the presence of finger grooves, or
shaping for comfortable holding, may indicate that a mano was intended
for intensive use, especially when compared to one lacking designed com-
fort features. Comparisons based on the presence or absence of mano com-
fort features may add support to conclusions made at the site level about
the duration or type of occupation. Combined with an analysis of the num-
ber of used surfaces on manos and the degree of wear, an argument can
be developed for how intensely particular mano assemblages were used.
At the site level, evidence for intensity of use might reflect occupation
strategy, recipes, and cooking techniques that rely on ground resources, or
even specialization in food production. For example, a settlement where
the manos are predominantly strategically designed with comfort features,
heavily worn, and have multiple surfaces was probably occupied for a
longer period than a settlement where the manos are predominantly ex-
pediently designed, lightly worn, and have only one surface. Alternatively,
at the first settlement, there may have been fewer grinders who used their
tools more intensely to feed the same number of people as at the second
settlement, where people were simply serviced by more grinders. Thus,
the grinders at the first settlement would have been more concerned with
comfort features and wear management than were the grinders at the sec-
ond settlement.
Measuring grinding surface area, counting the number of used sur-
faces, and assessing the amount of wear provide data useful for making
statements about food-grinding activities. It will not, however, directly
inform on what specifically was being ground. For that interpretation,
analyses of pollen and residue are needed. If such samples are not di-
rectly available from ground stone, macrobotanical studies can provide a
range of possible foodstuffs that might have been ground.
Cooking techniques are reconstructible through the presence of cook-
ing slabs, pikistones, griddles, firedogs, and to some extent, fire-cracked
rocks. The pots that rested on firedogs probably contained stews, mushes,
beverages, or other concoctions that required boiling or simmering. Cook-
ing slabs, griddles, and pikistones were used to bake dough or batter-
based recipes. Fire-cracked rocks are by-products of roasting or steaming
activities and may be easiest to recognize when associated with thermal
pits or when they occur in large quantities as refuse from cleaning out
thermal pits. Stone boiling also creates cracked rocks, and more research
is needed to recognize the attributes specific to their identification (see,
for example, Duncan and Doleman 1991). Clearly ground stone can be
used in a variety of ways to make inferences about behaviors relating to
food.
Manufacturing Activities
Grinding technology also had application in the industrious aspects of

prehistoric life, such as pottery production, tool manufacture, and the pro-
cessing of pigments. Certain polishing stones and pottery anvils can help
us analyze prehistoric pottery production. Pottery anvils, of course, indi-
cate that pots were shaped with the paddle-and-anvil technique. Use-wear
damage patterns on pottery polishing stones provide evidence for how
the pots were finished and for the nature of the finished pot itself. Stria-
tions were created by the rock temper used in some pots, and a lustrous
sheen forms through contact with hard clay surfaces. Wear facets are in-
dicative of extensive use. Polishing stone size and surface, and edge con-
figuration also provide evidence for pot configuration. For example, a nar-
row, rounded edge would be useful for finishing the underside of a rim,
but not the broad side of a jar. Thus, this suite of information gathered
from pottery anvils and polishing stones is evidence for the techniques
used in pottery production. It is important to be able to recognize the dif-
ferences between pottery polishing stones and those used to polish other
items, because at settlements where pots were shaped with the coil-and-
scrape method, the only stone evidence for pot production is polishing
stones. If all polishing stones recovered from a particular site are assumed
to have been for polishing pots, without support from a use-wear analy-
sis, the importance of pottery production may be overestimated. Polish-
ing stones with damage created through contact with more pliable or oily
surfaces may provide evidence for use in shaping wood or bone items.
Those with a lustrous sheen may have been used for polishing stone items.
Evidence for axe finishing may be overlooked if all polishing stones are
assumed to have been for polishing pots.
Several issues could be addressed with more directed research on
grooved abraders. Are there recognizable technological developments in
smoothing and shaping shafts with grooved tools? Is there a relationship
between groove orientation and material texture? Are there relationships
between burning or heat-cracking, material texture, and groove orienta-
tion? Are there regional differences among distributions of certain grooved
abrader designs? Is it possible to identify what types of shafts were pro-
cessed in specific grooves? Answers to these questions might be useful for
distinguishing more-meaningful subtypes of grooved abraders and for
understanding the various activities involving shaft tools.
Rinaldo (1959:249–250, 1964:75–78) chose to address variation in the
tools he called “arrow-shaft straighteners” by classifiying them according
to groove direction and the shape of the stone: (1) transversely grooved—
groove placed perpendicular to the length of the stone; (2) triangular—
groove placed across the apex of a triangular stone; and (3) ridged—a
raised ridge positioned perpendicular to the groove. Rinaldo (1964:78) sug-
gested that the three types represent a developmental sequence because
they seem to have occurred sequentially in the archaeological record at
Carter Ranch and Foote Canyon pueblos, but no attempt has been made
to further address his suggestions.
An analysis of grooved abraders from Point of Pines settlements
(Adams 1994a:100–106) revealed not only differences in groove orientation,
but also in groove width. The largest percentage (41.4 percent) have one or
more grooves across the width of the stone; 37.9 percent were placed length-
wise. A small percentage (6 percent) were grooved diagonally, and 3.4
percent had two grooves, one placed lengthwise and the other widthwise
so they intersect at right angles. It is interesting that the design of so many
of these abraders did not always optimize possible groove length, which
might have some as yet unrecognized functional or behavioral implication.
Measurements taken on 134 grooves from seven villages in the Point
of Pines area (Adams 1994a: Table 5.7) might provide evidence for the
types of shaft tools worked in the grooves. Groove widths range from 0.3
cm to 3.5 cm. The measurements of wooden items from the nearby cave
sites of Red Bow Cliff Dwelling, Ash Flat Cliff Dwelling, Tule Tubs Cave,
and Pine Flat Cave (Gifford 1980) were compared to determine if some
correlation could be made between artifact diameter and groove width.
The larger groove widths of 2.5 cm and 3.5 cm are equal to the diameters
of digging sticks. The smaller groove widths—0.4 cm, 0.5 cm, and 0.8
cm—are equal to the diameters of wooden awls. The pahos (prayer sticks)
found in the caves ranged in diameter from 0.4 cm to 1.1 cm, easily within
the groove widths of some abraders. Weaving tools measured 1.3 cm in
diameter, a flute 1.1 cm, and arrows 0.6 cm to 0.8 cm. Thus, variations in
groove widths could reflect the variety of wooden items shaped by
grooved abraders. It is interesting to note that arrows found in the nearby
caves were made of reed and were the only shaft tools not shaped by an
abrader (Gifford 1980:94), meaning an assumption that grooved abraders
reflect arrow-making technology is inappropriate for this study area.
Pigment technology is another area with tremendous potential for ad-
ditional research. The evidence we have to work with includes: (1) the raw
materials gathered for grinding into pigment; (2) the tools made specifi-
cally for reducing raw materials into powder or for mixing powders into
paint; (3) tools used secondarily for grinding pigment and making paint;
and (4) the processed powders and formed pigment cakes. As described
in Part 2, pigment processing technology is well documented in the ethno-
graphic literature. Carefully organized research, including petrographic
analyses of the minerals and organic analyses of the binders or other ad-
ditives in paints, would make a tremendous contribution to our under-
standing of technology in general. Recognizing which tools were used in
pigment processing is not always easy. Sometimes the pigment has been
worn away except for traces that remain in the deepest interstices of the
stone. Care must also be taken in distinguishing naturally occurring de-
posits of minerals from those intentionally ground for pigment. See, for
example, Logan and Fratt 1993, which describes experiments and micro-
scopic analyses conducted to find attributes distinctive to natural and cul-
tural deposits of minerals on ground stone tools.
Group Affiliation
The ability to distinguish ethnic or sociocultural groups by their material

culture may be one of the most exciting contributions of the technological
approach to ground stone analysis. As suggested previously, technology
is structured by self- and group-interests. It is a problem-solving tool
that chooses from among culturally derived solutions. In what ways can
ground stone assemblages be assessed to identify technological traditions
specific to particular groups?
Perhaps we should start with the basic focus of design theory. Mater-
ial selection and modification involve choices made from a myriad of de-
sign specifications and constraints to solve the problems of achieving a
specific end product. These choices are made within a system of values
specific to a particular group, and resultant tool designs reflect those values
(Horsfall 1987). Tool design is the embodiment of technological knowledge.
Methods for understanding technological knowledge were outlined by

Schiffer and Skibo, who define three basic components of technological
knowledge: recipes for action, teaching frameworks, and techno-science
(Schiffer and Skibo 1987:597–598). Briefly, recipes for action are the param-
eters within which raw materials are turned into working tools for use in
particular activities. A recipe for action is a summary of the materials,
tools, and behaviors needed to accomplish an activity.
Research on materials acquisition, quarrying activities, and manufac-
turing techniques provides some of the data required for reconstructing
recipes for action. For example, the quarry studies by Schneider (1996)
recognize the processes of selecting appropriate material, of selecting the
proper tools for manufacture, and of using those tools to rough out the
preferred tool configurations. Her data set is derived from the archaeo-
logical contexts of the tools and use-wear analyses. An example using eth-
nographic data is Hayden’s edited volume on Guatemalan lithics (Hay-
den 1987b). The chapters devoted to metate manufacture and use evaluate
mechanical, physical, and sociocultural factors influencing tool design
and use.
Teaching frameworks facilitate the intergenerational transmission of
recipes for action. While these transmissions may be verbal or nonverbal,
nonverbal transmissions may be the most common and may take the form
of demonstrations, apprenticeships, and imitation. This is how technolog-
ical traditions are created. For example, food grinding is a gender-specific
task. In the U.S. Southwest it is historically women’s work and is presumed
to have been that way in prehistory as well. We have many ethnographic
models from Puebloan and non-Puebloan groups to support this assump-
tion (Bartlett 1933; Euler and Dobyns 1983; Hough 1915; Simpson 1953;
Spier 1933; Stephen 1936; Underhill 1979). There are accounts of mothers
teaching daughters to grind, and of future daughters-in-law having to
grind to the satisfaction of potential mothers-in-law. In these situations,
transmission of technological knowledge is intergenerational. One possi-
ble refinement of Schiffer and Skibo’s (1987:597–598) definition of techno-
logical knowledge is the recognition that teaching frameworks are not
solely intergenerational transmissions: there are also opportunities for teach-
ing frameworks to operate interculturally through social mechanisms
such as intermarriage or migration, and perhaps even through slavery.
No matter how practitioners of one technological tradition become aware
of an alternative tradition, at some point they decide to adopt the alterna-
tive or not.
Repetition of a particular design is the clearest indicator of teaching
frameworks, especially if a particular design continues in use even when
faced with alternatives. The differing designs of axes and metates discussed
in Part 2 are the best examples of this. Those who learn to make a particu-
lar groove configuration on their axe head or design a particular metate
configuration can decide to change or not when made aware of alterna-
tives. Such a situation happened in the Point of Pines area when Anasazi
migrants brought axes and food-grinding equipment of different designs
(Adams 1994a). If the alternative design is more efficient, easier to hold,
or more durable than the existing design, then the decision to change is
primarily functional, and the decision to not change is sociocultural. If
the alternative and existing designs are equally efficient, easy to hold, and
durable, the decisions to change or not are perhaps both sociocultural.
This is a simplification of a very complex topic that has received much
theoretical attention, but the point of this discussion is to demonstrate
that specific tools can be analyzed in a manner that is useful for addressing
sociocultural issues. The ground stone analyst must look for and record
attributes that address design and redesign. Then analysis techniques that
identify patterns and changes in those patterns through time and across
space can provide a broad perspective on the development of technologi-
cal traditions.
Techno-science is the implicit understanding of the how and why things
respond to the application of recipes for action (Schiffer and Skibo 1987:
597). Techno-scientific understanding may become explicit only through
the application of scientific methods of observation and experimentation.
These are analytical constructs that may or may not be recognized by
technology practitioners. However, these analytical constructs make it
possible for us to understand whether or not certain designs are function-
ally superior to others, or how use alters a tool’s configuration from its
original design.
Lemonnier (1992:19) concludes that a “study of the relations between
technology and society must necessarily start from the study of differ-
ences, of variations in technological actions.” Following this logic, the best
ways to assess whether archaeological material provides the necessary in-
formation for studying the relation between technology and society, and
to recognize distinct technological traditions, are to assess variations both
within and among assemblages. From a technological perspective, the at-
tributes most appropriate for comparison are those related to design and
kinetics, with design relating to the broader sociocultural issues, and ki-
netics to more-individualized tool use.
Individuals
Archaeologists only occasionally address the role of individuals in pre-

history, but ground stone tools offer unique opportunities to highlight in-
dividual behavior. For example, a tool is manufactured by an individual.

Even if the design is culturally derived, the individual decides if the tool
will be used long enough to warrant the manufacture of comfort features.
The individual decides whether the natural texture of the stone is suffi-
cient for the task, or if it needs alteration before it can perform as required.
An individual’s movements work a stone tool against the contact surface
controlling the amount of pressure exerted and the direction of use. This
is an area that is wide open for research. Can we recognize through use-
wear or kinetics studies when more than one individual used a tool?
ANALYSIS STRATEGY
Ultimately, an analysis strategy should be structured to maximize the in-

formation obtainable from a specific collection to meet the needs of the
project’s research design. Not all ground stone assemblages are created
equal. Some are better suited to addressing research questions about food-
processing activities, and others for addressing questions about manufac-
turing activities. It is not always possible, however, to determine the best
research focus until well into the analysis. For example, at the Early Agri-
cultural site of Santa Cruz Bend, there were many more manos of differ-
ent designs than there were metates. It was not possible to determine much
about the food-processing activities by looking only at metates (Adams
1996); most of those recovered were of basin design and were found in ex-
terior pits. However, most of the manos had been used in flat/concave
metates and were found on pithouse floors or in pits in the floors. The
conclusion was that most of the metates had either been removed or were
located in an unexcavated portion of the site. In contrast, the assemblage
from a Pueblo site near Point of Pines (W:10:51) had an extensive floor as-
semblage of metates in bins, and assorted manos scattered across the
floors of several structures. Many of the manos were compatible with par-
ticular metates in each structure. From this assemblage, we can learn
much more about food-processing behavior than from the assemblage at
Santa Cruz Bend. To minimize speculation, analysis of each ground stone
assemblage must be tailored so that specific questions can be reasonably
answered with the data available.
This manual is intended to serve as an introduction to the analysis of
ground stone, a large class composed of many items used to perform a
wide range of activities. It is by no means the last word on the subject.
Many types and subtypes are not even mentioned. A technological approach
to ground stone analysis allows for growth and development of analysis
techniques as our understanding of how prehistoric tools were used in-
creases. We are not limited by the constraints of “form equals function.”
Instead, we can record and study attributes that relate to design, manufac-
ture, primary and secondary uses, and we can incorporate archaeological
context into the interpretive process.
The technological approach to analysis also benefits from building on
the research of disciplines outside of archaeology. The use of ethnography
as a model for archaeological research has been at times controversial,
mostly because of problems with analogic reasoning (see, for example,
Gould and Watson 1982; Wylie 1982). Ethnography, as discussed in the
next chapter, is an integral part of a technological approach to ground stone
analysis because of the range of interpretive models that can be con-
structed through comparative research. Principles and mechanisms identi-
fied through the science of tribology have contributed to our understand-
ing of how surfaces wear. The scientific method employed in experimental
research substantiates what might otherwise be speculation about how
tools were made and manipulated. Most important, this manual builds on
more than a hundred years of research that has paved the way for further
developing our understanding of prehistoric life.
This chapter has outlined a method for analyzing ground stone in a
manner that brings together old and new concepts. In conjunction with
the descriptions presented in Part 2, the goal is to integrate a relatively
new interest in social agency (sensu Dobres and Hoffman 1999b) with
methodological issues of science and experimentation, and the founda-
tional reconstructions of culture history—sort of a technology and tribol-
ogy meet the time/space continuum. Social agency is a useful concept
because it foregrounds individuals, sometimes called “social agents” or
“technical agents” (Dobres and Hoffman 1999a:8). Confronted with the task
of designing a tool, an individual can choose a design from the existing
technological tradition, choose to experiment with new designs, or choose
a design from a foreign technological tradition. The challenge for archae-
ologists is to be able to recognize developmental change, a natural growth
process, as distinct from acquired technology, a process of technological
interaction.
Chapter 3
Resources for Modeling Tool Use

and Technological Behavior
T here is a broad range of resources useful for modeling ground stone

tool use, including photographic archives, traditional written ethnogra-
phies, and ethnoarchaeological experimentation. Each resource provides a
distinct type of information, and certain caveats must be observed while
building analogies that bridge from these sources of information to inter-
pretations of prehistoric behavior. The use of analogy as a strategy for in-
terpreting the past has a long history of debate and comment (see, for ex-
ample, Gould and Watson 1982; Stahl 1993; Wylie 1982, 1985).
For interpreting the range of technological behaviors associated with
ground stone objects, we must take advantage of every device possible.
We must continually look for ethnographic and experimental examples of
tool life histories, and contexts of use and discard. Furthermore, we must
evaluate how each example may be similar to, or different from, the vari-
ous patterns found in the archeological record. Expressed in the terms com-
mon to the debate on the use of analogy, we must develop source-side cri-
teria for analogue creation and develop subject-side criteria for analogue
evaluation (see, for example, Stahl 1993:243–244 and Wylie 1985:101–105).
The key concepts for successful uses of analogy are comparison and con-
trast (Stahl 1993:246–252; Wylie 1985:97–105). What in the archaeological
record looks similar to what was modeled from the ethnographic record;
what is not similar? Can the same patterns be found in other ethnographies
or in other archaeological contexts? Sometimes the patterns that show up
in the archaeological record at one site and not in another, or not in the
ethnographic record, can be used to illustrate technological development
or geographical/environmental adaptations. As previously mentioned,
the literature is extensive and sometimes contentious on the uses and
abuses of analogy as a source for archaeological interpretation. In Part 2
59
the specific ethnographic and experimental models used to recognize pre-

historic technology and associated behavior are discussed in more detail.
PHOTOGRAPHS
Photographs, especially, are important sources of ethnographic and archae-

ological information, yet perhaps one of the most underused. For some
photograph collections this might not be a bad thing because photos must
be used with caution. As noted by Judd (1959:44–45), the photograph of
prehistoric grinding bins at Pueblo del Arroyo shows metates as they were
placed in position by the workers, not as they were found. Although use-
ful for recording the number of grinding bins and the range of variation
in metates, the photograph does not document which metates were used
in the bins. How many similar archaeological photographs have been staged
in other reports is a question that probably cannot be easily answered.
The same caution should be applied to ethnographic photographs.
For example, Edward S. Curtis staged photographs for the express pur-
pose of illustrating vanishing traditional lifestyles, not for scientifically
recording data (Fowler 2000:324). In either case it is important to remem-
ber that photographs need to be interpreted and not just accepted at face
value. Some photographers who have created useful photographic records
of Native Americans in the U.S. Southwest are Kate Cory, Frederick Mon-
sen, Adam Clark Vroman, and Laura Gilpin (Fowler 2000:363–364).
ETHNOGRAPHY
Traditional ethnographies provide a wealth of information that must be

gleaned with an awareness of the different jargons and varying levels of
distinctions used in describing tools. Sometimes tools are mentioned in
the course of describing an activity, but there is little actual description of
their form or material. The descriptions in Part 2 clarify some of the con-
fusions created by differing jargons and unclear descriptions. For exam-
ple, the noisemakers that are spun at the end of a long rope and some-
times called “bullroarers” are almost always made of wood; none of the
ethnographic reports mention any made of stone. However, the noise-
makers that look like two-hole buttons with a loop of string through the
holes are made of a variety of materials, including stone. Both types have
been referred to in various ethnographic descriptions as “whizzers” and
“buzzers.” The ground stone analyst must therefore carefully read the de-
scriptions to recognize which noisemaker is actually being described and
to determine if material type is even an important design characteristic.
3. M ODELING T OOL U SE 61
Ethnographic descriptions of tool use are valuable for their insights

into the physical and sociocultural implications of who uses the tools
when and where. Sometimes an ethnographic report may contain only a
sentence or two useful for understanding traditional stone technology. The
ethnographies that do impart technological information may describe the
mechanics of tool use, the locations of use, who uses the tools, and the rit-
uals surrounding tool manufacture, use, and discard. Given appropriately
detailed descriptions, a ground stone analyst can build one or more plau-
sible models of what archaeological patterns should be expected if the
same social/cultural/technological behaviors occurred in the past. For
example, if an analyst wants to model prehistoric food-grinding behavior
of the Western Pueblo, he or she can read descriptions of historic Puebloan
groups. Among most Puebloan groups some food-grinding tasks are so-
cial opportunities shared among several women who, although somehow
related, contribute their daily chores to different households (Bartlett 1933:4;
Hough 1915:62; Parsons 1939:21; Stephen 1936:882). Space in a particular
room is dedicated to the activity, and permanent grinding stations accom-
modate multiple grinders who can work simultaneously. If similar behavior
occurred in the prehistoric past, we could expect to find multiple grind-
ing stations positioned in locations common to more than one household.
Alternatively, among other historic non-Puebloan groups food grinding is
done by individuals working in or just outside their own houses (Spier
1933:52; Underhill 1979:67, 82). If food-grinding tasks were similarly indi-
vidualized in the prehistoric past, we would expect to find a single grind-
ing station somewhere within the household’s interior or exterior space.
When reading ethnographic descriptions, archaeologists must recognize
the range of possible behaviors that could have been associated with spe-
cific tool use. The ethnographic record should be gleaned for models of
possible behaviors, not for examples of the exact behaviors. Comparisons
with the archaeological record should also accommodate the recognition
of things, contexts, and associations that are not similar.
Clearly there are differences in technological traditions among mod-
ern Native American groups just as there were differences in the prehis-
toric past. Yet, for those studying ground stone, even today there are op-
portunities to watch a potter polish pots, or to talk with a woman who
uses traditional methods to grind maize and cook piki, tortillas, or breads.
Knowledge about some activities and tools is unfortunately lost forever
because not only are they not used today, but there are no written or pho-
tographic records of them. For example, there are no early historic de-
scriptions or photographs of stone axes being used to fell trees in the South-
western United States. Stone axes were so quickly replaced by metal ones
that by the time ethnographers began writing about traditional cultures,

stone axes were occasionally used as hammers or were relegated prima-
rily to ceremonial contexts. Other stone tools that were probably also rap-
idly replaced by metal tools include awls, reamers, files, and saws (J. Adams
1979:76; E. Adams and Hull 1980:23). Lacking U.S. sources, archaeologists
can also look for ethnographic analogies elsewhere, such as in Australia
or Africa, or conduct experiments to replicate specific tool use. Because
traditional ethnographers did not always describe material culture in a
manner useful to archaeologists, the subdiscipline of ethnoarchaeology
developed methods for framing research questions that can be explored
in traditional and nontraditional cultural settings. Experimentation has
also become a useful tool for modeling prehistoric behavior.
EXPERIMENTAL REPLICATION
Once the ethnographic literature has been reviewed for appropriate mod-
els, the analyst may want to evaluate the wide range of culturally based
solutions to similar problems with experiments that replicate the tools
and motions of use. Replication studies are useful not only for determin-
ing the best solution, but also for recognizing that sometimes the prehis-
toric technician made unexpected choices, or choices that created only a
satisfactory solution rather than the best solution (as discussed in the sec-
tion on design theory in Chapter 1). Replication studies, now a common
technique for understanding how things work, have been a learning method
for more than a hundred years (Vaughan 1985:3–6). Flaked lithic technol-
ogists have been much more aggressive with experimental research on
use wear, wear rates, and kinetics than have ground stone technologists
(see, for example, Hayden 1979; Hayden and Kamminga 1979; Keeley
1980; Unger-Hamilton 1984; Odell and Odell-Vereecken 1980; Tringham et
al. 1974; Vaughan 1985). In the 1970s and 1980s, the emphasis was on con-
trolling as many variables as possible, restricting the types of strokes, and
counting the number of strokes before certain types of wear became visible
on flaked edges or before retouch was needed. Some studies confronted
the problems of identifying use wear specific to different material types
(see Vaughan 1985 for a summary of these studies).
The relatively few experiments conducted with ground stone tools
have primarily sought to establish baseline patterns for use-wear compar-
isons with prehistoric tools, and to answer questions about wear rates, ef-
ficiency, and kinetics. For some tools, such as manos and metates, there
are fairly detailed ethnographic descriptions of various grinding pro-
cesses, and these can be used as guides for designing experiments to com-
pare and contrast various tools and techniques. For other tools (for example,
tabular tools) we may have to imagine how they were used and design
experiments that include a broader range of uses than would be necessary
if we had ethnographic descriptions as guides.
Two basic types of experiments have been designed to research is-
sues specific to grinding technology: those conducted in controlled labo-
ratory settings (see, for example, Wright 1993) and those conducted in
settings more typical of prehistoric use (see, for example, Mills 1993). Each
experimental setting has its advantages. Choice of setting should depend
on the research question being explored. For example, questions concerning
wear rates need to hold constant as many variables as possible to fairly
evaluate differences in material type, stroke type, stroke rate, surface area,
and so on. Mechanical devices can be employed to control and measure
changes (Wright 1993), but questions concerning kinetics need to replicate
how the tools were used by people in varying “natural” settings (Adams
1999; Mauldin 1993; Mills 1993). Some experiments, such as those designed
to explore efficiency, may require tests in both types of settings.
Experiments that attempt to replicate prehistoric tool use can be fur-
ther divided into those that employ native technicians who are presum-
ably knowledgeable about specific tool use (see, for example, Mauldin
1993) and those conducted by the analyst (see, for example, Adams 1999).
Replicative experiments such as those conducted with axes (Mills 1993)
and with manos and metates (Adams 1993a, 1999; Wright 1993) give re-
searchers the opportunity to use the tools themselves and come to under-
stand kinetics and use-wear patterns. Those conducted by native users
bring into play their expertise on technique (Mauldin 1993). There has
not yet been an evaluation that compares experiments conducted with
native participants to those conducted by analysts, and an experienced
tool user may very well produce different results than those of a novice;
documenting this distinction may have utility as well for evaluating the
archaeological record. The following brief summaries are examples of ex-
perimental research in both natural and laboratory settings conducted by
native technicians or analysts.
One experimental study involved a Bolivian grinder who replicated
possible prehistoric strategies for processing foods (Mauldin 1993). The goal
was to understand tool efficiency as it relates to tool morphology. Mauldin
(1993:319) achieved his goal by directing the woman to grind wheat, maize,
and Chenopodium with four different sizes of manos so that he could meas-
ure and compare results. The data documented that the woman processed
the foods faster on the tools with larger surface areas. Because there was
also patterning among the sizes of manos at different types of Bolivian
settlements (main houses, pastoral, and agricultural), Mauldin concluded

that mano size is a good predictor of subsistence strategy in Bolivia
(Mauldin 1993:318–320). The Bolivian grinder provided the source-side cri-
teria for a model that Mauldin (1993) applied to the archaeological record.
Additional comparisons with other archaeological settlement systems in
Bolivia would strengthen his subject-side evaluation of the model.
Lacking local ethnographic references for axe use, Mills (1993:395)
gleaned some from ethnographies of New Guinea and then created source-
side criteria for modeling possible prehistoric axe uses in southwestern
Colorado. The cross-cultural models he chose, in conjunction with exper-
iments and use-wear analyses, provide multiple lines of evidence for dis-
tinguishing the activities in which prehistoric axes were used (Mills 1993:
408–410). Mills (1993) compared prehistoric axes from Sand Canyon Pueblo
(in southwestern Colorado) to replicated axes used experimentally in a
variety of contexts. He manufactured axe heads, hafted them, and worked
with them to chop different species of trees and to grub sage out of the
dirt. His goal was to explore the possibility that the range of use wear vis-
ible on prehistoric axe heads was indicative of the range of activities in
which they were used. After comparing the experimental patterns with
those on the archaeological axes, Mills concluded that more of the prehis-
toric Sand Canyon axes had use-wear patterns similar to the replicated
axes used to grub sage than to those used to chop wood (Mills 1993:409–
410). He concluded, however, that additional experiments are needed to
enhance the insights gained from his studies (Mills 1993:410). Mills used
experimentation and ethnography to develop source-side criteria for model
building. His subject-side evaluation found some possibly interesting re-
sults, but he opted to hedge his conclusions pending more work.
Mills conducted his experiments in the natural environment. Wright
(1993) needed more control over environmental variables so that she could
focus strictly on the mechanisms of wear and measure wear rates. For her
analysis she made a mano-metate toolkit comparable to prehistoric toolk-
its in southwest Colorado and used them in a laboratory setting. Wright
(1993:346) proposed to quantify experimental wear rates to enable the mod-
eling of prehistoric wear rates, thereby determining life history and dis-
card rates. She used traditional Puebloan ethnographies to model kinetics
and duration of use, and her replicated tools were designed to simulate
the average size and material type of the prehistoric tools found at Duck-
foot in southwest Colorado. She documented that manos and metates of
certain materials do wear at measurable rates, and that manos wear out
faster than metates (Wright 1993: Table 2). Ultimately, she concluded that
determining prehistoric wear rates is next to impossible because of “the

inability to determine the initial weight and thickness of prehistoric manos
and metates before they were used to grind maize” (Wright 1993:353).
Her conclusions do not mean that her experiments failed, but that our
concepts of what can be quantified need modification.
The above examples are exploratory experiments useful not only for
setting a baseline against which others can be compared, but also for
identifying problems yet to be addressed. They show us that conceptual-
izations about how things might work cannot substitute for actual behav-
iors. Actually using the tools forces archaeologists to frame their questions
more carefully.
DESIGNING EXPERIMENTAL RESEARCH
There is no “best” way to design experiments for solving archaeologically

derived research questions. The following guidelines are distilled from
personal experience and general references on designing scientific exper-
iments (Diamond 1989; Robinson 2000). The usefulness of the guidelines
is illustrated with the description of a set of experiments organized to
demonstrate how tool configuration reflects how tools were designed and
used (Adams 1999:484–487, 1993a:337–340). One goal of the experiments
was to identify use and wear variables specific to tool design that can be
compared to those on items found in the archaeological record. A second
goal was to quantify output amounts to compare the relative efficiency of
each design.
With exploratory experiments it is appropriate to deliberately vary
factors just to see what happens—to see what the response variables are
(Robinson 2000:30). Factors are independent variables (things that can be
varied), such as selecting either coarse-grain or fine-grain material. Re-
sponse variables, as dependent variables, are indicators of a process. For ex-
ample, the striations or wear facets caused by one surface moving against
another are indicators of wear. The objective of an exploratory experiment
is to identify a starting point for the next level of experimentation with
the recognition of which response variables were created by particular
factors. If there is uncertainty about which factors caused which responses,
then a series of experiments should be designed that alternately hold all
but one factor constant. Finally, confirmatory experiments should be run
to verify that the results are replicable and not just random products. The ex-
periments used as an example in the following guidelines were exploratory,
and many more questions were generated than answered.
Guidelines
1. Define the problem. The first step in designing experimental research

is to define the problems to be addressed. For the following experiments,
I wanted to compare three specific mano and metate designs (Figure 3.1)
to learn about the strokes and other possible behaviors associated with
their use (hereinafter referred to as the design experiments). Three ques-
tions were posed: (1) Are tool designs and kinetics related in a way that
makes one design more comfortable to use than the other two? (2) Is one
design capable of grinding more product during a set amount of time,
making it more efficient than the others? (3) Are certain seeds and kernels
easier to grind with one design than the others? These questions address
certain understandings that have become accepted by archaeologists
without empirical confirmation, including beliefs that coarse-grain mate-
rials were selected for manos and metates used to grind maize, and fine-
grain materials were selected for those used to grind gathered grains and
seeds. Archaeologists have also assumed that trough manos and metates
were designed for grinding maize, and basin manos and metates for
grinding gathered grains and seeds.
2. Decide on an experiment strategy. Have there been previous ex-
periments upon which this experiment can build? If so, the archaeologist
must either select different factors to vary and evaluate those response
variables, or replicate the experiments to either confirm or confound the
previous results. If there are no prior experiments, a strategy including a
series of exploratory experiments is probably the best way to select rele-
vant factors. Even though other experiments had been conducted with
manos and metates (Mauldin 1993; Wright 1993), their problem orienta-
tion was different and did not identify factors appropriate to the ques-
tions to be answered with the design experiments. Therefore, the selected
strategy for the design experiments was one of exploration.
3. Select the factors and response variables that will be measured and
observed, and enumerate expectations for those selected. Once the exper-
iments begin, it may become immediately obvious which factors are im-
portant and which are inconsequential. Those that are inconsequential
can be dropped from further evaluation. For my design experiments I se-
lected a basin metate made from granite, a flat/concave metate made
from sandstone, and a trough metate made from vesicular basalt. The ma-
terials to be ground were selected to include a range of large and small
sizes, and oily and dry textures: amaranth seed, shelled raw sunflower
seeds, blue popcorn kernels, dried dent kernels, and soaked white flint
kernels (nixtamal, which is processed with lime). It was decided that dif-
3.1. Experimental manos and metates used to grind dried feed corn: (a) basin
design; (b) flat/concave design; (c) open-trough design.
ferences in design would be evaluated by measuring the amount per hour

of processed seeds and kernels. Other observed response variables in-
cluded the types of use-wear damage on the tools’ surfaces, and the types
of strokes used with each tool design. These responses were not quantifi-
able, but qualitative evaluations were made at the end of each experiment
run.
It was expected that the trough set would grind most efficiently be-
cause the surface area between the mano and metate was the largest of
the three designs. It was further expected that the basin set would be the
least efficient because the contact surface area is smallest. These expecta-
tions were based on previous experiments, such as those reported by
Mauldin (1993), who determined that mano size correlated with relative
efficiency. Because the basin mano is worked with only one hand, it was
expected to be the most tiring to use. The trough metate was expected to
be the least tiring because it was assumed to be more efficient. Expecta-

tions about material texture were that the coarse-grain material would be
the most efficient because it has the roughest surface texture.
4. Design a system for recording data. Pertinent information should
include the name of the experimenter, the date and time of the experi-
ment run, what happened during the run, and records of all quantifiable
observations. For the design experiments, tables were created with
columns for the initials of the experimenter, the experiment date, the time
at which the experiment run started and stopped, how much seed was
worked, and the quantity of processed meal. Additional notes were kept
describing the stroke used with each mano/metate set and the muscles
involved in the stroke. Special attention was given to where pressure was
exerted on the mano and how that related to the observed use-wear pat-
terns.
5. Perform the experiment runs and record the data. The design ex-
periments began with the careful measuring of equal quantities of each
seed or kernel type into each mano/metate set. The strokes used with the
manos and metates imitated those described in the ethnographic litera-
ture and those described to me by Hopi grinders. It was recognized that a
novice grinder may grind differently than an experienced one, and this
variable may be subject to future experimentation. Hopi women who
have ground food all their lives emphasize using rhythmic strokes and
using the whole body, not just the arms. After an hour of grinding with
each set, the nature and quantity of the product were evaluated, and ob-
servations about strokes and ease of grinding recorded. The experiments
continued for several hours, but the reported observations (Table 3.1)
used an hour to standardize the results.
6. Evaluate expectations. Did the results meet your expectations?
Why or why not? Was the experiment design faulty? My experiment test-
ing design revealed that seed grindability and ease of use varied with
each mano/metate design, but not quite as expected. Everything was eas-
ily ground with both the basin and trough sets, but with the flat/concave
mano and metate, the dried seeds kept falling off the surface. No matter
which tool was used, it was clearly easier to create a fine-textured flour
from soaked kernels than it was from dried kernels. However, even from
dried kernals, some fine flour was produced with all tools.
This experiment made it possible to observe that grinding with a
basin mano/metate set was the most tiring, as was expected. Further-
more, even though the trough design was more restrictive to specific mo-
tions and thus fatiguing, it was the most efficient for grinding dried foods.
Flat/concave and trough manos and metates were perhaps equally efficient
Table 3.1.
Results of grinding experiments conducted with different
mano/metate designs.
Basin Metate Flat/Concave Metate Trough Metate

(Granitic) (Sandstone) (Vesicular Basalt)
Dried Kernels
Amount/hour 1 cup 3/4 cup 2 cups
Results 75% fine flour; 25% 75% coarse meal; 90% fine flour; 10%
coarse meal 25% fine flour coarse meal
several rock fragments several rock fragments rock dust
Soaked Kernels
Amount/hour 2 cups 4 cups 4 cups
Results 75% masa; 25% fine 75% masa; 25% fine 90% masa; 10% kernel
flour flour fragments
few rock fragments no rock fragments no dust
Popcorn
Amount/hour 3/4 cup 3/4 cup 1 cup
Results 75% coarse meal; 100% coarse meal 75% coarse meal;
25% fine flour 25% fine meal
Sunflower Seeds,
raw shelled
Amount/hour 1-1/2 cups 2-1/2 cups 3 cups
Results 75% paste; 25% seed 75% paste; 25% seed 75% paste; 25% seed
fragments fragments fragments
no rock fragments no rock fragments no rock dust
Amaranth Seeds
Amount/hour 1-1/2 cups 3/4 cup 2 cups
Results 75% fine flour; 25% 50% fine flour; 50% 75% fine flour; 25%
seed frags seed frags seed fragments
Note: Grinding rates should not be considered indicative of prehistoric rates.

Source: Adams 1999: Table 2.
for processing soaked kernels and oily seeds. This was not an expected re-
sult. Nor was it expected that granular and vesicular materials would be
equally efficient in reducing seeds and dried kernels to flour, at least until
the granular material was worn smooth, when vesicular material was su-
perior. The conclusions suggested by these experiments are that vesicular
material is a better choice for grinding seeds and dried kernels because it
does not add rock grains to the flour (Adams 1999:487), and the flat/con-
cave design is the most comfortable to use for an extended period of time.
7. Communicate results. There is a certain responsibility to share the
results of experiments, even if they are only exploratory, so that others
can build on them and avoid unnecessary duplication. The results of my
design experiments were reported in American Antiquity (Adams 1999)
and communicated through several conference papers and lectures in
classes and public forums.
8. Evaluate the experiment and reformat the procedures to improve
the next series. Probably the most important step in the exploratory
process is the evaluation of what went wrong and what went right with
the individual experiment runs and the experiment process as a whole.
Were the correct factors chosen for observation? Was the selected unit of
measurement the most appropriate, and did it actually measure what it
was intended to measure? What factors need to be varied next, and what
are the expected response variables?
My design experiments identified many important factors that
should be systematically varied in the next series of experiments. Because
material texture does seem to be important, it should be evaluated inde-
pendently from tool morphology. The next series of experiments should
involve a set of each design made from the same material. Grinding times
should also be extended so that we can evaluate efficiency over a longer
period of time. This should bring into play the variables associated with
wear management. In a later series of experiments, more than one grinder
should run the experiments so that comparisons can be made to evaluate
the human agent using the tools. It might also be possible at this time to
decide that there has been enough experimentation: if after a certain num-
ber of runs the results are always the same, we can probably trust that we
have recognized significant relationships among factors. If trough manos
and metates of vesicular basalt always perform more efficiently than
basin manos and metates of the same material, then we can be comfort-
able with our conclusions and find ways to apply these conclusions to
patterns we see in the archaeological record.
The above guidelines and experiment descriptions illustrate how varied

the opportunities are for advancing our understanding of how tools
work, how decisions about material selection affect tool performance, and
how tool design relates to the motions of operation. The resources de-
scribed in this section for modeling ground stone tool use and technolog-
ical behavior are presented as examples of what can be done. There are
many more possibilities that any analyst can create. This is the fun part of
the research process: the analyst can create models and test them for ap-
plicability to the archaeological record, thereby increasing the potential
for expanding our understanding of prehistoric life.
Part 2 of this manual and the appendixes present the working tenets
of ground stone analysis and attempt to organize a very diverse category
of stone items that in reality have very little in common. The forms and
coding sheets in the appendixes are guides for observing and recording
important attributes. Used together, the various sections of this manual
will enable anyone to proceed with the task of analyzing ground stone ar-
tifacts using techniques designed to create meaningful interpretations
about the daily lives of prehistoric people.
Part 2
Artifact Descriptions
C hapter 1 includes a discussion about classification and the usefulness

of fuzzy set theory. The following definitions and descriptions build
on the fuzzy set concept. The purposes for defining types are to denote
which attributes are important for assigning membership to a set or sub-
set, to understand how items were used or how they functioned, and to
track the distribution of items through time and across space. The fuzzy
set concept as presented here is intended to help clarify some problems in
classification, especially those created by conflicts in terminology. Because
archaeologists have learned more since some of the typologies were first
developed, it becomes confusing to try to force new concepts into old def-
initions. Where appropriate, artifact descriptions include different terms
applied to the same object, and a discussion of recent research on specific
items.
The following descriptions emphasize tools used in the U.S. South-
west, but the analytical approach is universally applicable. The initial as-
sumption is that design reflects intended use. This assumption, coupled
with evidence for actual use and recovery context, makes it possible to re-
construct multiple behaviorally meaningful typologies. A technological
analysis moves beyond discussions of what types were present at a settle-
ment to evaluations of people’s interactions with their things. The bridge
between analyzing things and making interpretations about people’s ac-
tions is an assessment of the collective life histories of the recovered
items. Life-history categories (defined in Chapter 2) essentially include
design, type of use, amount of wear, type of use wear, and archaeological
context. Tracking life-history attributes makes it possible to evaluate and
compare these categories through time and across space.
What rapidly becomes obvious in reading the following descriptions
is that archaeologists know more about some artifacts than about others.
Also, certain types are better known or better described for some areas
in the U.S. Southwest. Reference is made occasionally to traditional,
73
74 A RTIFACT D ESCRIPTIONS
archaeologically defined, cultural distinctions such as Anasazi, Mogollon,

Hohokam, Salado, Sinagua, and Coconino. However, during the last 10
years, these terms—once so freely used—have become politically charged.
The goal here is to bridge what was intended by the use of these terms in
the past with what is possibly more politically correct.
The bridging terminology used throughout this manual derives from
the concept of technological traditions as defined in Chapter 2. Instead of
comparing objects and tools common to one group with those common to
another, comparisons can be made at the level of technological traditions.
For example, the Anasazi and Hohokam technological traditions devel-
oped different solutions for grinding food. Anasazi traditions developed
flat manos and metates set up in bins, whereas Hohokam traditions de-
veloped freestanding open-trough manos and metates. The comparisons
include tool design as well as well their position of use. This opens the
door wider for understanding different behaviors and results, even though
the tasks were basically the same, than if the comparisons are at the tool
level (for example, the Anasazi used flat metates, and the Hohokam used
open-trough metates). Then, as we recognize that technological traditions
can be passed by means of group interactions, we can begin to recognize
levels of those interactions. Labels such as Anasazi, Hohokam, and the oth-
ers become more archaeologically recognizable and distinctive as techno-
logical traditions than they are when considered solely as ethnic or group
labels.
The challenge to presenting a behaviorally and technologically mean-
ingful typology is in the structuring the categories. Part 2 is divided into
nine chapters based on what the items were designed to accomplish. The
functional categories include (1) abrading, smoothing, and polishing tools;
(2) grinding and pulverizing tools; (3) percussion tools; (4) hafted percus-
sion tools; (5) spinning tools; (6) perforating, cutting, and scraping tools;
(7) paraphernalia; (8) containers and container closures; and (9) structural
stones. Specific types and subtypes are defined within each category. Some
items can be sorted into more than one functional category and are some-
times distinguishable solely on the basis of use wear. For example, some
netherstones and lapstones may have been used for smoothing or polish-
ing, and others for grinding or pulverizing. The concepts of fuzzy bound-
aries between sets, and set intersections, are more useful in sorting such
types than the mutually exclusive concepts common to traditional scien-
tific classification as discussed in Chapter 1 (Adams and Adams 1991:47).
The items in some sets are further sorted into subtypes, and it is im-
portant to be consistent with classification tactics. For example, a tool that
was secondarily used should be classified according to its original design.
A RTIFACT D ESCRIPTIONS 75
If it is not possible to decide which design was original, it should be clas-

sified as the tool for which there is the most obvious wear or greatest de-
sign effort. Adherence to this kind of protocol will facilitate comparisons
between assemblages classified by different analysts. Finally, some of the
descriptions include discussions of ethnographic and archaeological ex-
amples, and of spatial and temporal patterns. Because research sometimes
generates more questions than answers, suggestions for future research
are also presented. This classificatory system is not perfect and is subject
to refinement as more is learned. The types selected for description here
should not be considered to be all that are known. This is a model for ar-
tifact classification, not an encyclopedia of types or of all sites where par-
ticular artifacts have been found.
4
Abrading, Smoothing, and

Polishing Tools
F rom the perspective of the tool user, abrading, smoothing, and polish-
ing are three distinct activities, each requiring a differently textured
tool. From the perspective of the tool analyst, the boundaries between
these tools are fuzzy, but it is the analyst’s job to interpret which function
(or functions) a particular tool performed (Figure 4.1). In all three activi-
ties tools are used to alter contact surfaces through the mechanisms of
abrasive wear, adhesive wear, and tribochemical wear. Although each one
relies more on one mechanism than the others, all mechanisms come into
play at some point in the process.
Abraders remove material from the contact surface through adhesive
and abrasive mechanisms. The damage is visible macroscopically as stria-
tions on both surfaces with some actual reduction in mass of the contact
surface. The damage to the abrader surface is dependent on the nature of
the contact surface. Hard-contact surfaces flatten the asperities that make
the abrader surface rough, and pliable surfaces round them. (Refer to Chap-
ter 2 for a more in-depth discussion of use-wear patterns.)
Smoothers rely more on adhesive and tribochemical mechanisms than
on abrasive mechanisms. Less material is loosened and removed from the
contact surface than with an abrader, and the loosened material is moved
to another surface location. Smoothers have finer surface textures than
abraders, but not as fine as polishers. The use wear on smoother surfaces
is minute striations and some spots of sheen. The contact surface ends up
with a more uniform texture than a surface worked with an abrader. Be-
cause there are no defined subsets of smoothers, they are not defined here
in any more detail, but are considered a continuum in the polishing task.
The tribochemical interactions between polishers and their contact
surfaces leave a sheen (commonly called “polish”) that is visible on both
the tool’s surface and the surface being polished. When viewed under a
77
4.1. Diagram of abraders, smoothers, and polishers expressed as sets and subsets.
4. A BRADING , S MOOTHING , AND P OLISHING T OOLS 79
microscope, however, most polished surfaces have multiple minute stria-

tions as well as sheen. The boundaries between abrading and smoothing,
and between smoothing and polishing, are fuzzier than the boundary be-
tween abrading and polishing, at least at the macroscopic level of analysis.
Not all abrading, smoothing, or polishing tools are handstones. Some
lapstones and netherstones alter contact surfaces according to their tex-
ture. Starting at the generic level of classification, handstones, lapstones,
and netherstones can be sorted into the functional abrasive or polishing
categories, depending on their surface textures and the nature of any
wear management. For netherstones and lapstones there are no clearly
defined subtypes to distinguish those that polish from those that abrade.
Instead they are classified as netherstones and lapstones, and are described
as having rough or smooth texture. Handstones are usually further classi-
fied as abraders and polishers.
At the design stage, material selection determines if a tool will abrade,
smooth, or polish the contact surface. If suitable material is not readily
available, existing material can be altered, either by smoothing a stone
that is too rough for the task, or roughening a stone that is too smooth.
Thus, manufacturing techniques are evidence of intended tool use, and
wear-management strategies are evidence of actual tool use. For example,
if a rough tool that was worn smooth was then roughened again to main-
tain its abrasive texture, then there is evidence for both intended and ac-
tual use. The factors that define tool design, including material selection
and manufactured attributes, are used to classify each tool. Variables related
to use wear and wear maintenance are used to discuss variations in tool
use.
Sometimes variations in design provide information about distinctive
technological traditions useful for understanding developments through
time or differences across space. The goal of the classification scheme pre-
sented here is to simplify the system and to subsume (under the categories
of abraders and polishing stones) several previously defined “types” that
merely exhibit variations in use and amount of wear.
ABRADERS AND SMOOTHERS
Abraders are handstones that have one or more rough surfaces useful for
removing material from contact surfaces, thereby altering their texture or
modifying their configuration. Previous classification schemes have used
terms such as “rubbing stones,” “smoothers,” “files,” “whetstones,” and
“rasps” for the tools included in this abrader category (see, for example,
Cosgrove and Cosgrove 1932:40; Haury 1976:283–284; Hayes and Lancaster
1975:157–158; Martin et al. 1961:102; Wheat 1954:123–124; Woodbury 1954:

98). Within the larger set of abraders are two general subsets: flat abraders
and grooved abraders (see Figure 4.1). Each subset has design variations,
requires unique technological knowledge, and presents opportunities for
researching prehistoric behavior.
Flat Abraders
The classification of flat abraders is as varied as the number of researchers

studying them. The defining feature, however, is a broad working surface
on stone coarse enough to remove material from the contact surface. Haury
(1976:284) uses the terms “files” and “abrasive blocks” to refer to what are
here considered flat abraders. Woodbury (1954:98) categorizes abraders
by shape (tabular, bar-shaped, and irregular) and uses the length limit of
about 10 cm to distinguish hand-held abraders from netherstones used
for other purposes.
From a technological perspective, the size of the tool is not as impor-
tant as whether or not it was an active or passive tool. An active hand-
stone is an abrader, and a passive stone is either a lapstone or netherstone.
Strategically designed flat abraders are generally manufactured to specific
shapes: rectangular, tabular, triangular, and so on. The shape of an expe-
diently designed flat abrader depends on the shape of the chosen rock until
it has been altered through use (Woodbury 1954:98). The surfaces remain
flat or become convex or concave depending on what they are used against
and how long they are used. Lightly used abraders retain their original sur-
face shape. Well-used flat abraders may end up anything but flat, whether
they are expediently or strategically designed. The sorting of flat abraders
should not confuse design with the shape that results from the item’s life
history.
Some flat abraders have V- or U-shaped grooves indicating a second-
ary use. If the groove is positioned on the flat abrasive surface and there is
no evidence of wear over the groove margins, then the groove indicates
sequential secondary use. If the groove margins have wear over them from
continued use of the flat surface, or if the groove is located on another
part of the tool, then the secondary use is concomitant. Whether the tool
should be classified as a flat abrader secondarily used as a grooved abrader,
or as a grooved abrader secondarily used as a flat abrader, should be de-
termined by the amount of wear: the design that seems to have been the
most used should be considered primary. Certain flat abraders may also
have been used against more than one type of contact surface. For exam-
ple, one surface may have smoothed wood, and another bone or stone.
Such differences are determined through a use-wear analysis (see Chapter

2 for use-wear descriptions). The important point to remember is that the
relative hardness of the contact surface alters the abrader surface in a pre-
dictable manner.
Archaeological and Ethnographic Comparisons
Some regional variations among flat designs can be considered regional

subtypes. For example, in the Point of Pines region, two subtypes are com-
mon: faceted and wedge. Wedge flat abraders are designed by selecting rocks
that naturally have a wedge shape or by manufacturing them to that shape.
The two converging flat surfaces are the tool’s working surfaces. The sur-
faces against which these stones were used were broad, flat, and pliable.
The use-wear damage patterns are most similar to experimental tools used
to shape wood (Adams 1989a:267, 1989b:268). These wedge abraders may
have been used like modern sanding blocks to alter the surfaces of broad
pieces of wood. The wedge shape allows pressure to be exerted both down
and forward as the abrader is moved across a flat surface with mainly flat
reciprocal strokes. Extensive use further enhances the wedge shape.
Faceted flat abraders and polishers are designed by selecting rocks that
naturally have a triangular profile or by manufacturing them to that shape.
The flat facet is manufactured or worn along the apex of the triangle and
is a long and narrow working surface. The base of the triangle is held in
the hand, which provides a large area against which pressure is applied
(Figure 4.2). The tool is primarily worked against the contact surface with
a reciprocal stroke that moves perpendicular to the length of the tool. How-
ever, striations on the facets of some flat abraders are evidence that move-
ment is occasionally parallel to the length of the tool. One interpretation
(Adams 1994a:101) is that at least some of these tools were used in axe
manufacture. The use-wear damage patterns on some faceted abraders
and polishers from Point of Pines area settlements are similar to those on
experimental tools used against other stone surfaces (Adams 1989a:264,
1989b:267). Most were found at Turkey Creek Pueblo in thirteenth-century
deposits that also had many axes in various stages of manufacture and
maintenance. Thus, use wear and archaeological context are both used to
infer tool use in this case (Adams 1994a:101). These abraders are also good
examples of tools whose function should not be determined by form. Tex-
ture is the attribute that distinguishes these similarly shaped abrading
and polishing tools.
Woodbury (1954:98–100) discusses the typological difficulties of defin-
ing flat abraders and their distribution in the Southwest, and he recognizes
4.2. Faceted flat abrader from the Point of Pines area. (Photograph courtesy
of Arizona State Museum, Catalogue Number A15,538. Geoffrey Ashley,
photographer.)
that items of different materials are shaped with flat abraders. Castetter
and Bell (1951:94) note that the Yumans shape digging sticks with a flat,
rough stone. The Maricopa shape their arrows with flat abraders and use
their teeth to straighten shafts (Spier 1933:134). Flat abraders appear to be
ubiquitous in the U.S. Southwest, although there has not been as much re-
search into their geographic and temporal distribution as there has been
for other types of abraders. Experimental research has begun to recog-
nize use-wear patterns distinctive to shaping bone, shell, stone, and wood
items (Adams 1993b). Perhaps systematic use-wear analyses will help
identify more specifically the varied uses of flat abraders and help recog-
nize other regional variations.
Grooved Abraders and Shaft Straighteners
Grooved abraders and shaft straighteners have smaller working surfaces

than do flat abraders. The grooves confine cylindrical or shaft-like objects
during their manufacture. Within the larger set of grooved abraders are
two subsets: those with V-shaped grooves, and those with U-shaped
grooves. Shaft straighteners are a subset of abraders with U-shaped grooves
(Figure 4.1).
Abraders with V-shaped grooves (Figure 4.3) are used to put points on
the end of tools such as awls and needles, and to dull the edges of lithic
tools. Unless the shape of a particular abrader is manufactured, it is of ex-
pedient design. Rinaldo (Di Peso et al. 1974:136–138) identified some
2 V-shaped grooves
groove not to end
dorsal side groove width margin
groove depth
thickness
ventral side
width
length
4.3. Abrader with V-shaped grooves, expediently designed. (Drafted by Chip

Colwell.) (Adams 1996: Figure 7.)
grooved abraders as used to straighten and smooth yarn, or to knock the

rough spines off bear grass or other basket-weaving material. V-shaped
grooves are worn rather than manufactured into the stone, and are usu-
ally the result of working a shaft against a flat surface to create a point. If
loosely cemented sandstone is chosen for the abrader, then the grooves
wear rapidly. In most cases groove depths can be used to compare relative
rates of wear on the same type of material. A well-worn V-shaped groove
can be as deep as 1 cm; a lightly worn groove is barely distinguishable
from the surrounding surface.
V-shaped grooves are often randomly placed on the surface and can
vary in size on the same tool. An extreme example is illustrated by Hayes
et al. (1981: Figure 156) from Gran Quivera, a Pueblo IV site in New Mex-
ico. Not all V-shaped grooves are on handheld tools. Many netherstones
have been found with V-shaped grooves worn in them; however, this is
usually a secondary use, in which case it is suggested that the netherstone
be coded as the original tool design. Some stones have random incised lines
that were probably not used to shape points or dull edges, but whose spe-
cific function is unknown. These are often referred to as grooved stones
and considered ritual or symbolic. Sometimes archaeological context helps
with the interpretations of these as either ritually or functionally created
items.
U-shaped grooves are designed for working slender rods such as wooden
spindles for spinning fiber; drills, wooden awls or other weaving tools;
wooden or reed shafts for arrows or prayer sticks; wooden construction
elements; and stone rods for awls, hairpins, or stone beads (Figures 4.4, 4.5)
(Jernigan 1978: Figure 95; Judd 1954:86, 119; Woodbury 1954:103). Groove
texture determines whether the shaft will be abraded, smoothed, or pol-
ished (Cosner 1951:147–148). If super-fine or cryptocrystalline material is
chosen for the tool, the groove smooths, hardens, and puts sheen on shafts.
For removing bark and leveling unwanted projections, coarser materials
are required. Eventually, use smooths the texture of the groove no matter
what material texture is chosen. The groove must then be reroughened, or
an abrasive agent such as loose sand added, if the intent is the continued
removal of material from the shaft.
Abraders with U-shaped grooves are more likely than those with V-
shaped grooves to be strategically designed. Grooves in relatively hard
stone are generally manufactured rather than worn, and deepen with re-
peated use. Unlike V-shaped grooves, U-shaped grooves are useful for
making inferences about the size of the shaped tool: the shaft diameter
can be no larger than the width of the groove (see Chapter 2, “Answering
Research Questions: Manufacturing Activities”). Groove configuration also
3 U-shaped grooves
groove not to end uniform groove
embellishment
margin groove width dorsal side
groove depth
thickness
ventral side
edge
end width
length
4.4. Grooved abrader or shaft straightener, strategically designed and embel-

lished with two ridges oriented perpendicular to the grooves. Granular material
defines this tool as an abrader; evidence of heat treatment on material of any
texture defines it as a shaft straightener. (Drafted by Chip Colwell.) (Adams
1996: Figure 6.)
4.5. Grooved abraders with stone tools that might have been shaped in the
grooves. (Adams 2000c: Figure 10.1.)
reflects the kinetics of abrader use. If the ends of the groove are deeper or
wider at the ends, then the shaft was held at an angle rather than flat
across the groove. Striations in the grooves indicate direction of use and
the nature of the contact surface or intermediate abrasive agents. If mate-
rial texture is smooth or the grooves were worn smooth, and there are
longitudinal striations in the groove, then either the shaft was relatively
more asperite, or an abrasive agent was added to the groove. The longitu-
dinal striations indicate a reciprocal stroke. If the striations are perpendi-
cular to the length of the groove, then the shaft was rolled in the groove.
A sheen in the groove is the buildup of tribochemical reaction products
that were not abraded away because the worked shaft was as smooth as
the groove.
Stone of any texture modified with a straight, U-shaped groove is
useful as a shaft straightener. However, material selection favors those that
retain heat or withstand repeated heating and cooling. Evidence of heat
alteration indicates the tool was used to straighten shafts (Simon and Rice
1996:571), but it is not clear if heating facilitates, hinders, or destroys the
buildup of residues visible as sheen. Thus, until experiments can identify
other useful attributes, texture and evidence of heating are the attributes
used to distinguish straighteners.
Some of the more carefully shaped grooved abraders and shaft straight-
eners are embellished with one or more ridges or incised lines positioned
relative to the grooves (Figure 4.4). Woodbury (1954:111) cites ethnographic
accounts of Yavapai heating arrow shafts in the groove and then bending
them over the ridge. Rinaldo (Martin et al. 1964:77) also suggests that the
ridge functioned as a fulcrum for bending reeds. Grooved abraders with
ridges were recovered from early Classic contexts within settlements along
Tonto Creek in central Arizona. The location of the ridges on the Tonto
Creek abraders varied from the center of one end, to off-center, to extend-
ing entirely across the abrader, interrupted by the grooves. None of the
ridges had evidence of any use wear (Adams 2000c). Perhaps, at least on
the Tonto Creek abraders, the ridges were simply decorations or identify-
ing marks.
Some grooved abraders, including some with ridges, have either in-
cised lines or short V-shaped grooves (Haury 1945:139, Plate 60a,b; Simon
and Rice 1996:577, Figure 9:21). Some of these lines are shallow with no
obvious use wear. Others are deep and have use wear consistent with the
working of smooth stone. Perhaps this is evidence of their use to dull the
bases of projectile points before they are hafted. If any grooved abraders
are to be identified specifically as arrow-shaping tools, the most likely
candidates would be those that show evidence of heating, have ridges with
use wear from bending pliable shafts, and have worn V-shaped grooves
from the dulling of projectile point bases in preparation for hafting.
Typological Confusion
Woodbury (1954:101–111) takes on the confusing issue of distinguishing

grooved abraders, shaft smoothers, and shaft straighteners. He notes the
conflicting use of jargon and the problems of clearly separating tools that
abrade from tools that smooth. Woodbury (1954:101) purposefully does
not use the term straightener because of these problems, and because any
grooved abrader could be used as a shaft straightener with the addition
of heat. He instead proposes that simple grooved abraders and shaft
smoothers should be distinguished on the basis of degree of tool manu-
facture and, among shaft smoothers, on the basis of manufactured shape:
elongated, loaf-shaped, ridged, and transversely grooved (Woodbury 1954:
101). From a technological perspective, abrading and smoothing are merely
relative degrees of the same abrasive process. A straightener, on the other
hand, is technologically distinct with the application of heat. It is there-
fore a more significant technological distinction to sort straighteners from

grooved abraders by the presence of thermal alteration, and to sort grooved
abraders based on their texture.
Hayes et al. (1981:123) note that the grooved abraders recovered from
Gran Quivera with evidence of heat treatment also have flat bottoms.
They determined that when placed on a heated comal, the flat-bottomed
abraders heat quickly and retain their heat, making them useful straight-
eners (Hayes et al. 1981:123). Perhaps following Kidder (1932:76–80), Ri-
naldo (Di Peso et al. 1974:86; Martin et al. 1964:77; Martin et al. 1961:102,
Figures 53, 64) used the terms “arrow-shaft abraders” and “arrow-shaft
straighteners.” Such terms prejudice our understanding of how these tools
were used, even if accompanied by qualifying statements about how the
tools alternatively might have been used to straighten or smooth other
shaft objects. Thus, it is suggested that “arrow” be dropped from the label
for this tool type.
Archaeological and Ethnographic Distributions
Opinions vary about whether grooved abraders found in the greater U.S.
Southwest originated locally or were influenced by Great Basin or Plains
abrading technology (Woodbury 1954:110–111). The main difference is
that outside the U.S. Southwest, some shaft abrading technology includes
paired stones with a hole through which the shaft is passed. (See Flenniken
and Ozbun 1988 for ethnographic and experimental research on paired
grooved abraders.) Woodbury (1954:109) notes that there is no evidence
among historic Puebloan groups for the use of either the paired or the sin-
gle grooved abrader. However, single shaft smoothers were recovered from
various Jeddito settlements that are similar to those used by Owens Val-
ley Paiute, Surprise Valley Paiute, Yavapai, Havasupai, Maricopa, and nu-
merous southern California tribes (Woodbury 1954:109). A historic photo-
graph shows an Apachean man working with a single handheld, grooved
abrader (Opler 1983: Figure 3). A 1932 Harrington photograph of a Paiute
man straightening a shaft with a heated shaft straightener is reproduced
in Liljeblad and Fowler 1986:419, Figure 3. In this demonstration the stone
was placed flat on the ground.
In discussing the history of abraders, Woodbury (1954:100,104, 110–
111) concludes that the more elaborately shaped shaft smoothers probably
did not occur as early in the archaeological record as did grooved and flat
abraders. He also surmises that some of the previously reported distribu-
tions through time of various abrader types are confused by misidentifi-
cation and lack of excavation of sites dating to certain time periods that
may misrepresent chronological developments (Woodbury 1954:99, 104).
A listing of all the sites and time periods in which various abrader types
have been found is impractical here, but it is sufficient to note that flat
abraders have been found in the earliest Cienega-phase contexts in south-
ern Arizona (Adams 1998:337) in the latest contexts of Pueblo III at Long
House in Mesa Verde (Wheeler 1980:247), in the Medio period at Casas
Grandes, Chihuahua (Di Peso et al. 1974:51–55), and among the historic
Maricopa (Spier 1933:134), Yumans (Castetter and Bell 1951:94), and Hopi
(Adams 1979:41). Abraders with V-shaped and U-shaped grooves are found
in both pithouse villages (Hayes and Lancaster 1975:158) and pueblos
(Wheeler 1980:248) at Mesa Verde, in late Pueblo contexts in the Mogollon/
Zuni area (Adams 1994a:304–305; Martin et al. 1961: 102–103; Roberts
1932:144; Wendorf 1950:63), and among early Classic contexts in upper
and lower Tonto Creek basins (Adams 1995:47, 2000c). They do seem to
be less common in the Phoenix and Tucson basins.
Abraders embellished with incised lines or ridges have been found at
Classic settlements in the lower Tonto Basin (Simon and McCartney 1994:
793), at Los Muertos (Haury 1945:139), at Winona Ruin near Flagstaff (Mc-
Gregor 1941:180–181), at Pueblo settlements in the Jeddito Valley (Wood-
bury 1954:111), in later deposits at Pecos Pueblo (Kidder 1932:79), and in
historic Hopi contexts (Adams 1979:43–45). One style identified as dis-
tinctive to the Gallina area in New Mexico has a ridge the length of the
tool, along the base of which is the groove for straightening. Perpendicu-
larly across the top of the ridge is an additional groove for shaft straight-
ening or smoothing (Hibben 1938:136). Similar groove and ridge configu-
rations are on grooved abraders from Gran Quivera (Hayes et al. 1981:
123, Figure 157). A shaft smoother with similar attributes was found at
Showlow Ruin (Haury 1931a:22, Plate 6, Figure Id, called “a polisher”).
Whether this indicates some connection between the two areas, or that
this configuration is not areally distinctive, needs to be explored.
In summation, within the set of tools that shape shafts there is con-
siderable variation and a wide distribution through time and across space.
The set itself has fuzzy boundaries between the sets of abraders, smoothers,
and polishers, and among handstones, lapstones, and netherstones (Fig-
ure 4.6). When these tools are classified using behaviorally meaningful
terms of design, use, and wear, it is possible to compare assemblages from
across the U.S. Southwest and derive inferences concerning what was made
and how with particular abraders. Some of the models useful for compar-
isons were derived from ethnographies, some from comparing archaeolog-
ical contexts, and some from experimental research.
Other interesting questions also remain to be researched. For example,
why do some grooves run the length of the stone, some across the width,
and some diagonally, as was noted for abraders at various Point of Pines
4.6. Diagram of handstone, netherstone, and lapstone types expressed as sets and
subsets.
settlements (Adams 1994a:102)? At Hooper Ranch it was noted that the

grooves on smoothers were placed along the length of the abrader,
whereas the grooves on abraders were placed across the width. Grooves
running both lengthwise and widthwise on the same piece, creating a T-
shaped groove, were found on abraders at Casas Grandes (Di Peso et al.
1974:92) at Point of Pines settlements (Adams 1994a:102), and probably at
others as well. Chapter 2 explores some additional research issues related
to grooved abraders in the section “Answering Research Questions: Man-
ufacturing Activities.”
POLISHERS
The smooth texture of a polisher alters the surfaces of other objects through
abrasive and tribochemical mechanisms (Adams 1993b). As explained in
Chapter 2, the operation of these mechanisms causes a buildup of residues
that have a sheen often described as “polish.” The large set of polishers
includes handstones with smooth surface texture involved in the final stages
of manufacturing other items (Figure 4.7), and some lapstones and nether-
stones smooth enough to polish the items worked upon them. Relatively
large polishers are often strategically designed with finger grips or grooves
that make them easier to hold. Smaller, water-worn pebbles are suitable
for polishing smaller objects without further modification, making them
of expedient design.
Polishers have been associated with the manufacture of pottery and
wood or bone items, and larger ones with the application of plaster to
walls and floors (Adams 1979:51; Kidder 1932: 63–65; Woodbury 1954:93).
For expediently designed polishers particularly, use-wear analysis is the
only way of determining the nature of the surfaces they polished. Experi-
ments conducted with river pebbles indicate that polishing wood and
bone produces use-wear patterns distinctive from those produced by pol-
ishing stone and pottery (Adams 1979:49–51, 1994a:119).
Pottery Polishers
When a smooth pottery polisher is rubbed against a dry clay surface, a

small amount of clay is transferred to the stone’s surface and stays there
until it is worn away with further use. The smooth, hard texture of the
two contacting surfaces allows for the buildup of tribochemical residues
on both surfaces, and thereby produces a sheen. The sheen is enhanced
if the clay is moistened and burnished with the stone until the pot surface
is dry again. The combination of moisture and frictional heat alters the
surface
edge
thickness
end
length edge
width
4.7. Polisher, expediently designed. Use wear can occur on edges and surfaces.
(Drafted by Chip Colwell.) (Adams 1996: Figure 27.)
nature of the reaction products with the clay surface. When the clay is
moist, the polisher’s surface becomes coated, and some of the clay pushes
into the lowest recesses of the stone. The sharpest asperities become
rounded rather than crushed. If the clay is tempered, especially with an-
gular quartz sand, striations may cut through any sheen on the polisher’s
surface, leaving macroscopically visible striations.
Pottery polishers are almost exclusively of expedient design, with the
most effort expended in selecting appropriately textured material. River
pebbles are commonly chosen because their smoothness does not scratch
the pot surface (Adams 1979:49–51). More granular pebbles or rocks are
chosen for their ability to smooth out imperfections and are not used to
create sheen. Of course, the clearest evidence for pottery polishing is to
find the remnants of clay on the stone’s surface (see, for example, Geib
and Callahan 1988). Otherwise, use-wear analysis may be the only hope
for distinguishing polishers used on pottery from those used on stone,
wood or bone. Pottery polishing is perhaps easiest to recognize among
well-worn tools with facets. Some extensively used stones have facets of
varying sizes for finishing different areas of a pot. For example, a smaller
surface is needed to get under the lip of a pot than is needed to polish its
broad sides. The size of the rock is not as important an attribute as the
size of the working surface. Some pottery polishers may have been sec-
ondarily used to polish stone, bone, or wood surfaces; use-wear analysis
is the only way to determine such secondary uses.
Archaeological and Ethnographic Distributions
Ethnographic accounts of Puebloan (Batkin 1987:20–21; Bunzel 1972:63;

Dillingham and Elliot 1992:10; Mills 1995:151, 154–155; Simpson 1953:75;
Stanislawski 1978:217) and non-Puebloan (Spier 1933:107) pottery pro-
duction document that it traditionally has been women’s work. These ac-
counts serve as useful models for prehistoric production modes as well.
The teaching frameworks employed may have varied slightly, but in gen-
eral, women probably taught their relatives and perhaps their neighbors
how to make pots (Stanislawski 1978:219; Stevenson 1904:373). Their in-
structions included the selection of the proper raw material as well as the
proper use of tools for finishing.
Archaeologically, pottery polishers have been found throughout the
Southwest. Their generic design makes it unproductive to look for tem-
poral or spatial differences. One potential area of inquiry is the compari-
son of pottery polishers associated with paddle-and-anvil techniques of
production with those associated with coil-and-scrape techniques to de-

termine whether there is a discernible difference.
Stone Polishers
Stone polishers are distinguishable from all other types of polishers by their
use-wear patterns more than by any other attribute. If the polisher and
the polished stone surface are similarly textured and durable, the sheen is
distinctively more reflective than the sheen created by polishing pottery.
If the polished stone surface is more asperite than the polisher, striations
form on the polisher’s surface. If the polished surface is relatively soft,
such as a sedimentary or weathered volcanic rock, the use wear on the
stone polisher may be hard to distinguish from that on a polisher used on
dried, tempered pottery. The size and shape of a stone polisher does some-
what reflect the nature of its use. For example, the faceted polishers that
were probably used to finish axes are much larger (approximately 10 cm by
6 cm) than those used to polish cruciforms (approximately 3 cm by 7 cm).
There has not yet been enough research into stone polishers to allow
for discussions about their distribution through time or across space.
Floor Polishers
Floor polishers can be distinguished from other polishers by their predom-

inately disk shape and relatively larger size (Figure 4.8). They are also
more often strategically designed than other polishers with finger grips or
grooves (Adams 1979:51; Kidder 1932:64; Woodbury 1954:90). Expedi-
ently designed floor polishers are suitably shaped natural stones. Floor
polishers most frequently have unilinear striations indicating the direc-
tion of use in reciprocal strokes. Multidirectional striations indicate it was
used in a variety of strokes. Sheen is more likely to be produced with bur-
nishing partially dried plaster than with its wet application. A particular
stone could have been used in both plaster application and burnishing,
and perhaps evidence of only the last use is visible on the recovered tool.
Tools specifically identified as floor polishers by the Hopi are distin-
guished by a pecked area in the center of the flat polishing surface (Adams
1979:51). Similar stones have been documented as used in the application
and finishing of plaster for both walls and floors in Hopi and Zia struc-
tures, a process that sometimes incorporates beef or goat blood in the
final finishing (Adams 1979:52; Woodbury 1954:91). Floor polishers have
been identified at Pindi Pueblo, and reference is made to Stevenson’s
surface length
surface width
central pecked area
perimeter
thickness
length
width
4.8. Floor polisher with central pecked area distinctive of stones identified
ethnographically as floor polishers. (Drafted by Chip Colwell.) (Adams 1996:
Figure 28.)
description of putting adobe on the floors at Zia with such stones (Stubbs
and Stallings 1953:113). Not enough work has been done with these tools
to understand their distribution through time or across space. More ex-
tensive use-wear analysis and experimentation should help distinguish
those involved in the application and burnishing of plaster from those
used to polish other surfaces.
HIDE-PROCESSING STONES
Hide-processing stones are considered a subset of abrading and polishing

tools because they roughen or otherwise alter the surface of hides. They
are usually strategically designed with finger grips to facilitate holding.
Some stones used against hides may have been originally designed as
manos; these should be classified as manos and described as secondarily
used for processing hides. I did not recognize hide-processing stones un-
til they were described by a Hopi moccasin maker during the analysis of
stone tools recovered from Walpi (see Chapter 1, “Determining Function”).
Those recognized among the Walpi collection range in size from 8 to 16
cm long, 5 to 10 cm wide, and 4 to 8 cm thick (Adams 1979:53–54). Because
hide-processing stones were not a commonly recognized tool type, it is
important to define distinctive attributes with a model.
Source-side criteria for building a model of hide-working behavior
using an abrasive stone were developed through experimentation and
ethnographic descriptions. Experimentation led to the recognition of use-
wear patterns distinctive to hide-processing stones (Adams 1988:310–314).
A use-wear analysis may be the only way to distinguish hide-processing
stones from other handstones (as described in Chapter 2). Hide working
is easiest to recognize through a microscopic examination of the surface.
The grains and margins should be rounded, and wear should penetrate
into the interstices.
The ethnographic criteria come from descriptions of hide working by
the Hopi (Kewanwytewa and Bartlett 1946), Puebloan groups (Underhill
1944), Navajo (Kluckhohn et al. 1971), Sioux (Belitz 1979), Crow (Laubin
and Laubin 1957), Blackfeet (Ewers 1945, 1958), and Comanche (Wallace
and Hoebel 1952). Only the Hopi accounts mention the use of a stone other
than a flaked-stone hide scraper in either the defleshing or dehairing task
(Adams 1979:53; Kewanwytewa and Bartlett 1946). The Sioux, Crow, Ute,
and Blackfeet use a stone to add texture to the hides. Yuman tribes pounded
hides with handstones to soften them (Spier 1933:125). The Hopi and Ute
use a stone for other purposes such as coloring or whitening. Consequently,
a hide-processing stone is defined as a handstone used at one or more

hide-processing tasks, including the application of tanning solutions, the
raising of the nap, or application of color (Adams 1988:313). No studies
have been done on the distribution of hide-processing stones through
time or across space. Thus, there has not been adequate subject-side eval-
uation of the analogue.
Chapter 5
Grinding and Pulverizing Tools
G rinding and pulverizing are actions that engage the mechanisms of

fatigue wear and abrasion more than those of adhesion or tribochem-
ical reactions. Fatigue wear is visible as crushed grains and impact frac-
tures from the collapse of the highest and weakest points on the stone’s
surface. Striations are caused by the reciprocal or circular movement of
one surface against the other. Any intermediate substance between the
surfaces also becomes ground, pulverized, or crushed. Included within
this functional set of grinding and pulverizing tools are manos, metates,
mortars, pitted and cupped stones, pestles, handstones, netherstones, grind-
ing slabs, lapstones, and palettes.
Handstones and netherstones can be conceptualized at two different
levels: one generic and the other more specific (see Figure 4.6). The generic
set of all handheld stones is labeled “handstones.” Within that generic set
are the abraders, smoothers, polishers, and hide processing stones described
in Chapter 4, as well as the manos and pestles described in this chapter.
Handstones that do not have attributes to identify them as members of a
more specific subset are referred to by the generic set label handstones. Sim-
ilarly, netherstones is the label for the set of all bottom stones upon which
substances or items are altered. Tools that lack the attributes to include
them in more specific subsets such as metates or mortars are referred to
by the more generic label netherstones (Figure 4.6). This distinction for
handstones and netherstones may seem confusing until applied to a col-
lection of artifacts.
The following descriptions of tools that grind or pulverize include
manos and metates in a single category yet describe mortars and pestles
separately. The reason behind this is simple, but often overlooked by ar-
chaeologists. Neither a mano nor a metate is functional by itself; one of
each is required to form the composite stone tool used to accomplish the
task. Also, there is no evidence that manos or metates were sometimes made
98
5. G RINDING AND P ULVERIZING T OOLS 99
of a nonlithic material. In contrast, although mortars require the use of a

pestle, there is ethnographic and archaeological evidence that one or the
other tool may be wooden (Kraybill 1977:492). Stone against wood contact
removes unwanted hulls and does less damage to the grain than stone
against stone contact, an important factor when the desired end product
is a whole grain or something other than flour or meal (Kraybill 1977:
492). Furthermore, pestles may function without mortars when the ground
or another type of netherstone supports the substances to be crushed (see,
for example, Sayles and Sayles 1948). With these caveats in mind, the fol-
lowing descriptions of grinding and pulverizing tools define techniques
for recognizing composite tools.
MANOS AND METATES
Manos and metates are two components of food-processing equipment.

The metate is the netherstone, and the mano is the smaller, handheld com-
ponent. One cannot be used without the other to accomplish the task
(Adams 1999). Use and maintenance determine the final configuration of
both mano and metate surfaces (Adams 1999:481; Roberts 1940:118). Metates
and manos can be of either expedient or strategic design. Expediently de-
signed metates have a grinding surface and no other modification. Strate-
gically designed metates can have simple modifications to make the metate
rest firmly on the ground or can be carefully manufactured into a specific
shape. Hayden (1987), working with a Guatemalan metate maker, evalu-
ated material choices for metates and the tools used to make the very elab-
orate three-legged metates, which are clearly on the extreme end of the
strategically designed scale.
Manos function properly when they are compatible in size and con-
figuration with the metates against which they are used (described in more
detail below). Expediently designed manos have one or more grinding
surfaces and no other modification. The availability of appropriately sized
and textured rocks may make further modification unnecessary. Strategi-
cally designed manos are pecked or ground into specific shapes. Some
are designed with comfort features such as finger grips (roughened areas to
make a smooth stone easier to hold) or finger grooves (depressions that
make wide tools more comfortable to grip). Other manos have the non-
working surface configured to fit comfortably in the hand, a modification
seen on manos sometimes referred to as “turtle backs.” The decision to
make a mano with comfort features may be influenced by processing
strategies that require intense use in long sessions. (The concept of use in-
tensity is discussed further below.)
Definitions of Subtypes
Because manos and metates are used together, the use wear on the surface
of one tool reflects that on the surface of the other (Adams 1999:480–481;
Woodbury 1954:60). A technological classification of manos, therefore, refers
to metate design (see, for example, Hayes and Lancaster 1975:152–154)
and differs from classification schemes that type manos as one-hand or
two-hand (Plog 1974:140; Rinaldo 1957:42–49; Woodbury 1954:67), large
or small (Diehl 1996:109; Hard et al. 1996:260), or by assemblage-specific
categories of Type I, Type II, and so on (Di Peso et al. 1974; Haury 1976:
281–282; Lancaster 1984:247–248, Rinaldo 1959:229–236).
Metates are classified in this manual according to the configuration of
their grinding surfaces rather than by overall shape (slab, block, etc. [Haury
1950:305–306]) or assemblage-specific categories (Type 1, Type 2, etc. [Di
Peso et al. 1974:162–171; Haury 1976:280–281]). Confusion in metate clas-
sification is noted by Rinaldo (1959:240): “strictly speaking . . . there remains
the question whether the slab metates at Foote Canyon are not simply
new metates that have not been used long enough to develop a trough.”
Similarly, Huckell (1998:120) notes a confusion created by some classifica-
tions that consider one-hand manos handstones because it was assumed
that they were not used to grind maize.
Rinaldo’s question, and the confusion noted by Huckell, can be ad-
dressed through analysis of how manos and metates are designed, and
how they wear concomitantly, keeping in mind that morphology does not
indicate what specific food substances are processed (Adams 1999; Wright
1994). A technological approach classifies artifacts according to design,
using amount of wear as one indicator of life history, thus eliminating Ri-
naldo’s confusion. To clear up some of the confusions, the following type
descriptions are suggested as the most useful for understanding the rela-
tion between tool design, kinetics, and wear.
Basin Design
Metates are classified as basin if their grinding surfaces are circular or el-
liptical basins (Figure 5.1). If the basin is open on one end it is a 3/4-basin
metate; if it is open on both ends it is an open-basin metate. These should
not be confused with 3/4- or open-trough metates, which are defined be-
low. Basin manos are manipulated with some combination of circular and
reciprocal strokes. These strokes may obliterate all evidence of basin man-
ufacture, making it difficult to distinguish some shallow basin metates
from some well-used flat/concave metates, as defined below (Adams
1993a:336–338, 1994a:75, 1999:481; Dick 1965a:110; Euler and Dobyns 1983:
254; Haury 1950:315–316; Woodbury 1954:50–51).
basin length
compatible mano
in basin
basin width margin
thickness
basin depth
base
length
width
5.1. Basin metate. Note that the upper portion of the basin is wider from grind-
ing with a broad, circular stroke, and the bottom portion of the basin is narrower
from grinding with a reciprocal stroke. (Drafted by Chip Colwell.) (Adams 1996:
Figure 17, 1999: Figure 2.)
5.2. Schematic of a circular stroke used in a basin metate: (a) assuming a right-
handed grinder, and a counter-clockwise movement, the away stroke puts
pressure on the area under the thumb; (b) the stroke across the top shifts the pres-
sure to the fingers; (c) the returning stroke shifts pressure to the right side of the
hand and around again to under the thumb. Note the location of the wear facets
compared to those illustrated for the reciprocal stroke in Figure 5.3. (Rendered
by Doug Gann.) (Adams 1993: Figure 3, 1996: Figure 4, 1999: Figure 3.)
The type of stroke used to manipulate the mano in a basin metate de-
termines the type of wear visible on the surfaces of both tools (Adams
1993a:337–338). Basin manos manipulated with circular rocking strokes
have wear facets on parts of their ends and edges, and multidirectional
striations on their surfaces (Figure 5.2). This stroke leaves circular or mul-
tidirectional striations within the metate basin as well. Basin manos used
in reciprocal rocking strokes have wear facets only on their edges, and lin-
ear striations perpendicular to their edges (Figure 5.3). Use wear on the
metate is visible as linear striations parallel to the length of the basin. Ex-
tensive use of this stroke can create a narrow channel in the bottom of the
5.3. Schematic of a reciprocal stroke used in a basin metate: (a) pressure is

greatest under the grinder’s palm on the away stroke; (b) less pressure is applied
under the fingers on the return stroke. Note the location of wear facets compared
to the circular stroke illustrated in Figure 5.2. (Rendered by Doug Gann.) (Adams
1993: Figure 4, 1996: Figure 5, 1999: Figure 4.)
basin. If both strokes are used, wear facets will be visible on all ends and
edges of the mano, and surface striations will be either or both linear and
multidirectional depending on the most recently used stroke. The result-
ant grinding surfaces are convex, with some basin manos resembling balls
(Figure 5.4).
Flat/Concave Design
Metates classified as flat/concave are used with manos shorter than the
metate width, and even though they start with flat surfaces, extensive use
can wear depressions deep enough to confine meal in the same way as
basin metates (Figure 5.5) (Euler and Dobyns 1983:264). A basin metate is
distinguishable from a flat/concave metate by the intentional shaping of
the basin and the narrowness and depth of the basin resulting from the
use of a smaller, rounder basin mano. The name flat/concave reflects the
progressive wear of the surfaces and distinguishes them from flat metates
that have flat surfaces because their manos are as long as the metates are
wide. Flat/concave metates also have been called “slab” and “block”
metates, but the same terms are used to refer to flat metates. Thus, the
term flat/concave is proposed to avoid confusion.
Flat/concave manos are generally longer (Table 5.1), with flatter grind-
ing surfaces than those of basin manos, and can be worked against metate
surfaces with several different strokes (Figure 5.6). Flat reciprocal strokes
keep a mano’s surface in contact with the metate surface at all times.
These strokes maintain relatively flat, edge-to-edge contours even if the
facet on proximal edge basin wear
dorsal side
thickness
grinding surface
ventral side
length
width
5.4. Basin mano. Note the wear facet on the proximal and distal edges from
use with a reciprocal rocking stroke. (Drafted by Chip Colwell.) (Adams 1996:
Figure 20.)
surface length
surface width
thickness surface depth
proximal end base
length
distal end compatible mano

width
5.5. Flat/concave metate. Note that the flat/concave mano is not as long as the
metate is wide. The wear on the metate does not extend completely across the
surface. (Drafted by Chip Colwell.) (Adams 1996: Figure 19, 1999: Figure 5.)
Table 5.1
Statistical description of mano types (in centimeters).
Interquartile Standard
Mano Type Smallest Largest Mean Median Range Deviation Number
Basin
length 03.9 018.2 011.4 011.20 010.0–13.20 02.36
area 36.7 249.9 113.2 106.21 089.1–138.6 39.00 102
Flat/concave
length 07.9 026.0 014.0 013.40 11.1–16.1 03.68
area 53.7 356.3 145.0 137.00 102.0–177.7 54.00 302
Trough
length 07.0 030.3 017.7 017.70 16.0–19.4 02.99
area 66.5 360.0 185.6 182.70 161.2–211.7 41.70 458
Flat
length 04.4 039.5 023.1 023.00 19.0–28.7 06.67
area 29.0 424.3 231.2 230.30 182.7–278.7 76.30 104
Total
length 03.9 039.5 016.5 016.30 12.8–19.2 04.86
area 29.0 424.3 170.1 170.00 125.8–205.0 59.75 966
Source: Adams 1999: Table 3.
deepening depression in a flat/concave metate surface creates a slightly

convex end-to-end contour on the mano surface. If the mano is rocked
during the away (or downward) stroke, a facet becomes worn on the prox-
imal edge; if it is rocked during the return stroke, a smaller facet becomes
worn on the proximal edge. Proximal facets are smaller because less pres-
sure is exerted on return strokes than on the away strokes (Figure 5.7).
Linear and multidirectional striations, and wear facets on mano ends and
edges develop on the flat/concave tools through the same mechanisms as
they do on basin tools. Hayes et al. (1981:119, Figure 148) illustrate one
stroke possible with a flat/concave mano and metate; they refer to the
manos as one-hand manos. See also Haury (1950:315) and Morris (1990:
181–182) for alternative views on stroke patterns.
Trough Design
Trough metates have intentionally manufactured rectangular basins or

troughs (Figure 5.8). Manos worked in trough metates can be moved only
with reciprocal strokes. Parallel abrasive striations formed on the mano
ends and the grinding surface match those on the edges and surface of the
trough metate (Adams 1993a:336–340). Because manos and metates wear
concomitantly, the manos progressively become shorter, and the metate
grinding surface
thickness
length
end
width
edge
5.6. Flat/concave mano. The profile is flat edge-to-edge and convex end-to-end
from use against a slightly concave metate surface. (Drafted by Chip Colwell.)
(Adams 1996: Figure 22.)
troughs narrower. Such troughs do not appear intentionally manufactured

because evidence of the initial shaping is usually destroyed by subsequent
use. If the trough is deep and has a rectangular cross section, it probably
was rewidened and used with a second, larger mano. If the trough width
is not maintained, the walls become sloped, making it easily confused
Downward Stroke
distal 2 adjacent
edge surfaces
a b c d e
Upward Stroke
f g h i j
5.7. Schematic drawing of mano surface configurations: (a–e) pressure on the
proximal edge during the downward part of the stroke creates a facet toward the
proximal edge; (f–h) lighter pressure on the distal edge during the upward part of
the stroke brings the meal back to the top of the metate and creates a small facet
toward the distal edge. As wear progresses, the facets enlarge, and if the proxi-
mal and distal edges are reversed (i–j), the wear is equalized and a second
adjacent surface is formed. (Redrawn by Ron Beckwith, based on Bartlett 1933:
Figure 8.) (Adams 1993: Figure 1; 1996: Figure 3.)
with certain basin metates. Remnants of an earlier, narrower trough can

be felt along the trough side as evidence of trough rewidening (Figure
5.8). Some deep troughs may have as many as three or four remnants of
narrower troughs. Sometimes the ridges created by trough rewidening
can be located more easily by touch than by sight.
The evidence for widening of troughs should not be confused with
that for reusing a trough metate with a mano shorter than the width of
the trough, as described by Schelberg (1997:1048, Figures 9.14, 9.15) for
metates recovered from various Chaco settlements. The shorter manos
wear a smaller trough within the original trough, leaving more prominent
ridges along the trough wall than the ridges created by trough rewidening.
Distinguishing an unmaintained trough metate from 3/4- or open-basin
metates is possible through use-wear analysis. A basin metate should have
use wear resulting from the circular and reciprocal strokes of a mano shorter
than the width of the basin, as described above; measurements taken at
the top of the basin should be wider in the middle than toward the ends.
An unmaintained trough metate should have use wear resulting from the
reciprocal stroke of a mano that is as wide as the trough; measurements
5.8. Open-trough metate. Note the remnant ridge from trough rewidening.
(Drafted by Chip Colwell.) (Adams 1996: Figure 18, 1999: Figure 6.)
taken at the top of the trough should be uniform along the trough lengths.
A mano used in a maintained trough has distinct corners between the
grinding surface and the ends that rub against the trough edges. Manos
used in unmaintained troughs have very curved surfaces with no distinc-
tion between the grinding surface and the ends. These have been called
“rocker manos.”
Trough manos and metates can be distinguished from flat/concave
manos and metates by the intentional shaping of the troughs and the
exclusively reciprocal mano strokes. Even though similar flat and rocking
reciprocal strokes are used with both trough and flat/concave manos,
trough manos have distinctive wear on their ends from rubbing against
trough borders (Figure 5.9). Flat/concave manos also can have multidi-
rectional striations from circular strokes.
Trough metates have several designed configurations: open, 3/4, and
Utah are the most commonly recognized. Closed troughs, less common than
other subtypes, have borders on both sides and both ends. They are dis-
tinguished from basin metates by strictly reciprocal mano strokes and
more-rectangular depressions. They are distinguished from shaped mor-
tars by size and distinctive use-wear patterns. The larger closed-trough
metates are worn by abrasion, whereas the smaller mortars are worn by
both impact fractures and abrasion. A fuzzy boundary exists between
larger shaped mortars and smaller closed-trough metates.
Open-trough metates have borders only along the sides, so that both
ends are open (Figures 5.8, 5.10a). Other terms used to refer to these metates
are “through-trough” and “trough open at both ends.” Trough bottoms
are flat if the mano was worked with a flat stroke, exerting a constant
amount of pressure for the length of the metate. Such a stroke is possible
only if the metate remained flat on the ground during use. Additional ev-
idence for a metate used in a flat position can be found on the metate bot-
tom if it was pecked and ground to create a stable base. If the metate was
propped, the downward stroke would have exerted more pressure toward
the longitudinal center, eventually creating a concave surface. The trough
narrows as the mano wears smaller, and eventually the trough sides be-
come sloped. As mentioned above, a deep trough with a rectangular pro-
file shows evidence of wear maintenance.
Three-quarter trough metates have borders on both sides and one end
(Figure 5.10b). These have also been referred to as “scoop metates” and
“metates with trough closed on one end.” Troughs wear in the same man-
ner described for open-trough metates; their configuration thus depends
on whether they were used flat on the ground or propped at an angle,
and on the amount of trough rewidening.
Utah-trough metates are also referred to as “3/4-trough metates with
shelves.” Some mano rests or shelves have deep depressions, while others
are only slightly depressed. Roberts (1931:154) notes that at Kiatuthlanna,
the 3/4-trough metates have shelves with depressions not as deep as those
from Utah. To further confuse some identifications, it was recognized that
at settlements along Tonto Creek (Figure 5.10c) and north of Flagstaff there
are several metates that looked like Utah-trough metates but are actually
3/4-trough metates redesigned into open troughs (Adams 1999, 2000c,
end
distal edge
proximal edge
trough wear
dorsal side
thickness
ventral side
width
length
5.9. Trough mano. Note the wear on the ends where they rub against the metate
trough. (Drafted by Chip Colwell.) (Adams 1996: Figure 21.)
5.10. Metates with compatible manos: (a) open-trough design; (b) 3/4-trough design; (c) 3/4-
trough redesigned into an open trough. All three metates and manos were recovered from the
same feature at a site along the upper Tonto Creek. (Adams 2000c: Figure 10.10.)
2001a). The proximal borders were battered away, lowering the surfaces,
but not as low as the original 3/4-troughs. Use-wear damage that extends
from the original troughs over the higher ends indicates that they were
worked with a mano after the border was lowered. It is important to rec-
ognize these design and redesign differences for those studies that try to
equate tool design with particular time and space distributions or with
specific groups.
Flat Design
Flat metates start out with flat or unshaped surfaces, and they remain flat
edge-to-edge because their manos are the same lengths as the widths of
the metate surfaces. Flat metates in bins are fixed in place at an angle (Fig-
ure 5.11). The pressures of the downward strokes eventually wear longi-
tudinal concavities, as described for propped trough metates. Flat metates
that rest firmly on the ground wear evenly across their surfaces. Flat manos
remain flat end-to-end and are distinguished from trough manos by the
lack of wear on their ends. Facets become worn on the proximal and dis-
tal edges of flat manos in the same manner as described for flat/concave
manos. Flat manos, more often than other mano types, have multiple grind-
ing surfaces, described further below.
Mano Profiles as Correlates for Strokes and Wear-Management Strategies
Manos are sometimes described by their profile shape: rectangular, wedge,

triangular, diamond, and so on (Figure 5.12). A mano’s profile reflects its
5.11. Open-trough and flat metates in multiple bins from Room 13 at W:10:51
in the Point of Pines area (Adams 1994: Figure 5.6.) Note the baffle sherds in the
north corners of at least two bins. These metates were not found in place in the
bins but were found scattered in the room. (Photograph courtesy of the Arizona
State Museum, University of Arizona, E. B. Sayles, photographer.)
5.12. Schematic of mano profiles illustrating the locations of mano surfaces: (a) a single sur-
face; those with convex upper surfaces are sometimes called “turtlebacks”; (b) two opposite
surfaces worn to a wedge profile; (c) two opposite surfaces worn to a rectangular profile;
(d) two adjacent surfaces; (e) two adjacent surfaces and one opposite surface worn to a trian-
gular profile; (f) four adjacent surfaces worn to a diamond profile. (Redrawn by Rob Ciaccio
based on illustrations from many references, such as Stubbs and Stallings 1953: Figure 66;
and Woodbury 1954: Figure 8.)
life history—how it was used and maintained—and should not be consid-

ered a mano type. This is an issue with which Kidder (1932:70–71) strug-
gled. He lamented the fact that he had not collected the appropriate infor-
mation from the manos left in the field at Pecos Pueblo to address the
relationship between wear and mano profile.
Bartlett’s (1933:15, Figure 8) discussion of two types of grinding strokes

(rocking and flat) illustrates how certain mano profiles are created. With a
rocking stroke the grinder exerts pressure on the proximal mano edge and
pushes downward while holding the distal edge of the mano slightly above
the metate surface. On the upward stroke the grinder rocks the mano for-
ward so that the distal edge is held against the metate and the proximal
edge is lifted up, carrying grain back to the top of the metate (see also
Spier 1933:127) (Figure 5.7). A wear facet is created early in the life history
of a mano worked with this stoke. The facet size increases with this stroke
until it takes up almost half of the mano surface, and the proximal edge
eventually becomes thinner than the distal edge. To counteract uneven
wearing, the grinder can rotate the mano so that the distal edge becomes
the proximal edge. Eventually two adjacent surfaces are created, and the
profile appears triangular. This shape provides a raised edge for the grinder
to grip, keeping the fingers from coming into contact with the metate sur-
face as the mano becomes thinner. Some extremely well used manos were
also flipped so that the opposite surface was worn and rotated in the same
way, creating a diamond profile with four adjacent surfaces.
A flat stroke maintains the mano surface in constant contact with the
metate. Pressure on the downward stroke wears the proximal edge more
rapidly than the distal edge, creating a wedge-shaped profile. By rotating
the mano so that the distal edge becomes proximal, the wear is balanced,
and a single flat surface is maintained, resulting in a rectangular profile.
The number of surfaces used against the metate can be increased to two if
the dorsal and ventral surfaces are rotated. This rotation does not make
the mano easier to hold as it wears thinner, but it does provide a usable
second surface when the first one has worn too smooth (Adams 1993a:
335–336). The opposite surfaces must be regularly rotated to maintain the
compatibility between both surfaces and the metate. If used enough, any
mano becomes too thin to hold comfortably and is essentially worn out.
Bartlett (1933: 4) notes that a Hopi grinder might resharpen her mano
every five days. This resharpening hastens wear and affects how fast manos
wear out. A grinder might want to manage wear for several reasons. It
might be more desirable to prolong mano use life rather than expend the
energy to procure new material, manufacture a new tool, and break in
that tool until it is compatible with an existing metate. Furthermore, a
particular mano might be more comfortable to use, or raw materials might
be scarce. Bartlett’s (1933) descriptions of grinding strokes are important
because they recognize that the grinder is concerned about wear and has
certain wear-management techniques for maintaining efficiency, and oth-
ers that prolong mano use life.
Grinding Stations
Metate positioning and the creation of grinding stations are part of grind-
ing technology development. Grinding stations can be classified as single
or multiple, and as movable or permanent. These attributes are important
because of how they relate to human behavior. Single grinding stations
are usable by only one person at a time. Multiple stations allow more than
one grinder to work simultaneously, thereby allowing for different types
of social interactions than does solitary grinding. Movable freestanding
metates can be used in a space intended for multiple functions. Perma-
nent features with constructed confinement features indicate a dedication
of space specifically to food-grinding activities. Perhaps the earliest exam-
ple of a permanent grinding station was found at Stone Pipe (a.d. 1–550)
in the Tucson Basin, where a basin metate was found plastered at an angle
within a shallow pit (Adams 1998:374). Woodbury (1954:62) notes an iso-
lated example of a permanently affixed metate on the floor of an early pit-
house in the La Plata District.
Evidence for the propping of metates has been noted in many descrip-
tions of floor assemblages (see, for example, Roberts 1931:154, Plate 5a).
Freestanding metates are positioned either flat on the ground or propped
at an angle. Basin and flat/concave metates are positioned flat on the floor
more often than are trough and flat metates. If no trivets were needed to
stabilize the metate and it was removed, no evidence of a grinding station
remains. The presence of metate trivets on an archaeological floor should
not be used to infer that metates were propped at an angle: the bases of
some metates are so irregular that rocks are needed to position their sur-
faces to keep the metates from rocking during use. Few reports actually
describe the stones used to prop metates. For an exception, see the de-
scription of “metate rests” recovered from Duckfoot, a late Pueblo I settle-
ment in southwest Colorado (Etzkorn 1993:163–164). Also see the defini-
tion of trivets in Chapter 12, “Structural Stones.”
Freestanding metates were commonly used by historic groups in Ari-
zona. Spier’s 1928 report on the Havasupai describes a woman grinding
food against what is probably a flat/concave metate resting flat on the
ground. She moved the small mano in multiple directions across a flat
metate. A historic photograph of a Maricopa woman shows her kneeling
behind a flat metate propped at a low angle, with a cloth spread to catch
the ground meal (Spier 1933:127, Plate III). The mano is as long as the metate
is wide, and the accompanying description indicates that the mano was
worked with a rocking stroke. The woman was outside in a setup that
was probably too ephemeral to become visible in the archaeological record.
It should be noted that the photograph may have been staged and may
not represent an actual grinding location.
Permanent grinding stations have two obvious types of confinement
features: bins and receptacles. Receptacles are pits excavated below the floor
that serve to confine the ground meal or flour. The metate rests either
against the edge of the pit or on the floor above it (Figure 5.13). Slabs
sometimes line the receptacle, but they do not create a floor-level box like
a bin. A bowl in the bottom of the receptacle facilitates removal of flour.
Floor-level adobe ridges confine flour and separate metates in grinding
stations with more than one receptacle.
Bins are slab-lined boxes situated so that the slabs extend both above
and below the floor (Figure 5.14). The metate (usually of flat design) is set
at an angle ranging from 20 to 35 degrees (Bartlett 1933:5; Judd 1954:133;
Wendorf 1950:29). The top of the metate is above the floor, and the bottom
is either on the floor or slightly below. (See, for example, the idealized
cross sections of bins illustrated by Bartlett [1933: Figure 5] and Brew [1974:
147, Figure 24], and the photographs of bins at Bailey [Mills et al. 1999:
200–201, Figure 6.30]). It is not unusual to find a stone platform inside the
bin where a bowl or basket can be placed to catch the meal or flour as it
comes off the metate. Sometimes bowls are sunk below the floor surface
at the end of the metate.
Bins occur as individual or multiple units, with five perhaps a maxi-
mum number. The bins in the set of four illustrated in Figure 5.11 have re-
movable baffle sherds positioned toward the distal end of the metate, pro-
viding easy access to the flour. Bins became a part of Anasazi grinding
technology sometime between a.d. 900 and 1000. See, for example, the
Pueblo II bins uncovered at Site 12 on Alkali Ridge (Brew 1974:147, Fig-
ures 24, 174).
The presence of multiple grinding stations has been noted both ar-
chaeologically (Figure 5.11) and ethnographically. The ethnographic de-
scriptions of grinding activities that involve several women working mul-
tiple bins always are in reference to maize grinding. When multiple grinders
work together, one grinder cracks and grinds the maize to a coarse texture
which is passed subsequently to other grinders who each reduce the meal
to a finer texture, ultimately producing the finest flour (Bartlett 1933:4;
Hough 1915:62; Mindeleff 1989:211; Parsons 1939:21; Stephen 1936:882).
Sometimes the metates have been described as being of varying tex-
tures, supposedly related to the coarseness of the meal or flour ground on
them. However, experiments have illustrated that producing fine-texture
flour is not as much a factor of stone texture as it is of time and the nature
of the processed seeds (Table 5.1). Thus, a single grinder working with
5.13. Adobe grinding receptacle from Turkey Creek Pueblo in the Point of Pines
area. An open-trough metate is plastered at an angle with a slab immediately
behind the proximal end. The metate rests at floor level, and the receptacle is
slightly below the floor. (Photograph courtesy of the Arizona State Museum,
University of Arizona, E.W. Haury, photographer.)
only one metate of any texture can produce flour or meal of any texture. If
the processing involves several grinders, each working the same batch of
flour on different metates, this is a social process, not a functional process.
In such social contexts the concepts of efficiency and use intensity probably
had more significance than they would in the context of a single grinder.
Concepts of Efficiency and Use Intensity
In order to fully understand how mano/metate morphology relates to the

behavior of prehistoric grinders, the concept of use intensity must be rec-
ognized as distinct from the concept of tool efficiency. Efficiency is meas-
ured by output of ground product per unit of time. Improved efficiency
has two behavioral implications and several results. The first implication
is that less time is spent grinding a particular quantity of grain. The pos-
sible results are: (1) the same number of people are fed the same amount,
(2) fewer people are fed more, or (3) surplus is created if fewer people are
5.14. Slab-lined mealing bins from W:10:51 in the Point of Pines area. Note
that both flat and open-trough metates are plastered at an angle within the bins.
These metates were found in place. The flat stones at the distal ends of the
metates are sunk below floor level. (Photograph courtesy of the Arizona State
Museum, University of Arizona, E. B. Sayles, photographer.)
fed the same amount. Whatever the result, the user of a more efficient
tool has more time available to participate in other activities. The second
implication is that the same amount of time is spent grinding a larger
quantity of grain. The possible results are: (1) more people are fed, (2)
more processed grain is added to the diet of the same number of people,
or (3) surplus is created if the same number of people are fed the same
amount. Any or all of these results may be the goal of an individual grinder,
and different goals may be desirable at different points in time.
Use intensity is measured by the amount of time spent at each grind-
ing task (Adams 1993a). For example, a mano used for three hours one
day is used intensively compared to a mano used for an hour a day over
three days, which is used extensively. Both have received three hours of
use, and it may not be possible to distinguish intensive and extensive use
in every context. Comfort features and wear-management strategies (ex-
plained below) may be more indicative of intensive than extensive use.
Increased use intensity does not require larger tools; however, tool design
may be governed by the limitations of human strength and endurance,
and the material’s resistance to wear (Horsfall 1987). Knowing these limi-
tations, grinders can select relatively wear resistant material, create fea-
tures that make the tools comfortable to use, and develop techniques to
manage wear (Adams 1993a:334–336). Increased use intensity has one be-
havioral implication and three results. The behavioral implication is that
more time is spent at a single grinding task. The possible results are: (1)
more people are fed, (2) the amount of processed grain is increased in the
diet of the same number of people, or (3) surplus is created if the same
number of people are fed the same amount. The important behavioral
factor is that an individual grinder spends many consecutive hours grind-
ing with the same mano and metate.
The correlates for grinding efficiency and intensity are subtle, but meas-
urable. All manos are efficient in that, in concert with the metate, they re-
duce foods to smaller particle size; thus, efficiency is a relative concept.
Some mano/metate sets grind more efficiently than others because they
are easier to maintain (Adams 1999:487) or because of their greater size, as
demonstrated by Mauldin (1993) with Bolivian food grinders. Size attrib-
utes can be measured and compared within and between assemblages. The
attributes useful for measuring intensity are the number of used surfaces,
amount of wear, evidence of wear management, and the presence of man-
ufactured comfort features (Adams 1993a). In a relative sense, extensively
used manos may be sorted out as those with moderate to heavy amounts
of wear and no visible comfort features or evidence of wear management.
Wear Rates
Experiments have illustrated that manos wear out faster than metates,
perhaps lasting as little as one or two years (Wright 1993:352, Table 2) de-
pending on material, amount of resharpening, and grinding intensity. Is-
sues concerning metate use life and material preference were addressed
during the study of Guatemalan Mayan food-grinding tools (Hayden
1987:13–17). The preferred basalt was more resistant to wear, lasting 30–
100 years, but it also was more expensive to obtain. The cheaper granitic
material lasted about 15 years. Economic and cultural reasons were cited
for why metates were moved from one village to another. Archaeological
floor assemblages sometimes have multiple manos that fit a single metate,
a single mano that fits more than one metate, or multiple manos that do
not fit any metates. It is common for site assemblages to have many more
manos than metates. Some manos may have been cached for future use,
or metates may have been removed from settlements upon abandonment
or later scavenged (see, for example, Adams 1998:364–374; Schlanger 1991).
Metates are one of the few tools with which it is possible to determine
the intentional ending of their designed use. Examples of “killed” metates
have been found in many places. (See, for example, those recovered from
Chaco Canyon settlements [Schelberg 1997:1068, Figure 9.24]). Haury
(1985b:244) thought that killing metates was more common in the south-
ern part of the U.S. Southwest than in the northern part. Evidence for
killing is usually in the form of a hole punched through the bottom. Holes
in the bottoms of metates have sometimes been referred to as having been
worn through the stone, but the chances of that are very unlikely. The
holes are usually in bottoms that are still a centimeter or two thick and
surrounded by flakes and impact fractures.
Development of Food-Grinding Technology
Manos and metates are perhaps the most ubiquitous ground stone items
in the U.S. Southwest. Traditionally they have been analyzed from an evo-
lutionary perspective. In this view, certain items evolved from earlier pro-
totypes, with particular archaeological examples considered transitional
between types. This thinking is clearly illustrated with Hohokam assem-
blages from Snaketown and Hodges Ruin. A line drawing charts their evo-
lution as it was understood after the 1930s excavation at Snaketown (Glad-
win et al. 1975: Figures 46, 47; Kelly 1978: Figure 7.2). Such illustrations
and their attendant trait lists became tools used by researchers to place
distinctive types into time categories.
The earliest hypothesis was that metate morphology changed with
the introduction of agriculture. Changes in tool forms through time were
interpreted as developments to improve grinding efficiency related to the
requirements of processing planted foods for consumption. Increases in
the frequency of more efficient tools were used to model an increasing re-
liance on planted foods (Haury 1950:317, 545, 1976:282; Plog 1974:139–
140). When defining the Cochise culture (8000 to 3000 b.p.) in southeast Ari-
zona, Sayles (1983:114; Sayles and Antevs 1941:27–30) described an evolu-
tionary scheme for manos and metates. Simply put, shallow basin metates
evolved from flat slab milling stones as an improved tool for grinding,
mashing, or pulverizing wild foods. The basin configuration predomi-
nated until the introduction of maize and trough metates (Haury 1950:
543–548, 1976:351–352). Martin and Rinaldo (1947:290, 316, 1950b:561–562;
Martin 1943:131; Rinaldo 1940:35, 1943:177, 1952:111), studying early Mogol-
lon settlements in western New Mexico, clearly stated the assumption that
basin metates and one-hand manos were equated with seed-gathering
economies, and trough metates and two-hand manos were introduced
with maize agriculture. Haury (1950:317, 545, 1976:282) voiced the same
idea based on his analysis of ground stone from Ventana Cave in Arizona.
We now know that maize agriculture began centuries before trough mano
and metate designs were used in the U.S. Southwest, and that initially
the processing tools used to incorporate agriculturally derived products
into the diet needed no design changes (Adams 1999; Wills 1988).
I have proposed elsewhere (Adams 1999) that design developments
were unrelated to how foods were acquired but were instead sensitive to
changes in recipes and the ways foods were processed. The trough design
was a solution to the problem of efficiently grinding dried seeds into flour.
The importance of flour may have increased for practical reasons. For ex-
ample, dried seeds store longer than fresh or soaked, and flour-based tor-
tillas, tamales, and breads cook faster, use less fuel, and are more versatile
than whole-grain stews. Alternatively, or concomitantly, there may have
been social reasons to encourage the design of more efficient tools for dry
grinding. Perhaps fewer grinders were available, or there were more
mouths to be fed by the labor of a single grinder.
There are, however, basic clinal distributions in metate designs that re-
flect the development of different technological traditions. Figure 5.15 il-
lustrates these trends on a relative scale. Among the assemblages from
the Middle Archaic and late preceramic periods (roughly 1200–400 b.c.),
the earliest food-grinding tools across the Southwest are of basin and flat/
concave designs (see, for example, those from White Dog Cave, Talus Vil-
lage, Tularosa Cave, Cordova Cave, Bat Cave, Ventana Cave, White Water
Draw, Double Adobe, and Santa Cruz Bend) (Adams 1999: Table 4; Huck-
ell 1995:61–63). At some point in the sixth or early seventh century a.d.,
trough metates were added to their choices of tool designs.
If we could take a survey of the U.S. Southwest between about a.d.
300 and 500, we would find evidence for the introduction of both 3/4-
and open-trough metate designs. For example, at the New Mexico settle-
ments of Galaz and SU, 3/4-trough metates are predominate in contexts
that date slightly earlier than a.d. 500, with one or two open-trough metates
in the same early contexts (Lancaster 1984: Figure 17.4; Martin and Rinaldo
1947:328, Figure 108). This may be as far north as the open-trough design
extended during this early time period. Only 3/4-trough metates have
been found in contexts that date between a.d. 300 and 600 at Flattop in
the Petrified Forest (Wendorf 1953a:60). Martin et al. (1949: Figure 14) il-
lustrate a metate from a Cochise deposit at Wet Leggett that they describe
as “open basin or trough type, trough open one end only.” However, a
use-wear analysis conducted on a metate from Valencia Vieja in the Tuc-
son Basin that had a similar surface configuration concluded that metates
5.15. Relative time line illustrating the occurrence of various metate designs in different
areas around the U.S. Southwest. (Rendered by Catherine Gilman.)
of this configuration have a 3/4-basin design, not a trough design (Adams

2000a). This illustrates some of the confusion created by a lack of under-
standing about how metates become worn through use with particular
manos and with particular motions.
By a.d. 500 the evidence for trough metates is more widespread (Fig-
ure 5.15). Three-quarter trough metates, including Utah-trough metates,
are the predominate type among Basketmaker III/Pueblo I (approximately
a.d. 500–900) and equivalent contexts in southern Utah, southwest Col-
orado, central and northern Arizona, and western New Mexico (compare,
for example, Alkali Ridge, Prayer Rock District, La Plata District, Shabik’
eshchee, Badger House Community, Duckfoot, Jeddito 264, Bear Ruin,
Bluff, and Crooked Ridge Village). Both open- and 3/4-trough designs
continued to coexist in some areas: the Pinelawn area in western New
Mexico (Martin and Rinaldo 1947:328, Figure 108; 1950a: 318, Figure 118;
Martin et al. 1949:152, Figure 54), the La Plata District (Morris 1939:133,
Plate 147) in northwest New Mexico/southwest Colorado, Three Circle–
phase deposits at Turkey Foot Ridge and Twin Bridges (Martin and Ri-
naldo 1950a) and farther south at Wind Mountain, (Woosley and McIn-
tyre 1996:213). However, 3/4-trough designs are more frequent in these
areas than open-trough designs.
It is interesting that no open-trough metates were reported from any
contexts at Casas Grandes (Paquimé) (Di Peso et al. 1974:45). This con-
trasts with the Tucson and Phoenix basins, where during the Tortolita and
Vahki phases, respectively (approximately a.d. 450–650), the earliest trough
designs are open (see, for example, Snaketown [Haury 1976] and Valencia
Vieja [Adams 2000a]). The few reports of 3/4-trough metates from early
contexts in the central basins of Arizona should be reevaluated. For ex-
ample, researchers looking at Pioneer-period assemblages have identified
3/4-trough metates used with handstones or one-hand manos, and these
have been considered transitional between basin and trough metates (Rapp
1995:140, 1998:301; Montero 1992:273). However, as mentioned previously
in reference to similar metates recovered from earlier deposits at Wet
Leggett, a use-wear analysis on an example of this type of metate sug-
gests that they are more likely to have been 3/4-basin metates than 3/4-
trough metates (Adams 2000a).
Around a.d. 1000 or perhaps a little earlier, open-trough designs be-
gin to appear in locations where they were not found previously, or they
become more abundant in places where they were previously scarce: the
Point of Pines area (Adams 1994a:90), the upper Little Colorado River
drainage (Martin et al. 1964:67), the Zuni area (Roberts 1932:140, Plate 49),
the Colorado Plateau (Woodbury 1954:53), Chaco (Judd 1954:135; Schelberg
1997:1074), the Flagstaff area (Adams 2001a), and Mesa Verde (Hayes and
Lancaster 1975:151, 185), to name a few. Martin (1979:71) places the ap-
pearance of open-trough metates in Mogollon contexts at about a.d. 800,
although they probably did not become more common than 3/4-trough
metates until later, perhaps post–a.d. 1000 (Rinaldo 1956:128). At most lo-
cations the two trough designs coexist, perhaps for generations.
That the open-trough design was sometimes preferred over the 3/4-
trough design is evident with the occasional redesigned trough, such as at
Higgins Flat Pueblo (Rinaldo 1956:73), Pueblo Bonito (Judd 1954: Plate
30), and in the Flagstaff area (Adams 2001a). In each of these places, 3/4-
trough is the indigenous design. In the Tonto Basin, where open-trough is
the indigenous design, 3/4-trough metates begin to show up in early Clas-
sic contexts. A few of these were redesigned into open troughs (Adams
2000c). At some point open-trough metates began to outnumber 3/4-
trough metates, especially in east-central Arizona and west-central New
Mexico. This point is during the Reserve phase in the Pinelawn Valley
(Martin and Rinaldo 1950a:450, 1950b:464, Figure 465), and the Reserve/
Tularosa phase in the Point of Pines area (Adams 1994a: 87–96). A single
3/4-trough metate from about this same time period was recovered from a
Classic-period context at Gibbon Springs in the Tucson Basin (Fratt 1996b:
301), and one or two were found, probably in Classic-period deposits, at
Las Colinas in Phoenix (Teague 1981: Table 26). At the San Pedro Valley
site of Second Canyon both 3/4- and open-trough metates were found, al-
though the 3/4-trough metates were associated with the later pithouse oc-
cupation, and the open- trough design with the surface structures (Franklin
1980:140); both contexts date after a.d. 1100.
The temporal and spatial distribution of flat metates has a somewhat
different trajectory than the trough metates. Among Pueblo II contexts at
Alkali Ridge (roughly a.d. 900–1050) in southeast Utah there are flat metates
in bins (Brew 1974:240). Flat metates in bins, Utah-trough metates, and 3/4-
trough metates were each found in contexts that date a.d. 945 at Gnat
Haven, a San Rafael Fremont site in Utah (Marwitt 1986:161, Figure 1). In
a slightly later (a.d. 1050–1100) context at Pueblo del Arroyo in Chaco
Canyon, there was a slab-lined bin with a flat metate that was identified
as foreign to Chaco Canyon (Judd 1959:135, 136). There is some confusion
concerning the classification of metates recovered from Pueblo Bonito
(Judd 1959:135). Pepper used terms such as “plain surface (slab) type”
and “tabular” to describe some trough metates, and according to Judd
(1959:135), these terms were misinterpreted to mean flat metates, such as
occur in bins, by Woodbury (1954:59) and Bartlett (1933:24).
Sometime during Pueblo III, probably by a.d. 1200, flat metates in
bins replaced free-standing open-trough metates at Mesa Verde settlements
(Swannack 1969:115). Swannack (1969:115, Figure 97) illustrates three open-

trough metates, which he calls “plain/trough,” that have been redesigned
through the removal of their borders into flat metates. The same thing
happened at Higgins Flat Pueblo in the Pinelawn area by the early to
mid-1200s (Rinaldo 1956:74). There are examples of unmodified trough
metates in bins in the Flagstaff area (Bartlett 1933:7–8, Figure 4; Colton
1946:284), at Mesa Verde (Hayes and Lancaster 1975:15), at various Chaco
Canyon settlements (Schelberg 1997:1040), and at the Point of Pines area
pueblo W:10:51 (Adams 1994a: Figure 256; Wendorf 1950:29, Plate VIII).
Judd (1959:44–45) notes that the illustration of multiple bins at Pueblo del
Arroyo with both trough and flat metates is not indicative of how they
were found, but of a reconstruction effort by workers who helped clean
the room for a photograph (noted in the Chapter 3 discussion of the perils
of using the photographic record).
Martin (1979:71, Figure 11) illustrates a multiple-receptacle station at
the Apache Creek site. Rinaldo (1959:173, 177, Figure 76) illustrates and
describes flour receptacles found in the floors of several masonry rooms
at Foote Canyon (approximately a.d. 1245–1350); they are pits ranging
up to 20 cm deep and 32 cm in diameter. Some have a stone-slab bottom,
and others have sunken bowls to catch the flour. These flour receptacles
occur both singly and in multiples. The metates are set into place at an an-
gle with adobe. Some of the receptacles are described as lined with small
slabs. Some confusion is created by both Martin’s and Rinaldo’s inter-
changeable use of the terms “receptacles” and “bins” (Martin et al. 1956:
38, Figures 16–19, 22, 23; Rinaldo 1959:173, 177).
Receptacles also have been found at Turkey Creek Pueblo in the Point
of Pines area from contexts that date post–a.d. 1225 (Johnson 1965:36).
Neither grinding bins nor receptacles are mentioned in Lowell’s (1991)
assessment of households at Turkey Creek, but archived photographs il-
lustrate that metates were set at an angle within adobe-ridged receptacles
(Figure 5.13). In east-central Arizona, slab-lined bins associated with
Anasazi grinding technology are found among Maverick Mountain phase
and other contexts after a.d. 1265 at Point of Pines Pueblo and two nearby
sites, W:10:50B, and W:10:51 (Adams 1994a:93) (Figure 5.14). Thus, it ap-
pears that receptacles may have been a local Mogollon technological de-
velopment in east-central Arizona and west-central New Mexico, and that
bins were an Anasazi technological tradition.
The trough metates in bins at the Point of Pines pueblos were proba-
bly reused and may be examples of tools scavenged by, or gifted to, prac-
titioners of a different technological tradition (Adams 1994a:222). The pho-
tographic evidence indicates that none of the trough metates plastered
into bins were very deep, and some of the trough edges may have broken
off or were intentionally removed. In some contexts it appears that two tra-
ditions may have coexisted. For example, note the presence of both free-
standing trough metates and flat metates in bins at Showlow Ruin (Haury
1931a:22), at several fourteenth- and fifteenth-century settlements (such
as Canyon Creek [Haury 1934:116, Plate 70]), in the Maverick Mountain–
phase rooms at Point of Pines Pueblo, and the Point of Pines–phase pueblo
W:10:51 (Adams 1994a:287, Figure 5).
Hundreds of flat metates were recovered from Awatovi, particularly
from Pueblo V contexts, but there was scant evidence of bins. Woodbury
(1954:64–65) speculates that this might be a circumstance of site-formation
processes, and that the bins in second-story rooms were destroyed when
the structures collapsed. Similar circumstances were noted as possible at
Aztec and at Pecos Pueblo (Kidder 1932:71; Woodbury 1954:64), at Hawikuh
(Smith et al. 1966:34), possibly at Pueblo Bonito (Judd 1954:133–135), and
at Te’ewi in the Chama Valley of New Mexico (Wendorf 1953b:68). The
prehistoric dismantling of bins from a variety of remodeling episodes has
also been noted for several settlements in Chaco Canyon (Schelberg 1997:
1055). All of the metates at the fifteenth-century Pueblo Tonque in New
Mexico were flat, but there was no mention (Barnett 1969) of mealing bins
in any of the excavated rooms. Kidder and Guernsey (1919:124) noted dis-
mantled bins at several settlements in northeastern Arizona and surmised
that the metates were removed because good stone material for metates
was rare. These are just subtle reminders that the archaeological record
does not always provide all the information we need to completely recon-
struct prehistoric activities at a particular site.
In summation, I believe that the different metate designs are different
technological traditions that represent distinct ways of making and using
food-processing equipment. This brief summary highlights the move-
ments of technological knowledge, probably via the migration of individ-
uals or small groups who replicated their learned traditions in the areas
where they relocated. Relative proportions of tool designs, as mentioned
previously, may reflect the size and type of migration. For example, the
one or two occurrences of open-trough metates at settlements with pre-
dominately 3/4-trough metates may be the result of the immigration of a
single household, or the intermarriage of a single woman. Within this
mode, the movement of the open-trough tradition seems to have been more
common to the north prior to a.d. 1000 than the movement of the 3/4-
trough tradition to the south. Increased frequencies of one metate design
in an area previously dominated by a different design could reflect the
migration of larger social units. Such meeting of different technologies
happened when northern Anasazi technological traditions using flat

metates in bins were brought south sometime after a.d. 1250 into the Point
of Pines area, where indigenous technology used free-standing trough
metates or grinding receptacles.
It also has been possible to recognize the concomitant development of
concepts concerning the placement and number of grinding stations. The
concept of permanent grinding stations led to two different designs: the
bin, originating with the Anasazi technological tradition, and the recepta-
cle, originating with the Mogollon technological tradition—two different
solutions to the problem of confining the product during processing. Both
traditions expanded the number of grinding stations by positioning them
next to each other. The stations were not scattered around the room, nor
situated so the grinders faced each other. Both traditions hit upon a single
solution for increasing the number of grinders who could work together
(the sociocultural implications of which are beyond the scope of this dis-
cussion). The final point to be made is that food-grinding technology in
particular, and grinding technology in general, varied through time and
across space in the U.S. Southwest—a fact that has been underutilized in
attempts to understand the dynamics of prehistory.
MORTARS
Mortars are designed with a basin that confines an intermediate substance

that is worked with a pestle in some combination of crushing, stirring, or
pounding strokes. Crushing is a low-impact action that depends mostly on
the weight of the pestle. Use-wear patterns include uniformly configured
impact fractures within the mortar basin. If stirring is part of the crushing
movement, striations are created across the margins of the fractures and,
to some extent, within the fractures. When wooden pestles are used or the
intermediate substances are particularly soft, the margins of the impact
fractures and any protruding rock grains become rounded. Pounding is a
high-impact action in which the tool is raised higher and brought more
forcefully into contact with the intermediate substance in the mortar. The
impact fractures generated by pounding are deeper and more jagged than
those resulting from crushing. However, if the intermediate substance
worked in the mortar is thick or sticky, it cushions the impact, keeping the
mortar and pestle out of contact with each other and lessening impact
damage to both tools.
The designed sizes and configurations of mortars vary greatly (Wheat
1955:118–119; Woodbury 1954). Expediently designed mortars have no mod-
ification other than the manufacture of the basin. Strategically designed
mortars range from having simple modifications that make them rest flat
on the ground to elaborate modifications that create geometric or morphic
shapes (Figure 5.16).
Mortar Subtypes
Pebble mortars are small, and only small quantities of material could have
been worked in their basins, which are usually less than 5 cm in diameter
and generally range from 1 to 2 cm deep. Striations occur more often than
impact fractures in some pebble-mortar basins and are probably the result
of mixing rather than crushing activities. Basin size and the small size of
any companion pestle are not conducive to the heavy crushing that can
happen in larger mortar basins with larger pestles. The rocks into which
the small basins are manufactured are not always further modified, and
despite their name, pebble mortars are not always made from pebbles.
Furthermore, other mortars sometimes have basins that are the same small
size as pebble mortars, creating fuzzy boundaries among the mortar types.
Perhaps a label descriptive of basin size should be used; however, the
term has been in the literature for at least 40 years and has some concep-
tual utility (see, for example, Martin 1979:71; Wheat 1955:118–119). Thus,
the term pebble mortar is retained here as a category of relative basin size
in contrast to the categories of rock, boulder, and bedrock or stationary
mortars. Woodbury (1954:116–119) uses the terms “roughly-shaped” and
“bowl-shaped” mortars, which are subsumed in this technological ap-
proach under the assessment of design corresponding with expediently
and strategically designed mortars.
Rock mortars have basins pecked into portable-size rocks that were
not always further modified. They are larger with larger basins (generally
greater than 5 cm in diameter and deeper than 2 cm) and were used with
larger pestles than used for pebble mortars (Figure 5.17). Boulder mortars
have basins that are not necessarily larger than those of other mortars, but
are pecked into rocks too large to be easily moved. The basins that some-
times occur in these rocks are often called “cupules.” Stationary mortars
have basins pecked into unmovable rocks and include those found in caves
or cliff overhangs, on bedrock outcrops (sometimes called “bedrock mor-
tars” [Woodbury 1954:117]), and those cemented into structure walls. Porta-
bility is the attribute that sorts stationary or bedrock mortars from the
others.
Basin size has not traditionally been used to define mortar types, even
though it might be a more meaningful attribute than the size and shape of
the rock from which it was manufactured. Stationary mortars with small
5.16. Shaped mortar from a Cienega-phase context at Santa Cruz Bend in the
Tucson Basin. Note the flattened margins between the vesicles on the basin bot-
tom, the result of use-wear damage from a stone pestle. (Adams 1998: Figure
10.17a.)
5.17. Rock mortar with compatible pestle from a Cienega-phase context at Stone
Pipe in the Tucson Basin. (Adams 1998: Figure 10.18.)
basins comparable to those of pebble mortars are not easily classifiable

within the traditional mortar categories. Fuzzy boundaries exist between
mortars and items that have been variously referred to as cupped or pit-
ted stones. Refer to the section below, “Pitted and Cupped Stones,” for
definitions and clarification of the classification of these items. As an addi-
tional caveat, careful consideration should be taken in identifying bedrock
mortars. Among historic Pueblo groups, a single basin pecked into a rock
outcrop is used to signify trails or boundaries. Use-wear analysis may be
the only way to distinguish them, but weathering may have obliterated
any evidence of use wear caused by a pestle. A single basin on an isolated
outcrop is probably more likely a marker than a processing tool.
Shaped mortars are carefully manufactured to specific shapes. This cate-
gory has no typical member and cannot be defined based on size. Use wear
in the basins is the most reliable attribute for distinguishing shaped mor-
tars from bowls or other containers (Figure 5.18). Some vessels identified
rim width basin width
basin depth
thickness
base
width
length
5.18. Shaped mortar. Note that the distinguishing attribute between this artifact
and a plain bowl is use wear in the basin. Without the use wear, this would have
been sorted into the container category. (Drafted by Chip Colwell.) (Adams 1996:
Figure 23.)
as trays have use wear in them from contact with pestles. Ferg (1998: Fig-
ure 14.11) illustrates two trays that were found with their triangular pes-
tles. This allows for questioning whether trays were designed as shaped
mortars, and those without use wear from pestles were simply unused,
or whether trays designed as containers had some secondary use as
mortars.
Mortars sometimes have pigment in their basins. The presence of pig-
ment and use wear in the basin-like depressions of some disks found at
Santa Cruz Bend in Tucson influenced the decision to call them mortars
similar to pebble mortars (Figure 5.19) and to consider them as something
other than unfinished perforated disks (Adams 1996). Other Santa Cruz
Bend disk-shaped mortars and some pebble mortars have basins that are
unused and unburned, but with surrounding margins or rims that are
smoke-blackened (Figure 5.20a). These might have been something other
than unused mortars or unfinished perforated disks. The possibility of
their having been oil-burning lamps was explored because of their simi-
larity to Old World oil lamps found in prehistoric cave deposits that date
to at least 17,500 years ago (de Beaune and White 1993). An exploratory
experiment kept a cotton wick burning in a small, shallow basin filled
with corn oil for at least 15 minutes (Adams 1996). Thus, a lamp function
for this configuration cannot be totally discounted. With better wick tech-
nology, the lamp could have burned longer. The use of oil-burning lamps
has not been explored in the prehistoric U.S. Southwest, and the possibil-
ity deserves further research.
Most of what is known about mortar use comes from ethnographic re-
search, and this is a good case for developing source-side criteria that can
be used in model building. For example, the assumption is that prehistoric
women used mortars and pestles in food-processing activities because
that is who used them historically (Castetter and Bell 1951:96; Doelle 1976;
Euler and Dobyns 1983:259; Fontana 1983: Figure 6; Jackson 1991; Spier
1933:57, 96; Webb 1959:12). There are fewer descriptions of who makes the
mortars, although Jackson (1991) reports that Western Mono women in
California manufacture two types of bedrock mortars: “starter mortars,”
the shallow basins used in the initial cracking and pulverizing part of the
task, and “finishing mortars,” the deeper basins in which flour is reduced
to the desired texture. Mono women manufacture their own mortar basins
to the desired size (Jackson 1991:307). It is unclear, however, if a starter
mortar eventually wears enough to become a finishing mortar.
basin width margin
basin depth
thickness
length
width
5.19. Disk mortar. Use wear in the depression allowed for the classification of
this tool as a mortar. Are those without use wear unused disk mortars, or unfin-
ished donut stones? (Drafted by Chip Colwell.) (Adams 1996: Figure 15.)
5.20. A disk mortar (a) with soot around the basin, and (b) a mortar with a
handle. Or were these oil-burning lamps? (Adams 1998: Figure 10.9c,e.)
The range of resources processed in various types of mortars is also

gleaned from ethnographic sources. Acorn flour is the product most com-
monly processed by Western Mono women in the bedrock mortars, al-
though lesser amounts of other seeds, fruits, vegetal remains, clay, meat,
and pigment are also processed in them (Jackson 1991:305). Small, portable
mortars were also used by the Mono, but mostly for meat and pigments.
Both bedrock and portable mortars have been used by the Walapai, Mari-
copa, Pima, and other non-Pueblo groups to crush the pods of mesquite
beans (Castetter and Bell 1937; Doelle 1976:53–68; Euler and Dobyns 1983:
259; Jackson 1991; Spier 1933:51). Once the pods were broken apart with
mortars and pestles, they could be reduced further with a mano and
metate (Castetter and Bell 1937:24–26, 1951:184, 185; Spier 1933:51). Food-
processing mortars are identified by Hopi as used to pound dried meat to
soften it for those who had no teeth (Adams 1979:25). A few chemical
analyses have been able to detect the presence of animal immunoglobins
on mortars and pestles in California, adding another line of evidence for
inferring what was processed with prehistoric mortars and pestles (Yohe
et al. 1991:663).
Research at the historic Hopi village of Walpi (Adams 1979:25–26)
helped identify attributes that distinguish food-processing mortars from
vessels used to water eagles. Those identified as food-processing mortars
have flat to rounded bottoms, variously shaped rims, and deep, conical,
or cylindrical basins that have both impact fractures and numerous stria-
tions caused by the pestle. Those identified as eagle watering bowls (Fig-
ure 5.21) (Adams 1979:26, Plate 5) have flat bottoms, square rims, and
broad, deep basins damaged only through manufacturing (impact frac-
tures and only minor striations). A similarly shaped bowl was described
and illustrated from a pithouse context at Kiatuthlanna (Roberts 1931: Fig-
ure 34); however, because the size is not indicated, it is hard to determine
if this bowl could have served as a watering bowl. As described by Di
Peso et al. (1974:206, 274–7), the stone bowls found outside the macaw
pens at Casas Grandes are similar to the eagle watering bowls from Walpi.
Woodbury (1954:118–119) discusses the difficulties in distinguishing
mortars from bowls or other vessels. He notes that other researchers have
concluded that bowls were probably never as plentiful in the northern
Southwest as they were in the south, especially among Hohokam tradi-
tions. Boulder and pebble mortars are both common in central Arizona at
sites attributed to the Mogollon (Martin 1943: Figure 66; Wheat 1955:118–
119). It is more difficult to date the associations of bedrock mortars, which
have been found all over the U.S. Southwest. A possible association of
bedrock mortars with an early preceramic occupation at Ventana Cave
was recognized by Haury (1950:320).
One final piece of information derived from ethnographic references
provides a possible explanation for how mortars reach the end of their
use lives. Euler and Dobyns (1983:259–262) describe mortar and pestle
use by the Pai and illustrate a deep basin mortar that was broken, probably
5.21. Eagle watering bowl. The lack of use wear in the basin, the broad, squared
border, and the flat bottom distinguish this from a mortar. (Adams 1979: Plate
5d.)
upon the death of its user. The intentional destruction of personal tools
has also been mentioned for ending the use of metates, and the tools are
referred to as having been “killed” (Haury 1985b:244; Schelberg 1997:1068).
PITTED AND CUPPED STONES
This category is included here only because “pitted” and “cupped” stones
are so commonly mentioned in the archaeological literature. A careful tech-
nological analysis should be able to sort most of them into more meaning-
ful categories, such as pebble mortars, anvils, spindle bases, or fire-drill
hearths. Pitted and cupped stones have often been considered an enigma
in that their function is unclear. They are commonly classified on the basis
of having a single common attribute: one or more small depressions called
pits, cups, or cupules. These terms are what cause most of the confusion.
Sometimes the pits are nothing more than a concentration of impact frac-
tures that may be from the incomplete manufacture of a basin or from ex-
tensive use as an anvil. In such cases the impact fractures have sharp edges
and no other use-wear damage.
If the impact fractures were made to serve as finger grips, they should
be rounded, smoothed, and perhaps shiny, depending on the amount of
handling. Rounded impact fractures also can result from the spinning of a
wooden spindle. (See Chapter 8, “Spinning Tools,” for a discussion on
spindle bases and fire-drill hearths.) If the impact fractures are concen-
trated enough to form a basin, and there is no other damage in the basin,
the item is either a container or an unused mortar. Obviously, if there is
pigment in the basin and use wear from a pestle, it is a small mortar used
to process coloring agents. A container of processed pigment should have
no use wear in the basin from a pestle, but may have impact fractures
from the manufacture of the basin.
Small mortar basins are sometimes manufactured into the surfaces of
other tools. The important attributes to evaluate for recognizing primary
and secondary uses are the locations and orientations of impact fractures
and striations. For example, a basin in the surface of a mano is an indica-
tion of secondary use. If the impact fractures that create the basin are
sharp and intrude over the striations from mano use, the basin is from a
sequential secondary use. If the striations round over the basin edge or
level the basin margin, the mano continued in use after the basin was
manufactured, making the basin a feature of concomitant secondary use.
Archaeological and Experimental Comparisons
A series of experiments was conducted to compare the pits in netherstones

used during nut cracking and spinning a bow drill (Rowe 1995). The pre-
liminary results are reassuring. The cupule that confined the bow drill was
more conical and had a smoother interior than the nutting-stone cupule.
If there were impact fractures, they were remnants from cupule manufac-
ture (Rowe 1995:14). Thus, use-wear analysis may be the only way to dis-
cern how a pitted or cupped stone was used. However, even that might
not be enough to distinguish between spindle bases and some fire-drill
hearths. (Refer to Chapter 8, “Spinning Tools” for additional descriptions
of these tool types.)
Cupped stones have been recovered from such diverse locations as
Pecos Pueblo (Kidder 1932:66), various Mogollon sites (Wheat 1955:120),
and Awatovi (Woodbury 1954:184). Woodbury (1954:184) rejects the pos-
sibility of their use as bases for bow or pump drills primarily because the
depressions lacked macroscopically obvious use wear. However, it is un-
clear how much or what type of use wear he expects to be visible. For use
with a pump drill the basin may have served merely to hold the item be-
ing drilled. Use with a spindle whorl may create wear so light that it is
not macroscopically visible. Woodbury (1954:90) suggests that in some
cases pits may be remnants of secondary use as lithic anvils or platforms

for holding other items as they are worked.
Hayes and Lancaster (1975:156–157, Figure 197) describe a collection
of “pitted rubbing stones” from the Badger House Community in Mesa
Verde. They are not satisfied with the postulation that the pits served as fin-
ger grips, because there are no impact fractures from their use as pounding
stones, as was found on those recovered from Big Juniper House (Swan-
nack 1969:126–130, Figures 116, 117). The “pitted pounding stones” from
Mug House also have use wear from their use as hammerstones (Rohn
1971:210, Figure 246). Many pits are found as features of secondary use on
manos, handstones, and polishing stones (Wheeler 1980:282; Hayes and
Lancaster 1975:156–157; Rohn 1971:210). In southwest Colorado, the pit-
ted stones are most common in Pueblo I contexts and less so in Pueblo III
contexts (Hayes and Lancaster 1975:156). Those from Awatovi were in con-
texts dating Pueblo III–V (Woodbury 1954:91).
PESTLES
Pestles are handstones used to pulverize, or to crush and grind. They vary
in the complexity of their design, especially in size and shape. Some are of
expedient design, where appropriately sized pebbles or cobbles were se-
lected and used without further modification. Others are of strategic de-
sign, carefully manufactured to specific shapes, or with finger grips or
notches for holding. Pestle size is probably most often related to intended
function. Larger, heavier pestles are used to break and crush materials,
whereas smaller, lighter pestles are used to crush, grind, and stir. Pestle
manufacturing techniques are described by Schneider (1996:306) based
on evidence recovered from quarries in southwestern Arizona and south-
eastern California. The presence of large and small hammerstones at the
quarries, and the flake scars and impact fractures on the pestle blanks in-
dicate that the larger hammerstones were used to rough out the pestle
forms, and the smaller hammerstones to do the final shaping.
The configuration of the surface against which pestles were used can
be determined by the location and nature of use-wear damage. An ethno-
graphic report of a Maricopa woman using a stone pestle to crush chunks
of pottery clay on a blanket (Sayles and Sayles 1948:29) should make us
more carefully assess the damage patterns on archaeological pestles and
not assume that all pestles were used in mortars. The location of the dam-
age helps identify whether the pestle was used in the basin of a mortar or
against a flat surface. In general, use-wear damage includes impact frac-
tures, chips, and abrasion. Pestles used on flat surfaces, such as the ground,
5.22. Pestles from Santa Cruz Bend: (a) used against a flat, hard surface; (b) and
(c) used in stone mortars. (Adams 1998: Figure 10.19.)
have wear concentrated on the flattest part of their distal ends (Figure
5.22a). Those used in mortars have damage on their distal ends and along
the sides that came in contact with the mortar basins (Figure 5.22b, c). The
damage from contact with stone mortars is angular and chipped, whereas
damage caused by wooden mortars is rounded (refer to Chapter 2, for
use-wear descriptions).
The nature of the damage also reflects the type of stroke used with
the pestle. Small, evenly spaced impact fractures are created when the
weight of the stone supplies most of the force used to crush intermediate
substances. Deep impact fractures and chips reflect the use of a forceful
downward or pounding stroke. Striations indicate circular or reciprocal
movements of the pestle in the mortar basin to grind the intermediate
substances. Remnants of pigment are evidence that some pestles were used
to process substances other than food (Adams 1998:385). Pestles need not
be of stone, and use-wear patterns in some mortar basins suggest the work-
ing of wooden pestles, as described in the section on mortars.
One carefully shaped pestle subtype has been commonly referred to
as a “tooth pestle” because of its resemblance to a tooth (Figure 5.23), or a
“notched pestle” because of a notch in the proximal end (Ferg 1998:589–
595). The shape is generally triangular with a broad distal end that is used
against the mortar, and a narrower proximal end on which it is some-
times possible to detect hand wear. Some have a biconical hole drilled
through the body, and some have both a notch and a hole. Most are made
of basalt, but quartzite has also been used. The groove and the hole may
be attributed to a design for secondary use as an abrader or straightener
(Adams 1998:385). Because other triangular pestles lack the notch, the hole,
or both, I propose that tooth and notched pestles be classified as triangular
pestles, with the notches and holes considered attributes of secondary use.
Triangular pestles with various configurations of notches and holes appear
early in southern Arizona and have been found in Cienega-phase deposits
(800 b.c.–a.d. 150) (Adams 1998:385, 2000b; Ferg 1998:589–595).
Other secondary uses have been noted for some pestles that have grind-
ing surfaces on their sides (Adams 1995:84). More than one possibility can
explain this configuration. One is that the pestle was used concomitantly
in a two-step processing activity that required both the crushing action of
a pestle and the grinding action of a handstone. For example, the pestle
could have been used to crush mesquite pods in a mortar and then also
used against a metate to further reduce the mesquite meal to flour. The
second explanation is that a pestle could have been used in one activity,
such as to crush mesquite pods in a bedrock mortar, and then used in an
unrelated activity, such as to grind pigment against a netherstone. The
pigment on only the grinding surface may be the only evidence that the
pestle was used in two different processing activities.
Even without use-wear analysis, we can speculate that wooden mortars

were used prehistorically because they were used by historic tribes (see,
for example, Castetter and Bell 1951: 96 for the Yuman tribes; and Russell
1975:109 for the Pima). One historic photograph depicts a Pima woman in
front of a house with a large stone pestle in a tall wooden mortar used to
process mesquite beans (Fontana 1983:132, Figure 6; Russell 1975). In an-
other photograph, a stone pestle in a wooden mortar is clearly illustrated
proximal end
distal end
outer hole
diameter
inner hole
diameter thickness
biconical hole
groove width
groove depth
length
width
5.23. Pestle used in more than one activity. Use wear in the notch and the hole
suggest that it was used against a pliable surface, perhaps a wooden shaft. The
distal end has use wear from contact against a stone mortar. Such pestles are
sometimes referred to as tooth pestles. (Drafted by Chip Colwell.) (Adams 1996:
Figure 26.)
as part of a Pima woman’s pottery-making kit (Hayden 1959:10–16, Fig-

ures 1–3). Thus, food processing may not be the only activity in which
stone pestles and wooden mortars were used together. As with manos
and metates, the use of food-processing mortars was probably by women,
assuming that contemporary practices extended into the past.
Certain long, slender, similarly cylindrical pestles or pestle-like tools
deserve special description. A cache of 11 such tools was recovered from a
pithouse at Crooked Ridge Village in the Point of Pines area (Adams 1994a:
121–122, 150; Wheat 1954:117, 1955:120). Even though the tools are all
similarly shaped, the use-wear patterns on each are different. Wheat
(1954:117, 1955:120) postulated that at least some of the tools in the cache
were digging tools. I have suggested that the cache was of tools used in
different stages of pithouse construction (Adams 1994a:121–122). Five have
hoe-like blades and striations consistent with their use in soil. These could
have been digging implements used to dig postholes. Others have impact
fractures from their use as hammerstones, but the use-wear damage is
consistent with contact against wooden surfaces; perhaps these were used
to split planks or pound beams into place. One has no use wear on the
end but has a large concave facet worn along one edge; such use wear is
consistent with wood contact, and may have resulted from burnishing
beams. The lesson learned from this cache is that not everything that is
long and cylindrical was used in the same manner.
HANDSTONES
As explained in the introduction to this chapter, the term handstone has

two meanings. One refers to the large set of all handheld tools, and the
second to the tools lacking specific attributes that allow them to be sorted
into defined subsets such as manos, abraders, polishing stones, pestles,
and so on. Examples include the small stones used to process pigments or
to mix other substances on lapstones and netherstones. They are usually
expediently designed from small, smooth pebbles that are approximately
5 cm to 20 cm long. Handstones that are strategically designed have fin-
ger grips or grooves to make them easier to hold. The stones selected for
use range in texture from smooth to very coarse. The orientation of stria-
tions and sheen should provide adequate evidence for the nature of the
contact surface. Hide-processing stones are an example of a handstone
type that was defined through use-wear analysis and ethnographic analogy.
In some classification schemes the category of handstone includes
small manos (see, for example, Rinaldo 1956:55, 1959:227). Confusion is
created by assuming that because small manos were sometimes used to
grind things other than maize, they should be considered handstones and
distinct from larger manos, which are assumed by some to have been
used only to process maize (Diehl 1996:106, 109; Haury 1976:281; Lan-
caster 1984:247; LeBlanc 1983:42; Rinaldo 1959:227). As clarified by the
previous discussion on manos and metates, and by previous experiments
(Adams 1999), in the typology presented, any handheld stone used against
a metate of any design is classified as a mano. If there is no evidence of
contact with a metate, it is a handstone. Evidence of design, use wear,
and kinetics can be used to help sort manos from handstones. The most
difficult distinction to make is that between a handstone used on a flat
netherstone, and an expediently designed flat/concave mano; often the
presence of a wear facet from a rocking stroke distinguishes the mano.
Handstones are perhaps too ubiquitous for a worthwhile analysis of their
distribution through time and across space. Also, because they have been
lumped with manos in past classifications, it would take considerable ef-
fort to sort out handstones and evaluate their distributions.
NETHERSTONES
The two different uses of the term netherstone are discussed in the intro-
duction to this chapter. As with handstone, the term has both generic and
specific meanings. Netherstones are bottom stones against which something
was worked (Figure 5.24). In a generic sense, netherstones comprise a large
set within which tools such as metates, mortars, lapstones, lithic anvils,
and others are subsets (Figure 4.6). Those that cannot be sorted into sub-
sets are labeled netherstones. Netherstones are expediently designed un-
less effort has been expended in making them a particular shape, or they
have been modified to sit more securely on the ground. They vary in size
but are generally too large to be handheld. Use-wear patterns include abra-
sion, impact fractures, sheen, or a combination of these. They are not all
necessarily worked with a handstone. Some serve as working surfaces
upon which other items are shaped.
Ladd (1979: Figure 5) illustrates a Zuni man shaping a string of beads
against a flat netherstone. Given this possible use, a netherstone’s surface
should be carefully examined for use-wear patterns consistent with such
an activity. Experimentation is needed to identify a range of other possible
uses. One example of inferred use of a netherstone based on archaeologi-
cal context comes from Showlow, where Haury (1931a:23, Plate 11, Figure
2) found a netherstone that was considered a “potter’s kneading board”
because of its association with other pottery-making tools. The nether-
stone contained traces of clay, adding even more evidence to support the
5.24. Netherstone with use wear from a smooth handstone. (Adams 1998: Figure
10.16.)
inference. Occasionally it has been possible to recognize the handstones that

were probably used with specific netherstones (see, for example, Adams
2000a, 2001b). In these situations the handstones and netherstones were
recovered from the same contexts and have compatible surface configura-
tions and use-wear damage. A few have even been recognized as having
the same shade of pigment on them from pigment-processing activities.
These archaeological associations and ethnographic examples demonstrate

the range of possible uses for netherstones, which means that even though
they may seem plain and uninteresting, it is possible to make behavioral
inferences through use-wear analysis and analogy.
GRINDING SLABS
Grinding slabs are a subset of netherstones (Figure 4.6) and are tradition-
ally distinguished by their tabular or slab shape. The slab is usually the
natural shape of the material from which it was made, although some are
strategically designed as slabs, manufactured through pecking and grind-
ing. Grinding slabs are too large to be handheld, making them distinct from
lapstones, even though both are worn through abrasive and percussive
activities. Thus, there are fuzzy boundaries between grinding slabs, nether-
stones, and lapstones. The categorization of palettes, grinding slabs, and
mortars by Woodbury (1954:113–116) illustrates the confusion and fuzzy
boundaries between these types. Woodbury (1954:113) admits that some
of the items he categorized as grinding slabs are more similar to shallow
mortars, thus compounding category confusion. Grinding slabs probably
have not been consistently enough sorted out from netherstones to allow
for meaningful comparisons of their distributions through time and across
space.
LAPSTONES
Lapstones are handheld netherstones that serve as bases upon which other
items were shaped or intermediate substances processed with small hand-
stones. The sets netherstones and handstones intersect to create a cate-
gory labeled “lapstones” (Figure 4.6). Lapstones are more often expediently
than strategically designed, with the most effort expended in selecting
the proper material texture for their intended use. The sizes range from
small, flat pebbles that are 5 cm in diameter to larger pieces that are 15 cm
on a side. By definition, lapstones are never larger than what can be held
in one hand. Sheen, abrasion, or impact fractures are left on lapstone sur-
faces depending on the nature of the work done on them. Those that are
used to shape other objects either polish or abrade them depending on
lapstone surface texture. Lapstones differ from flat abraders and polishers
in that they are the bottom, passive stones against which something is
worked, whereas abraders and polishers are the active stone worked against
another surface. (Refer to Chapter 4, “Abrading, Smoothing, and Polish-
ing Tools,” for additional information on these tools.)
Some lapstones have evidence of their use in pigment processing, either

for grinding minerals into powdered pigment, or mixing the pigment with
a binder or vehicle such as water or oil to create paint (Adams 1998:385).
Past descriptions have used the term “proto-palette” for such lapstones,
but this creates confusion by assuming an evolutionary sequence leading
directly to the elaborate palettes found in later Hohokam deposits (some
of which may not have been pigment-processing tools), and leaves un-
classified some tools of the same simple configuration that occur in all
time periods but lack pigment. As defined here, lapstones are distinguished
from palettes by their simpler design. Palettes are embellished with bor-
ders and sometimes decorated with incised or carved designs, whereas
lapstones are plain, without borders or decorations. Lapstones used in
pigment processing are distinguished from mortars by the lack of a basin.
Extensive use may result in worn depressions, but these are easily distin-
guishable from the manufactured basins of mortars. Mortars are also dis-
tinctive in being worked with a pestle, in contrast to lapstones, which are
worked with a handstone. Time and space distributions of lapstones as
described here have not yet been done. Future research topics might in-
clude comparisons of the distributions of lapstones and palettes, which
could then be used to clarify any postulated relationships between the
two tool types.
PALETTES
As used here, the term palette labels a fuzzy subset within the set of lap-
stones (Figure 4.6). Woodbury (1954:114) also notes that types such as lap-
stones, palettes, and mortars grade into each other, thereby causing con-
fusion in classification. The fact that some palettes have depressions worn
in them from use with a small, smooth handstone increases the fuzziness,
as discussed in the previous section. To clear up some of this confusion,
palettes are defined here as specialized lapstones embellished with bor-
ders and made from schist, phyllite, or similarly tabular material. More
often than not, palette borders are decorated with incised lines of varying
motifs (Figure 5.25). As defined from the excavations at Snaketown, palettes
must have had ritual significance because of their pervasive association
with mortuary contexts; some also have evidence of use with censers, and
others have various minerals burned on them (Haury 1976:288; Hawley
1975:282–289). Woodbury (1954:112–115) infers that palettes were used to
process pigments, which is valid within the typical dictionary definition
of the term but may not be true of those found in the archaeological record.
border
basin width width
thickness basin depth
base
width
length
5.25. Palette. Impact fractures in the bottom indicate that this palette received
some use. The border distinguishes it from a lapstone. (Drafted by Chip Colwell.)
(Adams 1996: Figure 25.)
It should be noted that Hawley’s (1975:282–289) chemical analysis of

the residues on palettes from Snaketown revealed varying compositions
of lead silicates and lead oxides. He hypothesizes that powdered lead car-
bonate mixed with vegetal juices or gums was applied to the palettes,
which were then heated, causing lively chemical interactions and leaving
crusty residues (Hawley 1975:284). At temperatures of around 300 degrees
centigrade, the residues are strongly red. The caveat is that red residues
on palettes may not be from processing pigment. Rather, they could be
remnants of lead oxides, especially if there is a crusty texture to some of
the residues. More chemical studies like Hawley’s are needed to distin-
guish those palettes with pigment from those with lead residues. It can be
assumed that palettes with the crusty residues were subjected to high tem-
peratures, even if there is no evidence of smoke blackening. Hawley (1975:
285) speculates that the palettes were heated after the fire had burned
down.
Archaeological Comparisons
Roberts’s (1940:122, Plate 43) discussion of palettes as a Mogollon- or

Hohokam-derived item in the White Water District, causes confusion when
he includes sandstone mortars with pigment in the category of palettes,
which he sees as more of an item of Anasazi technology. Woodbury (1954:
114) fails in his attempt to make an evolutionary connection between Ho-
hokam palettes and the bordered mortars from Awatovi that he calls pal-
ettes. Only through a more careful use of terms and more detailed analy-
ses will it be possible to sort out the different technological traditions that
produced the varying configurations. Such analyses are currently being
conducted by Devin White and Steven Lekson (White and Lekson, per-
sonal communication, 2000). Figure 5.26 is the most recent illustration of
the attributes used to describe and analyze palettes.
A chronological sequence of palettes recovered from Snaketown il-
lustrates how the elaborateness of palette decoration reached a peak
through time (Haury 1976:286). The palettes found in Colonial and Seden-
tary contexts at Meddler Point in the lower Tonto Basin (Adams 1995:84,
Figure 2.10e) and at Las Colinas in the Phoenix Basin (Euler and Gregory
1988:305) fit well with the trends defined at Snaketown. One broken palette
from Meddler Point was partially reconstructed from pieces found in two
different cremation features (Adams 1995:84, Figure 2.10e). At Julian Wash,
a Rincon-phase site in the Tucson Basin, the palettes were also associated
with cremations, some intentionally broken and covered with lead residues.
In contrast, those from Los Morteros, another Rincon-phase site in the
Tucson Basin, are less elaborate and are more often associated with struc-
ture floors and fill than mortuary contexts (Dart 1995). If they lost their as-
sociation with mortuary ritual somewhere along the line, their presence
among the paraphernalia of Pima medicine men suggests that they re-
mained ritually used into historic times (Russell 1975:112, Figure 33).
Palettes found in the Mimbres area and at the Galaz Ruin have pig-
ment stains and striations that are interpreted as having been caused by
grinding or abrading uses (Anyon and LeBlanc 1984:272). These palettes
are more common in Late Pithouse (a.d. 550–1000) than in Classic Mim-
bres contexts (a.d. 1000–1150). Whole palettes were associated with buri-
als. Those with the most elaborate decorations were found on pithouse
floors (Anyon and LeBlanc 1984:273). Although a similar developmental
sequence is noted among palettes in the Mimbres and Hohokam areas, it
5.26. Terms used to describe palette attributes based on those originally used by Haury
(1975: Figure 49) and refined by Devin White. (Illustration generated by Devin White.)
is striking that Mimbres palettes are associated with burials, and Hohokam
palettes with cremations (Anyon and LeBlanc 1984:273). One unusual
palette from Galaz is illustrated with four supports on the base (Anyon
and LeBlanc 1984: Figure 19.2).
A possible use of palettes as snuff trays has been postulated, espe-
cially for those that have no evidence of residues and were not broken or
burned in mortuary rituals (Lowell 1990). This possibility stems from ethno-
graphic research into South American snuffing associated with curing
and divination (Lowell 1990:326). In these historic contexts small mortars
are also part of the processing or storage paraphernalia for the drugs that
were used on the snuff trays (Lowell 1990:327–328). Other associated para-
phernalia include bone tubes for inhaling or storing powder.
Chapter 6
Percussion Tools
T he purpose of percussion tools is to chip or smash away the unwanted

parts of other items. The forceful strokes used with percussion tools
cause impact fractures that damage surfaces through the mechanisms of
fatigue wear. Hammerstones, pecking stones, and choppers are all varia-
tions on a design scheme, differing primarily in edge preparation, kinet-
ics, and to some degree, size. The configuration of the percussion tool and
the force of the stroke dictate the nature of the damage done to the contact
surface. Rounded tool surfaces create broader, shallower impact fractures
on the contact surface than those created by sharp edges. The rounded
surfaces are themselves damaged by shallow to deep impact fractures,
depending on the stroke force. Damage on edges appears as both impact
fractures and flake scars. Certain subsets of handstones, netherstones, and
lapstones (Figure 4.6) were also used in percussion activities and are rec-
ognized by distinctive use-wear patterns, as described in the following
sections.
HAMMERSTONES
Generally, hammerstones are irregularly shaped rocks selected for their use-
ful size and weight, and they are often expediently designed for use with-
out further modification (Haury 1976:279). Strategically designed ham-
merstones have finger grips or other modifications for holding. Primarily
natural edges, but sometimes broad surfaces, are used with forceful strokes
against other surfaces. Hammerstones are sometimes large enough to re-
quire the use of two hands, such as those identified as used in quarrying
activities (see, for example, Schneider 1996:306). They are wielded with
more forceful strokes than pecking stones. The severity of the impact frac-
tures and flake scars becomes greater with more extensive use and with
more forceful strokes. Some hammerstones serve as billets used in flaked
151
tool manufacture. This use generally creates a broad, beveled edge covered
with impact fractures.
In ethnographic documentation of the tools used in Guatemalan metate

manufacture, size was recognized as functionally distinct (Hayden and Nel-
son 1981; Hayden 1987). Two-handed implements were used for initial
quarrying and initial shaping, and progressively smaller tools for shaping
and finishing. The amount of damage to the shaping tools was directly
correlated to the force of the blows. Perhaps together, size and damage
patterns might be useful for constructing a typology. However, as pointed
out with other tool types, size alone is not a good variable for comparing
assemblages, especially if factors such as raw material sizes limit tool con-
figuration.
There is a fuzzy boundary between hammerstones and mauls. Both
have the same general percussion function and incur the same use-wear
damage, but they are easily distinguished by the presence of a groove
that allows maul to be hafted. Wheeler (1980:245) makes a similar distinc-
tion between what he calls “hammerstones” and “hammer heads.” He
distinguishes hammer heads from maul heads on the basis of size. The
preference here is to include all grooved hammers in the set of mauls and
let size remain a nonclassificatory, descriptive variable. As Woodbury (1954:
91–92) notes, the inconsistent use of terms and the insufficient descrip-
tions of hammerstones in past reports make it impossible to discuss dis-
tributions and changes through time in percussion tools. It is perhaps suf-
ficient to know that they have been found throughout the U.S. Southwest
in contexts that date as early as the Late Archaic and as recently as the his-
toric period (see, for example, Adams 1979:57, in which they are called
“pounding stones,” 2001b; Haury 1976:279–280; Hayes and Lancaster 1975:
149–150, Figure 187; Hayes et al. 1981:108; Kidder 1932:60–63; Swannack
1969:127–130, Figures 113–115; Wheeler 1980:243–245, Figures 274–275;
Woodbury 1954:85–93).
PECKING STONES
Pecking stones are handstones used in light-duty percussion activities. They

have a fuzzy boundary with hammerstones (Figure 4.6). The basic design
of hammerstones and pecking stones is the same, and both are used with
forceful strokes that create impact fractures and chips. The main differ-
ence is that pecking stones are used with less force than hammerstones,
6. P ERCUSSION T OOLS 153
and the resulting impact fractures are more uniform and evenly spread
across pecking stone surfaces than those on the surfaces of hammerstones
and mauls. Pecking stones are used to shape manos, metates, and other
large objects, or to renovate worn grinding surfaces. Some have distinc-
tive use-wear patterns, with impact fractures rounded from contact with a
pliable intermediate substance. Such use wear may be distinctive of pulp-
ing stones used to smash fruits or split leaves for fiber extraction (Bernard-
Shaw 1990:190–192). Experimentation is needed for comparing use-wear
patterns produced by different contact situations.
CHOPPERS
Choppers are pebbles, cobbles, or small rocks that have been modified
through flaking to create an edge. The sharp edge distinguishes a chopper
from a pecking stone or hammerstone. If there is no other modification,
the design is considered expedient; if there are finger grips or other mod-
ifications that make the tool comfortable to hold, the design is strategic.
Forceful chopper strokes focus impact along the narrow edge and chip
away material from both the contact surface and the tool’s edge. The re-
sulting flake scars and sharp impact fractures dull the chopper’s edges.
Inefficient edges can be resharpened through additional flake removal.
This wear-management strategy maintains tool efficiency and prolongs
tool use life. Like hammerstones and pecking stones, choppers have a
long temporal and wide spatial distribution that has not received enough
research to recognize any meaningful pattern.
CHISELS
Chisels are designed with an edge useful for gouging depressions into rela-
tively soft materials. This action requires more force and a slightly sharper
edge angle, and removes more material than scraping. Haury (1945:138)
describes the manufacturing process for a chisel whereby a thin pebble is
broken and the edge ground to the desired working angle. Unless some
other feature is manufactured to make the tool easier to hold or to modify
it for hafting, it is expediently designed. Chisel edges become damaged
by shallow impact fractures and chips that subsequently become abraded
and rounded through extensive use. Abrasion progressively dulls the edges
to the point that techniques such as flaking or grinding are required to re-
store efficiency. Impact fractures on the distal end of a chisel may be evi-
dence for the use of a hammerstone to generate a more forceful gouging
motion. Chisels are apparently rare in both archaeological and ethnographic
contexts in the U.S. Southwest. Additional archaeological research is needed

to better understand their spatial and temporal distributions. Similarly,
ethnographic references outside the U.S. Southwest should be explored to
expand the behavioral models for their use.
Di Peso et al. (1974:86) describe a large chisel recovered from Casas
Grandes that may have been used in stone working, although neither con-
text nor use wear supported such a conclusion. Stone chisels should be
recognizable by distinctive use-wear patterns: the proximal end should
have impact fractures from blows with a hammerstone or maul, and the
distal end or bit edge should have impact fractures and flake scares. Such
tools also might be called “wedges,” and it is not too difficult to visualize
their usefulness in quarrying activities. The caveat with this description is
that it is possible to visualize many activities, tool types, and resultant
use-wear patterns. It is important that empirical evidence be gathered to
support the identification of specific activities, tool types, and wear pat-
terns through experimentation, ethnographic analogy, and archaeological
context.
CRUSHERS AND FERGOLITHS
Crushers are tools worked against a netherstone or the ground to reduce

intermediate substances. The crushing induced with these tools is prima-
rily the result of the stone’s weight rather than force applied by the user.
They are generally broader and heavier than most pestles, although their
functions are similar. Crushers are either expediently or strategically de-
signed, ranging from barely altered cobbles to carefully crafted T-shaped
stones. The working edge is either a broad convex surface or, in some
cases, a flaked, chopper-like edge. Sometimes the broad sides have been
secondarily used in grinding activities. Well-used crushers have wear on
their handles from being gripped by the user.
These large stones were first classified as crushers by Di Peso (1951:
114, Figure 53) after one was found at Babocomari Village in southeastern
Arizona (Ferg 1991:1). Others have been found, most commonly in con-
texts dated to a.d. 1100–1150, at University Ruin and Hodges in Tucson,
and scattered elsewhere around southern Arizona, perhaps as far north as
central Arizona, but not in the Phoenix Basin (Ferg 1991:3). The term fer-
golith is commonly used now to refer to these items because of the special
interest afforded them by Alan Ferg, a researcher at the Arizona State Mu-
seum. Ferg (1991:3) has pondered their use after looking at nearly 60 of
them, and while originally he concluded that they were not functional (Ferg
1991), more recent use-wear analyses have convinced him of their use as
crushers (Alan Ferg, personal communication, 2001).
POTTERY ANVILS
Pottery anvils (Figure 6.1) are handheld tools that provide firm bases against
which paddles are used to bond the clay, thin the walls, and generally
guide the shaping of a pot (Rice 1987:140–143; Russell 1975:126–127, Plate
XVII; Shepard 1956:59). The percussive paddle strokes are cushioned by
the clay, reducing the damage that might otherwise impact the anvil’s
surface. Some pottery anvils have either a groove or handle to facilitate
use and are thus of strategic design. Expediently designed pottery anvils
are rocks or cobbles selected for their appropriate size and shape, and
might be hard to recognize. Additional research and experimentation are
needed to determine if some handstones, especially those with encircling
grooves, might be pottery anvils (see, for example, Fratt 1996a:8.46–8.50,
Figure 8.6; Haury 1976:281, Figure 14.10c).
The paddle-and-anvil technique of shaping pottery is distinct from
the scraping technique. Among stone tools, the difference is discernible
only by the presence of pottery anvils. It is not possible to distinguish
stones used to polish pots manufactured with one technique or the other.
The presence of pottery anvils at a settlement can be used to help under-
stand technological traditions associated with pottery production.

Historically, the paddle-and-anvil technique of pottery manufacture has
been documented among Southern Paiute, Yuman groups, Pima, and Pa-
gago (Spier 1933:107; Wendorf 1953a:165). Russell (1975:126–127, Plate xvii)
describes a Piman woman coiling a pot and then blending the coils with a
paddle and anvil. Ethnographic accounts of Puebloan (Dillingham and
Elliot 1992:10; Simpson 1953:75; Stanislawski 1978:217) and non-Puebloan
(Russell 1975:126–127; Spier 1933:107) groups document the fact that pot-
tery production was women’s work. The teaching frameworks employed
may have varied slightly, but generally, women taught their relatives and
perhaps their neighbors how to make pottery (Stanislawski 1978:219).
Their instructions included the proper selection of raw material, as well
as the proper selection and use of manufacturing tools. The traditional as-
sumption is that this was also true in the prehistoric past.
Prehistorically, there is evidence for the paddle-and-anvil technique
among the Hohokam and also Patayan groups, which as defined by Colton
finger groove
thickness groove depth groove width
width
length
6.1. Pottery anvil. The encircling groove has wear from being handheld. Both
surfaces were used, with one slightly more convex than the other. (Drafted by
Rob Ciaccio.) (Adams 2001a: Figure 7.)
included the Cohonina and possibly the Sinagua (Wendorf 1953a:172). In

the Flagstaff area, pottery anvils occur at Lizard Man Village (Kamp and
Whittaker 1990:110, 1999:119, Figure 72), at several settlements along U.S.
89 (Adams 2001a), and at Kinnikinnick (Connor 1943:377). Anvils occur
on Perry Mesa at Agua Fria sites that are interpreted as a blend of Ho-
hokam and Sinagua people (Fiero et al. 1980:75). They also have been re-
covered from settlements in the lower Tonto Basin (Adams 1999; Simon
and McCartney 1994:794; Watson 1997:422, Figure 10.11). At the Hohokam
type site, Snaketown, pottery anvils were recovered from the early exca-
vations (Gladwin et al. 1975: Plate XLV), but not from the reexcavations
(Haury 1976:197). However, a handstone illustrated by Haury (1976: Fig-
ure 14.10c) looks very similar to a pottery anvil. The time range of pottery
anvils at Snaketown is not identified. Hohokam settlements in the Tucson
Basin, such as Hodges Ruin (Kelly 1978:91–93), have pottery anvils that
date to the Rillito and Rincon phases, between a.d. 850 and 1150.
LITHIC ANVILS
Lithic anvils (Figure 6.2) serve as working surfaces in the production of

flaked tools (Haury 1976:278). They are usually of expedient design, al-
though some that are strategically designed have been manufactured to a
specific shape or altered on the bottom to enhance stability or adjust the
angle of the surface. A flat surface is the only requirement for a functional
anvil. Size varies and is more likely an attribute related to available raw
material than to a specific design. Anvils can be distinguished from lap-
stones primarily by use wear, but also by their position of use. Anvils rest
on the floor during use, whereas lapstones are handheld. They can be dis-
tinguished from other netherstones by their use-wear patterns. Anvils have
impact fractures and striations deep enough to be considered gouges that
result from supporting cores during flake removal. The force of the ham-
merstone pushes the core into the anvil each time a flake is detached.
Netherstones and lapstones generally have shallower impact fractures
and more uniformly shallow striations created by things that are shaped
on them.
Rowe (1995) experimented with bipolar flaking, nut cracking, and bow
drilling to examine differences in use wear and the size of the resulting
depressions, variously called pits, cupules, depressions, and basins. Distin-
guishing lithic anvils from nutting anvils may be the most difficult task,
especially among lightly used tools. The depressions in the experimental
nutting stones grew larger and rougher as work progressed, reaching a size
that corresponded to the size of the cracked nut (Rowe 1995:13). Some
6.2. Lithic anvil, expediently designed, with wear concentrated in the lower por-
tion of the surface. Note the distinctive impact fractures and striations that
damage the surface. (Drafted by Chip Colwell.) (Adams 1996: Figure 8.)
stray impact fractures were also visible around the depression. In con-
trast, the depressions on experimental lithic anvils were more irregular
and damaged. Labels such as “cupped stones” or “nutting stones” have
been used in the past, but it is perhaps more useful to classify them as
netherstones and attempt to determine their more-specific uses through
use-wear analysis.
Distinguishing well-used anvils from pebble mortars is possible
through a technological analysis emphasizing design and use wear. If the
small depression is a manufactured attribute rather than a worn attribute,
the tool should be classified as a pebble mortar. Manufactured basins are
usually more regular in outline than depressions worn through percus-
sion activities. Also, striations in the basin from either pestle use or man-
ufacture help identify a mortar. A problem with these distinctions may
arise if the mortar was used to crush salt or crack some other hard sub-
stance, leaving only impact fractures in the basin. Then the identification
becomes an educated guess.
Information on the distributions of lithic anvils through time and
across space is not as plentiful as it is for other types. The reasons for this
are varied, including a lack of attention to their classification as distinct
from netherstones in general, and a belief that they “have little or no
meaning in a taxonomic sense” (Haury 1976:278–279). Nevertheless, a tech-
nological perspective will impart the usefulness of recognizing and ana-
lyzing these tools in behavioral, rather than taxonomic, terms.
Chapter 7
Hafted Percussion Tools
I ncluded in the set of hafted tools are axes, mauls, picks, adzes, mattocks,
hoes, and tchamahias. Each forms a distinct subset, yet the boundaries
between them are fuzzy in certain aspects of design and use (Figure 7.1).
Following the lead of Wheeler (1980:254) and Di Peso et al. (1974), the
term head refers to the stone part of the tool. Other terms specific to hafted
percussion tools are illustrated in Figure 7.2. The main characteristic dis-
tinguishing the subsets of hafted percussion tools is the angle at which
the head is seated. Hoe, adze, and mattock heads are hafted with the work-
ing edge perpendicular to the handle to facilitate grubbing or pulling mo-
tions. Axe, maul, and pick heads are hafted so that the working end of the
head is parallel to the handle. The heads are set at various angles to facil-
itate chopping and pounding motions. Axe, maul, and pick heads are dis-
tinguished by the shape of the impacting edge. Axes have sharp bit edges,
picks have pointed bit edges, and mauls are blunt with no edge. The bound-
ary between axes and mauls becomes fuzzy because worn axes were sec-
ondarily used as mauls.
Methods used to attach handles are distinct, usually involving features
such as notches, grooves that completely encircle the head (full groove),
or grooves that only partially encircle the head (3/4 groove, 5/8 groove).
These hafting techniques are distinctive of technological traditions that
have significant time and space distributions, as discussed below. Within
each hafting subset are variations that have their own significance. The
following sections describe each head type separately, discuss the varia-
tions in hafting techniques, and summarize their distributions through
time and across space.
AXES
Axe heads are modified for hafting with notches, 3/4 grooves, or full
grooves. Secondary features associated with each technique serve to
160
7. H AFTED P ERCUSSION T OOLS 161
7.1. Diagram of hafted tool types expressed in terms of sets and subsets.
a
bit side groove poll side
poll end
bit edge
bit poll
poll face
bit face
b
poll side groove poll side
poll end poll end
poll poll
poll face
poll face
7.2. Terms used to describe the parts of hafted percussion tools: (a) full-groove
axe head; (b) full-groove maul head. (Adapted by Rob Ciaccio from an illustra-
tion by Michael Donaldson [Adams 1979: Figure 3].)
securely seat the handle, but they are not found on every axe head. These
features include ridges on one or both sides of a 3/4 groove, a smaller
groove positioned perpendicularly to the 3/4 groove on the ungrooved
edge (called a wedge groove), a ridge within a full groove (called a spiral
groove), and repositioned grooves. What is known about the handles asso-
ciated with the various grooving and notching techniques comes from
axes found in dry caves and cliff dwellings. See, for example, hafted axes
illustrated by Fewkes, Haury, Morris, and Rohn (Fewkes 1909: Figure 12,
1911: Figure 20; Haury 1934:118, Plate 71, 1945:132; Morris 1928:86–89,
Figures 5, 8, 1939: Plate 151; Rohn 1971:215, Figure 250). These hafting de-
sign differences appear to be related more to differing technological knowl-
edge about attaching handles than to functional variation.
Axe-manufacturing techniques can be discerned from unfinished axes
and the tools used in their manufacture. For example, Haury (1945:130,
Plate 43) explains that at Los Muertos, water-worn pebbles were chosen
for axe heads, and hammerstones were used to rough out the basic shapes.
The heads were finished with stones that smoothed at least the bit, if not
the entire head. Netherstones with a shallow groove the same width as
axe bits have been found at settlements along Tonto Creek (Figure 2.15) at
Foote Canyon, Higgins Flat, Tla kii, and Big Juniper House (Haury 1985a:
116, Figure 57a; Hough 1914:4; Rinaldo 1959:244; Swannack 1969:126). Sta-
tionary sandstone outcrops and boulders, and masonry building stones
are marked with the grooves from where axe bits were formed and sharp-
ened (Hough 1907:22; Judd 1954:119; Martin et al. 1952:37–38; Swannack
1969:126). Although axe heads are strategically designed, the results have
been described as ranging from crude to elaborate. Some axe heads appear
crude and rough because the impact fractures created during manufacture
were never smoothed. Others are elegantly polished to an intense sheen.
Hough (1907:22) speculated that the grooves on axes were left rough to
help secure the handle.
Axes that were secondarily used as mauls should be categorized as
axes because that is the purpose for which they were originally designed.
Hafted tools designed as mauls are distinguished from axes by having two
polls of relatively equal size and shape (Figure 7.2). Axes have a roundish
poll, and a bit that tapers to a sharpened bit edge (Figure 7.3). The bit of
an axe head is usually longer than the poll, but extensive use can wear the
bit until it is nearly the same size. However, there is still almost always a
discernible difference in size or shape between the bit and the poll of an
axe head. There are some discussions in the literature about bit length as a
culturally distinctive attribute (Di Peso et al. 1974:41) or a chronological
development in design (Kelly 1978:92). Before making such assertions,
consideration should be given to original bit length and amount of wear
on measured axe heads. For example, unused axe heads from sites in the
Tonto/Roosevelt basins range in size from 10 cm to 25 cm. Groove widths
range from 2 cm to 6 cm. What is left for measurement today on used axe
heads may be the result of decades of wear and edge maintenance.
poll 3/4 Groove bit bit

edge
bit
thickness edge
length
groove
groove width
groove
depth
bit length
width poll length
length
bit
edge
length
bit
edge
width
7.3. Axe head with 3/4-groove design. Note that the bit is longer and shaped to
an edge, and that the poll is rounded. (Drafted by Chip Colwell.) (Adams 1996:
Figure 10a.)
Techniques for maintaining axes can be recognized by evidence of re-

sharpening and regrooving. Resharpening prolongs axe use life, but tool
use life also can be extended without resharpening if they are secondarily
used as hammers or mauls (Kidder 1932:58, Woodbury 1954:27–28, 32–
33). Resharpening can be recognized by an asymmetrical bit profile and
striations that overlay the original polish, manufacture impact fractures,
and any deep damage resulting from use of the originally sharp edge (Fig-
ure 7.4). The wear and resharpening of bit edges can be described in terms
of relative sharpness and edge configuration. Edges that are still sharp
from original manufacture are sharp. Dull axes have no evidence of re-
sharpening and are no longer useful without resharpening. Those that
have been resharpened and remain sharp are resharpened-sharp. Edges
that were resharpened but are now dull are resharpened-dull. Worn-out
axes are have no potential for resharpening. This categorization of edge
configuration allows for comparisons of the extent of axe use.
Some axes are regrooved as a maintenance strategy to keep worn-out
axe heads functional by rebalancing the bit and poll ends (Adams 1994a:
131; Haury 1945:132; Woodbury 1954:34). This strategy is recognizable as
two grooves of the same configuration placed side-by-side. The new groove
is usually shallower than the original. Sometimes a new 3/4-groove is
placed so that the opposite narrow edge is grooved. Occasionally a 3/4
groove is made to function as a full groove by extending the original groove
onto the ungrooved edge. The extended groove is usually slightly narrower
and shallower than the original groove. Some full-groove heads were mod-
ified with a new 3/4 groove positioned either to the poll or bit side of the
original groove. Axes with two different groove configurations may be ev-
idence of two technological traditions coming in contact, as discussed in
more detail below (see also Adams 1994a).
Archaeological context has been used in the Four Corners area to sug-
gest that axe heads were commonly collected and stored for future use
(Larralde and Schlanger 1994:10). Similar behaviors may account for the
accumulations of axe heads with varying amounts of use at Turkey Creek
Pueblo in the Point of Pines area (Adams 1994a:126–130). The differing bit
lengths and edge configurations discussed above may be useful for track-
ing this curatorial behavior. Sites with higher frequencies of resharpened-
dull and worn-out edges may have been the final depository for axes no
longer considered worth the effort to maintain.
Archaeological context and the unusual use-wear patterns of some
axes have caused speculation about their use to dig clay (Haury 1945:
132). Experimental replication has also been used to accumulate evidence
for suggesting that some axes served as hoes used to grub bushes while
7.4. Profiles of bit edges in various stages of wear: (a) sharp and unused; (b) dull;
(c) resharpened–sharp; (d) resharpened–dull; (e) worn out. Note the angles on the bit
that result from resharpening the edge.
clearing agricultural fields (Mills 1993). The use-wear patterns on the ex-
perimental tools used to grub in the dirt are strikingly similar to those
found on some prehistoric axe bits. Thus, although stone axes may have
been primarily designed for chopping wood, that may not have been
their only use. Descriptions of experimental axe manufacture and use can
be found in Pond 1930, where the observations of an elderly non-native
stone worker were recorded. Morris (1939:137) describes hafting a prehis-
toric axe and using it to cut down a cottonwood tree 10 cm in diameter
and trimming the branches, a process that took about six minutes.
Axe Subtypes
Notched axe heads usually have two opposing notches positioned on each
narrow edge of the axe head. The addition of a third notch across the end
of the poll is uncommon, but probably served to help secure the handle
on the few examples that have been recovered (Adams 2001a; Etzkorn
1993:169; Woodbury 1954:37).
Axe heads with a 3/4 groove are most commonly hafted with a J-
shaped wooden handle (Haury 1945:132). The bend in the J fits over the
groove, and the long end hangs below the ungrooved edge. Thus, the un-
grooved edge of the axe head is closest to the user’s hand (Figure 7.5).
7.5. Three types of handles attached to axe heads: (a) full-groove axe with dou-
ble-wrapped handle; (b) 3/4-groove axe with wrapped handle—note the wedge
groove on the proximal edge; (c) 3/4-groove axe with a J-handle. (Adapted by
Rob Ciaccio from an illustration by Michael Donaldson [Adams 1979: Figure 2].)
The branch selected for the handle is about 3 cm in diameter and 25 to 45

cm long. Axe heads with a 3/4 groove most commonly have the groove
positioned across two faces and one edge (Figure 7.3). Those that have the
groove positioned across two edges and one face were hafted for use with
a different stroke and should be classified as adzes. A wedge groove was
added to some axe heads so that a piece of wood could be wedged between
the head and the loop in the wooden handle, thereby improving handle
stability.
Some 3/4-groove axes have ridges that border the groove on the poll,
bit, or both sides. For an example of one with ridges on both sides of the
groove, see Haury’s report on Tla kii (Haury 1985a:116, Figure 57). Axes
with a single ridge have been recovered from Rincon- and Tanque Verde–
phase contexts in the Gila Basin (Kelly 1978:92) and from Tortolita-phase
contexts in the Tucson Basin (Adams 2000a). The ridges may have served
to more securely hold the handle in the groove. There has been specula-
tion that ridges on some axes may be nothing more than remnants of a
larger bit that remain after repeated resharpening (Pond 1930:87; Roberts
1931:156). The ridges on 3/4-groove axes in the U.S. Southwest appear to
have been intentionally manufactured and not remnants of a once larger
bit. However, this could be tested with experimentation and a compara-
tive use-wear analysis of archaeological specimens.
Full-groove axes have handles that wrap around the head one or two
times so that both long ends of the stick are held in the user’s hand (Fewkes
1909:39, Figure 12, 1911:65, Plate 20; Kidder 1932:45, Figure 23; Morris
1919:177, Plate 34, 1939:136, Plate 151). Full-groove axe heads are com-
pletely encircled by a groove (Figure 7.2a). Occasional variations include
the spiral groove, a full groove with a ridge in the center to hold a doubly
wrapped handle. Axe heads with a spiral groove also have a diagonally
positioned groove that perhaps alters the striking motion (Kidder 1932:50,
Figures 25–30; Stubbs and Stallings 1953:105, Plate 20i; Wendorf 1953b:
73). This is another topic that would benefit from experimental and com-
parative research.
Double-bit axes were not as common in the U.S. Southwest as single-
bit axe heads. For this reason they are traditionally classified by their bit
configuration rather than by the groove configuration used to classify sin-
gle-bit axe heads. The concept of the double bit seems to crosscut techno-
logical traditions: there are examples of double-bit axe heads with notches,
3/4, and full grooves in the same time and space distributions as those
with single bits. Double-bit axe heads found in the Four Corners area are
hafted with either notches or full grooves. See, for example, those recov-
ered from Long House (Wheeler 1980:256, Figure 296) and Spruce Tree
House (Fewkes 1909: Figure 2). Three-quarter groove, double-bit axe heads
have been found in the Casas Grandes area (Di Peso et al. 1974:82), at
Galaz (Anyon and LeBlanc 1984:276, Figure 19.3J), in the Point of Pines
area (Adams 1994a:136), and in the Tonto/Roosevelt basins (Adams 1995:
65, 2000c; Simon and Rice 1996: Figure 9.19). Some found in the Tonto
Basin appear to have never been used or even hafted (Adams 1995:65,
2000c). Unused axe heads and axe-manufacturing tools found at Tonto
Basin sites may be evidence for the local manufacture of both double- and
single-bit axe heads. The largest concentration of double-bit axe heads (a
total of 10) was found at Los Muertos (Haury 1945:132, Plates 47, 48). All
have 3/4 grooves and are generally somewhat similar to the singe-bit axe
heads from the same site.
Flaked double-bit axes are considered “war clubs” by the Tewa (Jeançon
1923:18), and archaeologists like to speculate about the use of axes as
weapons. Hill (1982:111) describes Santa Clara “war clubs” as both “axe-
shaped” and “ball-shaped.” There is no specific description of the heads
or the use wear to help distinguish them from axe or maul heads. Cos-
grove and Cosgrove (1932) identify “clubheads” as distinct from axe and
maul heads at Swartz Ruin, but it is unclear how the identification was
made.
Ethnographic accounts of axe use by the Hopi and the Tohono O’odham
indicate that most historic stone axes were collected from prehistoric sites
and employed in the manufacture of metates or masonry blocks (Hough
1918:270–271; Russell 1975:110; Woodbury 1954:40–42). Axes were also
collected from prehistoric sites for use as components of ritual altars or
ceremonies at Hopi (Hough 1918:271; Stephen 1936:43, Woodbury 1954:
41). Because there are no eyewitness accounts of stone axe use in the U.S.
Southwest, there is no documentation about whether axe manufacture
and use were gender specific. However, ethnographic accounts of axe and
adze manufacture and use in other parts of the world indicate that these
are male-related activities (Gould 1977; Gould et al. 1971). Perhaps the
same was true in the prehistoric U.S. Southwest.
Hafting technology seems to have some differences in temporal and
spatial distribution, somewhat similar to those outlined for metate de-
signs. Haury (1976:291) developed an evolutionary scheme for 3/4-groove
axe heads based on those recovered from Snaketown, and he concluded
that axes arrived in southern Arizona prior to the Sweetwater phase (about
a.d. 550), probably from farther south in western Mexico (Haury 1976:291).
Woodbury (1954:29) noted the conflicting evidence from northern Mexico,

and that some researchers postulated that 3/4-groove axes came there
from the U.S. Southwest.
In southern Arizona, 3/4-groove axe heads continued to be the indige-
nous design throughout prehistory, although the consensus is that the
configuration of the bordering ridges changes through time. Those with
ridges on both sides of the groove predominate in Pioneer and early Colo-
nial contexts (Kelly 1978:92). A single ridge becomes common around a.d.
1000. By the Classic period (around a.d. 1100), 3/4-groove axe heads with-
out ridges and wedge grooves were common at Snaketown, Los Muertos,
Las Acequias (Haury 1945:130, 167, 1976:291), Casa Grande (Fewkes 1912:
123–124), and in the Tonto Basin (Adams 1995:264–265, 1999; Simon and
McCartney 1994:792–793; Simon and Rice 1996:564, Figure 9.16; Towner
1994:524, 528; Watson 1997:418–421, Figure 10.6). Recent research, however,
may shed some different light on this subject: in the Tucson Basin there is
evidence for the local manufacture of axes at Valencia Vieja during the
Tortolita phase (a.d. 450–700) (Adams 2000a). One well-used 3/4-groove
axe head from this site has a wedge groove but only one ridge, and one
nearly finished axe head has no ridges. Thus, it is probably time to reeval-
uate the changes in 3/4-groove axe-head design through time and across
space.
The common belief is that in east-central Arizona hafted axes were
not part of Mogollon technology until late—after a.d. 1000 (Wheat 1955:
123–124, during Mogollon V). Although, a 3/4-groove axe head with ridges
and a wedge groove typical of contemporary Hohokam technology was
found at Crooked Ridge in a Circle Prairie-phase (about a.d. 550) struc-
ture that also housed a few other tools of Hohokam technology (Adams
1994a:248; Wheat 1954: Figure 51a, 1955:123). Thus, axes were not unknown
to these early residents in the Point of Pines area. Axe heads lacking ridges
and wedge grooves were plentiful by the Reserve/Tularosa phase (around
a.d. 1100) among later Point of Pines area settlements (Adams 1994a:126–
128). The occasional presence of axe heads in earlier contexts and the fact
that they are plentiful by a.d. 1100 beg the question about axe-head cura-
tion as discussed by Larralde and Schlanger (1994). They proposed that
axe heads were highly valued and highly curated tools whose distribu-
tion in the archaeological record is probably skewed because of storage
and caching behaviors. Perhaps axes seem scarcer in early Mogollon con-
texts than they really were.
The 3/4-groove design was introduced north of the Mogollon Rim
during Pueblo I or early Pueblo II (after a.d. 900). Axes hafted by means
of a 3/4 quarter groove first occurred in Flagstaff-area Sinagua sites during
the late Pueblo II period (Woodbury 1954:30, based on Colton’s revisions).

McGregor (1941:173–174) noted that 3/4-groove axe heads were missing
from Winona-phase contexts, where he most expected to find them, but
that they were found in Angell-, Padre-, and Elden-phase contexts among
the Winona/Ridge Ruin sites. All of the axe heads recovered from Lizard
Man are 3/4-groove (Kamp and Whittaker 1999: Figure 71), as are those
found at Elden (Peter J. Pilles Jr., personal communication, 1999). Thus, by
a.d. 1000 or a little later, 3/4-groove axes were common in the Flagstaff
area. At Wupatki, most of the axe heads were 3/4-groove; however, a cou-
ple of full-groove and notched axes were recovered as well (Stanislawski
1963:109–110). Similarly, at Montezuma’s Castle most were 3/4-groove
axe heads, with a few full-groove heads (Stanislawski 1963:111).
The most commonly accepted hypothesis is that the full-groove axe
design originated in the northern Southwest (Haury 1945:131), perhaps as
early as a.d. 700, working its way south into the Mogollon area by the
mid- to late a.d. 1200s (Woodbury 1954:36). Gladwin disputes the north
as the center of origin because of full-groove axe heads found at early set-
tlements farther south (Woodbury 1954:46). Martin (1979:71) places the
earliest occurrences of full-groove axes in eastern Arizona and western
New Mexico at about a.d. 900. The full-groove axe heads found at Galaz
are of local material, but evidence is inconclusive about how early in the
archaeological record these show up, perhaps not prior to Classic Mimbres
(a.d. 1000–1150). All axes at Galaz were recovered from Classic-period
contexts, but this may be a result of sampling bias (Anyon and LeBlanc
1984:276). A few full-groove axe heads were found at Casas Grandes in
northern Mexico in contexts dating to the Viejo period (approximately a.d.
700–1060) (Di Peso et al. 1974).
If the full-groove technique developed in the north, the San Juan Basin
of southwestern Colorado and southeastern Utah is proposed as the most
likely location, even if only because it was better known than other areas
at the time Woodbury (1954:35) was writing. The full groove was viewed
by Woodbury (1954:35) as a development deriving from an earlier hafting
technique used with notched axe heads. Axe heads found at the Williford
sites in Colorado appear transitional, with both notches and a groove on
the same axe head (Woodbury 1954:35). Notched axe heads are found in
Basketmaker III (prior to a.d. 750) sites in the Four Corners area. Based
on the illustrated heads from Duckfoot, a late Pueblo I settlement, it ap-
pears most were notched, including one double-bitted head, but there
were also a few full-groove axe heads (Etzkorn 1993: Figure 4.7). At Alkali
Ridge, full-groove axes are more abundant than notched axes by Pueblo
III (Brew 1974:239). The same is true for axes at Pecos Pueblo (Kidder 1932:
45–54). Both notched- and full-groove axe heads occur at Mesa Verde sites,
with full-groove heads the most common by the time the mesa was aban-
doned, around a.d. 1275 (Hayes and Lancaster 1975:148–149; Rohn 1971:
212–214; Swannack 1969:130–133; Wheeler 1980:256–257). At Chaco Canyon
settlements, full-groove axes heads predominate, with 3/4-groove heads
recovered from a few contexts (Breternitz 1997:993).
From yet another perspective, Stubbs and Stallings (1953:104–105) hy-
pothesize that the full-groove axe originated in the upper Little Colorado
region and spread north and south from there, but they present no spe-
cific sites and dates to bolster their argument. They do note, however, that
full-groove axes arrived late in the upper Rio Grande region with the im-
migrants from Mesa Verde in the thirteenth century (Stubbs and Stallings
1953:105). A single full-groove axe head was found at NA 682 in Big Hawk
Valley (Smith 1952:119), and a few from Wupatki demonstrate how rare
they are in the Flagstaff area at any time. Full-groove axe heads are rarely
found at Hohokam settlements and are usually considered intrusive when
they are. It seems plausible to consider the 3/4-groove techniques as dis-
tinctive of Hohokam and Mogollon technology, and full-groove techniques
distinctive of Anasazi technology. The ridges bordering grooves are not
attributes of Mogollon technology.
Full-groove and 3/4-groove hafting technologies co-occurred in sev-
eral places: Swartz (Cosgrove and Cosgrove 1932:41), Galaz (Anyon and
LeBlanc 1984:276), several Point of Pines area settlements (Adams 1994a:
124–137), Hooper Ranch Pueblo (Martin et al. 1961:102), Carter Ranch
Pueblo (Rinaldo 1964:74), and Higgins Flat Pueblo (Rinaldo 1956:87–89).
This meeting of two different technological traditions occurred in the Point
of Pines area, especially at Turkey Creek Pueblo (Adams 1994a:126–131),
sometime after a.d. 1100 and continued with the movement of Anasazi
populations (Maverick Mountain phase, Haury 1958; Lindsay 1987) into
the area around a.d. 1265. Some tools were redesigned from one hafting
technique to the other. These redesigned tools may reflect one behavioral
reaction to the commingling of different groups with differing technological
knowledge (Adams 1994a). A few scavenged or gifted axe heads were
modified. If they did not possess the “correct” groove orientation for
“proper” handle attachment, they were regrooved. Some 3/4-groove heads
were made into full-groove heads, and some full-groove heads were mod-
ified with a new 3/4 groove. Judd (1959:141) recognized regrooved axes
of both types at Pueblo del Arroyo. Thus, this is a technological reaction
that probably occurred more often than we have realized.
The spiral groove appears to have been a Pueblo IV modification that
was perhaps manufactured exclusively at New Mexico pueblos such as
Pecos and Pindi (Stubbs and Stallings 1953:105, Plate 20i; Wendorf 1953b:
73). The finest axe heads are commonly manufactured from an unusual
material called sillimanite or fibrolite that comes from a source in the San-
gre de Cristo Mountains (Kidder 1932:50; Wendorf 1953b:75; Woodbury
1954:37). These specialized axe heads have been recovered from proto-
historic and historic deposits at the pueblos of Pecos (Kidder 1932:50, Fig-
ure 30), Picuris (Dick 1965b:184), Paa-ko (Lambert 1954:126, Plate XXV),
and Tonque in New Mexico (Barnett 1969:115), and even farther away in
Arizona at the Hopi pueblos of Walpi and Awatovi (Adams 1979:37; Wood-
bury 1954:37).
Axe-hafting technology seems to be one of the best examples of a pre-
historic technological tradition related to ground stone working. There are
clear differences in the techniques used to attach handles, and these tech-
niques first appear in different places, among different cultural groups.
There is evidence that these techniques moved with people as they moved
across the landscape, and most interestingly, there is evidence of reactions
to the meeting of different technological traditions.
MAULS
Maul heads are large rocks grooved for hafting wooden handles in much
the same manner that axes are grooved. See, for example, the illustrations
of full-groove mauls with double-wrapped wooden handles recovered
from Aztec (Morris 1919: Figure 9) and Mug House (Rohn 1971: Figure
245). Maul heads can be distinguished from similarly grooved axes by the
lack of a bit (Figure 7.2b). The hafting groove of a maul head essentially
separates two equally sized and shaped polls (Woodbury 1954:40). They
are more often of expedient design than axe heads.
Mauls can function in any activity that requires impact force, such as
pounding stakes in the ground, driving wedges through wood, procuring
or early-stage processing of some food resources, and even in killing small
animals or people (Di Peso et al. 1974:154; Woodbury 1954:48–49). Use-wear
damage includes impact fractures and chips. As discussed more fully above,
hammerstones and mauls function similarly and have similar use-wear
patterns, but differ in that hammerstones are handheld tools and so lack
the groove for hafting.
The section on hammerstones should also be referred to for a discus-
sion of percussion tools from various Mesa Verde settlements that were
classified as either “hammer heads” or “mauls” on the basis of size
(Wheeler 1980:245). Size may have some functional distinction similar to
our modern distinctions between hammers and mallets of different weights.
However, it is recommended that size distinctions be evaluated only for

tools that have the same material source: large in one area may be medium
or small in another because of available raw material. Size categories are
therefore difficult to compare across different assemblages. Experimenta-
tion and use-wear analysis may at some point provide better criteria for
functional classifications similar to those recognized by Hayden and Nel-
son (1981) among the Guatemalan Maya.
Woodbury (1954:44–47) subdivides the “grooved hammer” category
by shape and other morphological attributes, such as pits and ridges along
the grooves that he calls “lipped full-grooved hammers.” Such divisions
perhaps confuse description with classification. The technique preferred
here is to classify maul heads by groove configuration and emphasize de-
scriptive details to define typical or unusual members of the subsets.
From a technological standpoint, a distinction should be made be-
tween tools originally designed as axes and secondarily used as mauls,
and those designed as mauls. The fact that secondarily used axes were
classified as mauls in the past makes it difficult to sort out the temporal
and geographic distributions of the two tool types. Woodbury (1954:48)
postulates that 3/4-groove mauls co-occurred with 3/4-groove axes, and
that both spread throughout the Southwest in the same manner. Stubbs
and Stallings (1953:104) hypothesize that full-groove mauls occurred ear-
lier and served as the prototype for full-groove axes. Obviously, more re-
search is needed to reconstruct the covariance in temporal and spatial dis-
tributions of variously grooved and notched maul and axe heads.
Archaeological Distributions
Mauls with various groove configurations occur in Mogollon assemblages

from the earliest to the latest periods (Haury 1985a:116–117, 1985b:235–
236, 243; Wheat 1955:122). Wheat (1955:122) identifies a groove technique
that encircles seven-eighths of the tool’s circumference. He calls this tech-
nique most typical of Mogollon groups, but he does not discuss whether
it is common to all branches of the Mogollon or is specific to the Point of
Pines area, where he first recognized it (Wheat 1954:127, 140). In Martin’s
Mogollon reports, only one 7/8-groove maul head, from a Tularosa-phase
site, is described (Martin et al. 1957:68). Nothing similar is illustrated by
Woodbury (1954) for Anasazi technology or by Di Peso et al. (1974), Glad-
win et al. (1975), or Haury (1976) for technology in southern Arizona/
northern Mexico. Three-quarter groove mauls were recovered in early
contexts at Bear Ruin and may have been a part of Hohokam technology
as early as the Colonial period at Snaketown. Roberts (1931:155) contends

that 3/4-groove mauls were an early Pueblo development in the Little Col-
orado River. They remained common in historic Hopi assemblages until
replaced by metal tools (J. L. Adams 1979:115; E. C. Adams 1982:117; Wood-
bury [1954:43–49] calls them “grooved hammers”).
Full-groove mauls are a Basketmaker III or perhaps earlier develop-
ment in the Four Corners area and are probably more typical of Anasazi
technology than of any other (Woodbury 1954:48). Mauls found in the
Four Corners region at Mesa Verde and Alkali Ridge settlements were
hafted by means of either notches or full grooves (Brew 1974:237; Hayes
and Lancaster 1975:148–149; Rohn 1971:210, Figure 245; Swannack 1969:
126–127; Wheeler 1980:245). Full-groove mauls were found in early sites
in New Mexico, as exemplified by those at Shabik’eshchee, a Basketmaker
III village in Chaco Canyon (Roberts 1929:134–135). Kidder (1932:54–55)
sorts some mauls recovered from Pecos Pueblo into two smaller subsets:
pemmican pounders and grooved club heads, distinguished primarily by
size. Kidder uses ethnographic analogy to identify pemmican pounders,
and archaeological context (one found in a warrior’s grave) to identify
club heads.
Woodbury (1954:40–42, 48–49) discusses the possible use of mauls
and axes as weapons of war and cites Hopi oral traditions and some ethno-
graphic identifications as evidence. Rohn (1971:248–249, Figure 298) dis-
tinguishes club heads from those of axes and mauls on the basis of a more
brittle material used to manufacture club heads. The tools he calls club
heads have flake scars and impact fractures, and some are stained with
pigment or decorated with incised symbols (Rohn 1971:248–249). Trying
to distinguish war and hunting weapons from other hafted percussion
tools is probably futile. The same tools could have served in any or all
tasks that require impact force, including hitting people or animals. The
presence of decoration may be a more reliable indication of tools that had
ritualized uses, such as those associated with hunting or warfare.
PICKS
Picks are axe-like tools both in the way they are hafted and in the general
motions used to swing them. They are generally expediently designed
with no additional effort expended beyond the manufacture of the bit and
the groove or notches for hafting. The most distinctive attribute is the
configuration of the bit. A pick bit tapers to a point that focuses the im-
pact and damages the contact surface with impact fractures and chips.
Picks are most commonly associated with quarrying or mining activities,

in which contact surfaces are much harder than those contacted by axes
that chop wood.
Picks have been found in association with mines at Zacatecas and the
salt mines in the Verde Valley (Di Peso et al. 1974:96). Hayes et al. (1981:
107) also describe picks found at Gran Quivera, but there is no specula-
tion about their use. They also have been found in Pueblo II contexts at
Alkali Ridge in the Four Corners area (Brew 1974:235, Figure 51c). Picks
appear to have been less common than axes and mauls, and there has not
yet been an extensive analysis of their spatial or temporal distributions.
Some handheld tools have been identified as picks, but this is an un-
clearly defined category. The handheld pick illustrated by Haury (1945:
Plate 32) looks like a pestle that has been redesigned with a chopping
edge, perhaps for a secondary use. There is an example of this same tool
configuration from an early context at Crooked Ridge Village (Adams 1994a:
122–123; Wheat 1954:117, 1955:120). Five of the 11 pestle-like tools cached
in a pithouse were probably used as digging tools (Wheat 1954:117, 1955:
120). See the discussion of pestles in Chapter 5 for additional description
of these tools. Perhaps experimentation and comparative use-wear analy-
ses can help identify distinctive attributes for the classification of hand-
held picks.
ADZES
Adze heads are similar to axe heads in terms of their cutting edges, but
differ in the orientation of the groove around the head, and the orienta-
tion of the blade to the handle (Haury 1945:133). The blade on an adze sits
perpendicular, or at an acute angle, to the handle rather than parallel to it
as an axe blade does. Thus, if hafted with a 3/4 groove, the groove is
across two edges and one side. An adze is also distinct from a plane, pri-
marily in the application of force and configuration of the edge. An adze
has a sturdier, more steeply angled edge that gouges and removes larger
quantities of material than a plane. A plane, with its thinner edge, scrapes
or slices through material. (Planes are described in Chapter 9, “Perforating,
Cutting, and Scraping Tools.”) Adzes are strategically designed only if
they are shaped with more than just the notches or grooves required for
handle attachment, and the edge created by flaking or grinding.
The largest single collection of adze heads was found at Los Muertos
(Haury 1945:133). They range in size from 11 to 15 cm, and in weight from
less than 140 to 2,000 g. Haury speculates that the adze may have origi-
nated among Puebloan groups, but concluded that there was not enough
evidence to discuss their spatial and temporal distributions (Haury 1945:
133).
MATTOCKS
Although not a commonly defined tool type, a mattock head is defined by

Di Peso et al. (1974:100) as more carefully manufactured than a pick head,
but not as carefully as an axe head, and with an edge more like an axe
than a pick. The mattock head is hafted by means of a groove across both
edges and one broad side so that the edge is perpendicular to the handle,
like that of an adze. The difference between an adze and a mattock seems
to be in the relatively larger size and greater weight of a mattock head.
Their uses were perhaps different as well, in that an adze is a woodwork-
ing tool, and a mattock is a quarrying or grubbing tool. The difference be-
tween a pick and a mattock is the shape of the bit: a pick has pointed bit,
and a mattock has an edge. These design attributes are perhaps more use-
ful for making classification-level distinctions than the amount of care
used in manufacture. Mattock heads are of expedient design unless the tool
has been shaped beyond the creation of the hafting features and blade.
There has not yet been enough research on mattock distributions to eval-
uate their development through time and across space.
HOES
Two different tool designs have been identified as hoe blades. One design
is a thin tabular tool that may or may not be modified for handling or
handle attachment. The other design is a thick, more robust tool that has
notches or is partially grooved for handle attachment (Figure 7.6). Both
are thought to have been used in the soil for tilling, weeding, or creating
water-diversion ditches. The thin blades are sometimes shaped through
flaking. They seem more common in southern and central Arizona than in
other parts of the U.S. Southwest in contexts that date as early at the Colo-
nial period, and are most abundant in Classic-period contexts (Haury 1945:
137). Hoes have been found at Snaketown (Haury 1976:285), Casas Grandes
(Di Peso et al. 1974:360), Los Muertos, other Salt-Gila region sites (Haury
1945:134–136), and in the Flagstaff area (McGregor 1941:171, Figure 59D;
Peter J. Pilles Jr., personal communication, 2000).
Elsewhere hoe blades are similar to notched axe heads, with a slightly
different positioning of the handle to accommodate a different stroke. They
bit
edge
notch
length
thickness
notch
width
notch
depth
width
7.6. Hoe with notches for hafting. (Drafted by Chip Colwell.) (Adams 1996:
Figure 16.)
have been found at Anasazi and ancestral Hopi settlements in the Ho-
mol’ovi area (Woodbury 1954:166–170). On the Colorado Plateau, some
manos have been redesigned as hoe blades with the addition of notches
on the edges to accommodate a haft (Seibert 1987). Woodbury (1954:169)
made the observation that as designed, the hafted hoes would not be strong
enough to work clayey soils but would be better suited for working sandy
soils—a hypothesis that could use some experimental testing.
TCHAMAHIAS
Items found primarily in the Four Corners area and labeled with the Kerean
term tchamahia or “tcamahia” (Wheeler 1980:284) are thought to have been
used as hoes (Judd 1959:138–139). They are long, thin blade tools that may
or may not have been hafted. Some tchamahias have projections that might
have been useful for hafting a handle, and Morris (1939) found remnants
of wood on one in the La Plata District. Wheeler (1980:285, Figures 329–
331) describes and illustrates tchamahia designs recovered from Long
House. Some are modified for hafting, a few have a hafting attribute he
calls a “tang,” and others show no evidence of either hafting or use. Judd
(1959:138–139) considered the tchamahia with a handle for hafting from
Pueblo del Arroyo as reworked probably for a secondary use, even though
he considers their original handheld use to be that of hoes, at least until
Pueblo III.
Tchamahias on Hopi and Zuni altars are usually made from hornstone
or black slate (Parsons 1939:333). Stephen (1936:597, Figure 332, Plate XVII)
illustrates a “chimahia” among altar paraphernalia of the Hopi. Parsons
(1939:194, 333) notes that the term is also used to refer to other altar stones
representing beings. The term “chamahia” is used by Laguna to refer to
altar stones that represent mountains (Parsons 1939:194). Thus, histori-
cally there has developed a confusion of ethnographic terms and archaeo-
logical types.
The tchamahias that archaeologists recognize are most often made
from an unusual material called “hornfels” or “hornstone” that has two
known sources in the Four Corners region: one west of the La Plata Valley
south of Mesa Verde, and the other in Arizona south of the San Juan River
(Morris 1939:139; Rohn 1971:247–248; Woodbury 1954:167). They are most
common among assemblages in the San Juan/Mesa Verde area dating to
Pueblo II and III contexts. One was found slightly farther north at Alkali
Ridge in a Pueblo II context (Brew 1974:241), and a few have been found
as far south as Awatovi in later, Pueblo IV and V contexts (Woodbury
1954:165–170). There is speculation that the tchamahia ceased to function
as a hoe at some point in the prehistoric past and instead became a sym-
bol important enough to have been included in certain ceremonial altars
of prehistoric and historic Puebloan groups (Wheeler 1980:285; Woodbury
1954:166–170). Rohn (1971:248) discusses the various interpretations of
tchamahia use, ranging from hoes to skinning knives, and concludes that
none of these uses seem particularly plausible.
Chapter 8
Spinning Tools
T he items included in this category are parts of larger tools. They each
function in different ways to facilitate the rotation of a spindle. Such
composite tools have utility in a variety of activities. Those described here
were perhaps used in processing fibers, drilling holes in various types of
items, and starting fires.
FIRE-DRILL HEARTHS
A fire-drill hearth is the bottom part of a fire-starting kit. Small, sometimes

notched cupules serve to contain the spark created by friction from a ro-
tating fire drill (Figure 8.1). A hearth may have either single or multiple
cupules which are generally no more than 1 cm deep and not much larger
than the diameter of the drill—approximately 1 cm. Grooves cut into some
cupules allow tiny pieces of flammable material to come into contact with
the spark, which can then be fanned into a flame. Neither the spark nor
the friction generated by the spinning shaft blacken the cupule. Only when
the spark ignites a piece of fiber does burning actually happen within the
cupule. Usually, before the flame starts, the smoking fiber is poured out
onto flammable material so that the final stages of starting a fire happen
outside the hearth.
Ethnographic accounts of Puebloan and non-Puebloan fire-starting de-

scribe men as the tool users (Hough 1890:531–532, 1915:164; Liljeblad and
Fowler 1986:421, Figure 5; Spier 1933:78; Underhill 1946:89). Simpson
(1953:82) describes Hopi fire-making techniques using a sandstone tablet
and a cottonwood drill, with juniper bark serving as the combustible ma-
terial. Fewkes (1920:595) describes the very ritualized Hopi New Fire Cer-
emony in which multiple fire-starters use both wood and stone hearths
180
8. S PINNING T OOLS 181
cupule width
dorsal side margin
thickness
cupule depth
ventral side
width
length
8.1. Fire-drill hearth with a cupule where a drill was twirled to create a spark.
(Drafted by Rob Ciaccio.) (Adams 2000c: Figure 16.)
with corn pollen in the cupules. The drills are rotated between the palms,
and the hands push downward along the drill shaft. A relief driller takes
over about every 15 seconds to keep the drill spinning rapidly. Such a
technique used on a stone hearth produced a spark in 1 minute 20 sec-
onds. While other participants sing, the spark is gently blown onto cedar
bark and fanned into flames. Fewkes (1920:606) goes on to explain the Hopi
view that the drill is male and the hearth female, and thus the fire repre-
sents the life generated by the two together. A very stylized stone hearth
is illustrated by Fewkes (1920: Plate 8).
The fire-drill hearths described in ethnographic contexts are most of-
ten made from wood (Hough 1890:531–532; Russell 1975:102–103, Figure
18a; Spier 1933:129), and we know that wooden ones were used prehis-
torically because of those found in cliff dwellings at Mesa Verde (Rohn
1971:234, Figure 276). However, most often only stone hearths survive in
archaeological contexts such as at Casas Grandes (Di Peso et al. 1974:205),
at settlements near Flagstaff (Adams 2001a), at several Point of Pines area
sites (Adams 1994a:137–138), and at sites in northeastern Arizona (Kidder
and Guernsey 1919:120, Plate 50). Pumice and vesicular basalt are com-
mon choices for fire-drill hearth material. Too few have been identified to
get a clear sense of their temporal and spatial distributions. Those few
that have been identified come from contexts as early as a.d. 800, and
they obviously continue in historic use.
SPINDLE BASES
Spindle bases are the netherstones used with spindle whorls (Barnett 1969:
123, called “drill bases”; Di Peso et al. 1974:138). They provide a firm sur-
face, and sometimes a depression that confines but does not restrict the
movement of rotating spindles. The difference between a fire-drill hearth
and a spindle base is the size of the cupule. The important feature for a
fire-drill hearth is friction. The spindle must rub against the walls of the
cupule to generate the frictional spark. This is not a desirable feature for
spinning fibers. Thus the cupule in a spindle base is usually larger with
more-sloped sides.
Spindle bases can be of either strategic or expedient design. Those
that are expediently designed have no modification other than the creation
of a basin to confine the spindle. Some possible spindle bases have been
found that take advantage of a natural vesicle as the basin (see, for exam-
ple, Adams 2001a). Use-wear analysis is then the only way to identify the
use of such naturally formed basins. The rotation of a spindle causes cir-
cular striations within the basin. The resiliency of the wood rounds the
stone’s grains and can produce a low-luster sheen with extensive use. Curvi-
linear striations form if loose grains get caught between the stone and the
spinning shaft.
Possible spindle bases have been described at sites in the Roosevelt
Basin (Simon and McCartney 1994: Figure 26.9a; Simon and Rice 1996:
562); in the Flagstaff area (Adams 2001a), in New Mexico (Barnett 1969:
8. S PINNING T OOLS 183
123), and at Casas Grandes (Di Peso et al. 1974:138). Some of the items
classified as pitted or cupped stones could also have served as spindle
bases and not been recognized as such ( see the Chapter 5 discussion of
pitted and cupped stones). Further experimental and comparative research
is needed to recognize the spatial and temporal distributions of spindle
bases. If they can become confidently recognized, valuable data can be
recorded about the extent of fiber production.
WHORLS
Whorls are thin disks perforated with a centrally located hole that fits over
a spindle shaft. They are the flywheels that maintain the spinning momen-
tum and are strategically designed only if there is more modification than
just drilling the hole. Whorls range in size from 4 to 12 cm in diameter,
and .3 to 2 cm thick. Stone is but one of many material types used to make
whorls. Russell (1975:148–149) footnotes a description by Lieutenant Emory
of a Piman woman spinning cotton using a wooden whorl. There is evi-
dence of wood whorls from cliff dwellings at Mesa Verde (Rohn 1971:228,
Figure 266). Both wood and horn whorls are among the remains recovered
from the historic Hopi village of Walpi (Adams 1979:17, 52, Plates 4, 12).
Molded clay and reworked sherds have also been used to make whorls.
Wilcox (1987:145–162) and Greenwald (1988:150) discuss perforated stone
disks and the possibility that differences in whorl sizes relate to the types
of fibers processed. However, Teague (1998:45) questions the direct rela-
tionship between whorl size and the types of fibers spun.
The spindle/whorl configuration also may have been used by jewelry
makers in hand drills, or even pump drills, as has been illustrated for
contemporary jewelry makers (Judd 1954: Plate 20; Ladd 1979: Figure 5;
Strong 1979: Figure 6). Hoebel (1979: Figure 3) shows a close-up of a pump
drill with a wooden whorl. Whorls on short spindles were used as toy
tops by the Hopi (Culin 1975:743). It seems unwise at this point to infer
the precise use of a specific stone whorl other than as a flywheel. Further-
more, it must be recognized that stone whorls alone do not adequately re-
flect the presence of a specific activity such as fiber production, jewelry
manufacturing, or play.
Chapter 9
Perforating, Cutting, and

Scraping Tools
T he awls, reamers, saws, files, planes, and other tabular tools included
in this set are generally made to remove material from contacting sur-
faces by means of honed edges or points. In order to remove material from
the contact surfaces, they rely on: (1) pressure, in contrast to percussion
tools, which use impact forces; and (2) edges or points, in contrast to
abraders, which rely on surfaces. The set of perforating, cutting, and
scraping tools has a fuzzy boundary with flaked tools such as scrapers
and knives.
Reamers and awls are perforating tools easily distinguished from the
rest because they serve to create or enlarge holes. Saws and files can be
differentiated from each other through careful definition of edge configu-
ration. They can be distinguished from planes by the orientation of use
wear to their edges, and sometimes by the handles that are created, espe-
cially for planes. The subset of tabular tools subsumes a highly variable
group, including items labeled “agave knives” and “tabular hoes.” More
research is needed to better understand the uses of all the tool types in
this set. Experimental research would help recognize the purposes of dif-
ferent designs, the evidence for contact with various types of surfaces,
and the distinct strokes associated with each tool type. There is an inter-
esting lack of ethnographic description for modeling the use of these
tools. As discussed in Chapter 3, much is missing from the ethnographic
record because many stone tools were replaced by metal tools before ethno-
graphers could document their uses.
AWLS
Stone awls are small conical or cylindrical pieces of stone sharpened to a

point (Figure 9.1). Most stone awls are of strategic design, having been
184
9. P ERFORATING , C UTTING , AND S CRAPING T OOLS 185
carefully worked to a specific shape beyond manufacture of the point.

The materials chosen for awls are usually supra-fine or cryptocrystalline.
Some abraders with U-shaped grooves have use wear consistent with the
manufacture of stone awls (Adams 2000c) (Figure 4.5). The long, deep
striations that are visible on the awl without the aid of magnification are
usually remnants from the manufacture of the awl point. Macroscopically
visible use-wear patterns include the smoothing of any edges or projec-
tions that punctured the contact surface. Microscopically, use-wear patterns
are visible as the rounded edges of naturally angular or rough grains;
short, shallow striations; and perhaps a sheen. Short, shallow striations
are created through contact with minute particles in the perforated sur-
face that are at least as durable as the stone from which the awl was man-
ufactured. If use was extensive enough and the penetrated surfaces soft
enough, a sheen will have formed on the parts of the awl that contacted
the surfaces.
Stone awls may have been used in much the same way as bone or
wooden awls: for stringing corn cobs, for poking holes in leather or bas-
ketry, or for use in weaving textiles (Adams 1980:3–8, 2000c). It was noted
that among the pointed tools from sites in the Tonto Basin, some were
either unused awls or perhaps hair or clothing pins (Adams 2000c) (Fig-
ure 9.1c). Others have use-wear along one side from contact against pli-
able surfaces and might have been used similarly to bone weaving tools.
Experimental research is needed to be able to recognize the use-wear pat-
terns that would develop from use in hair, clothing, or weaving.
REAMERS
Reamers are either cylindrically shaped or have cylindrical projections used

in a rotary motion to shape or enlarge holes in other items. Most are of ex-
pedient design, although texture is an important attribute for the selected
rock. Granular material abrades and enlarges the hole, whereas smooth
material polishes the hole. Awls and reamers can be distinguished by the
direction of the striations. Those on awls are parallel to the long axis of
the tool. The striations on reamers run concentric to the long axis.
Reamers are most often associated with shell working in archaeologi-
cal contexts. Bracelets and rings are shaped with them (Haury 1976:284;
Simon and Rice 1996:573). Those used to shape bracelets are thicker in cir-
cumference (3–5 cm) than those used to shape rings (about .5–2 cm). Haury
(1976:284) notes that at Snaketown, most reamers came from Sacaton-phase
deposits. Additional research is needed to confirm the temporal and spa-
tial distributions of reamers.
width
thickness
length
a b c
9.1. Stone perforating tools: (a) awl; (b) weaving tool; (c) pin. (Adams 1996:
Figure 9.)
SAWS AND FILES

Saws and files have long edges and are generally used with cutting or
slicing motions. Cutting motions create striations parallel to the edges and
slicing motions create striations diagonal to the edges. Saws and files are
distinguishable from each other by edge configuration and performance
(see for example Kidder 1932:82–83, Figure 58). Saws have serrated edges
that rip and slice through material, whereas files have straight edges that
abrade through material. These tools are rarely hafted, although some have
been modified for holding – called “backed” or “backing.” Backed tools
are modified either intentionally through abrasion to smooth sharp edges
or through use-wear caused by a soft material wrapped along the edge to
cushion the grip.
Clarification
If the above distinction between saws and files is to be maintained, clari-
fications need to be made in light of how these terms have been applied
in past descriptions. Among the ground stone tools from Los Muertos,
Haury (1945:138, Figure 82) classifies a tabular tool with a serrated edge
as a saw. Rinaldo (1956:101–103, Figure 54a–c, i–k) defines saws as “chipped
cutting” tools with serrated edges. Both of these definitions fit with that
given above. However, different definitions have been applied in differ-
ent contexts. The term “saw” was also used by Haury (1976:284–285) to
label tools from Snaketown that have smooth, ground edges marked with
abrasive striations, probably from cutting soft stone. He suggests that they
were used to manufacture palettes and illustrates their postulated use to
create grooves on opposing surfaces of tabular material (Haury 1976: Fig-
ure 14.15). The concomitant temporal and spatial distributions of palettes
and phyllite saws among pre-Classic Snaketown contexts seems more
than coincidental to Haury (1976:284–285). Saws with smooth cutting edges
have also been described from assemblages at Long House and Big Juniper
House in Mesa Verde (Swannack 1969:124, Figure 108; Wheeler 1980:252–
253, Figure 290), Swarts Ruin (Cosgrove and Cosgrove 1932:46), Higgins
Flat Pueblo (Rinaldo 1956:93, Figure 50h–j), and Pueblo Bonito (Judd 1954:
125–125, Figure 35). Tabular tools with serrated and smooth edges from
the Rincon- and Rillito-phase deposits at Hodges Ruin in the Tucson Basin
were both identified as saws (Kelly 1978:87–88, Figure 6.3). A use-wear
analysis of these tools would help clarify whether the smooth edges started
out serrated and were worn smooth, or if they were designed smooth.
Kidder (1932:82–83) notes that tabular tools with smooth edges from
Pecos Pueblo could have been used to score stone but also that an experi-
mental “file” (his term for these tools) quickly cut into dry wood. Other
experiments illustrate that the smooth edges were more useful for cutting
bone and wood than for cutting stone (Cosgrove and Cosgrove 1932:46;
Wheeler 1980:253). Judd (1954: 123, Figure 34) uses the term “file” to de-
scribe a completely different tool configuration from Pueblo Bonito that is
more similar to modern metal files than any files described in other reports.
The observations made so far about the various uses of saws and files
provide baseline hypotheses that would benefit from structured experi-
mental testing. Obviously, they have a broad spatial and temporal distri-
bution, but additional research is needed to determine if there are meaning-
ful variations. We also need to consistently sort these tools into the proper
categories. Files should have smooth edges, and saws should have ser-
rated edges. The distinction is important from a technological standpoint
because of the different techniques involved in edge manufacture and be-
cause of the different performance characteristics of the finished edges.
PLANES
Planes are tools with angled working edges employed to scrape material
away from pliable surfaces, such as when shaping wood or defleshing
items such as yucca or agave leafs for fiber extraction. They are generally
strategically designed and are either backed or have handles to allow the
user to exert pressure behind the edge (Figure 9.2). Use wear is recogniz-
able as striations oriented perpendicularly to the edge. Tiny flake scars
may be removed from the edge depending on edge angle, the amount of
pressure exerted during use, and the resiliency of the contact surface. After
extensive use, the scratches and flake scars become rounded, and the edge
loses its sharpness. Sharpness is restorable through both flaking and grind-
ing, although grinding seems to be the more commonly chosen method.
Planes, chisels, and abraders are all designed to remove material from
pliable surfaces, but they are distinguishable by the configuration of their
working surfaces and by their performance characteristics. Planes and chis-
els both function along sharp edges. The difference is that a plane slices
into a surface through pressure, whereas a chisel gouges through impact
force. The damage to the working edge of a plane is from abrasion. The
working edge of a chisel shows more damage in the form of impact frac-
tures and large flake scars than found on the edge of a plane. Abrasive ac-
tion is part of how both planes and abraders shape or alter the configura-
tion of whatever surface they are worked against. The difference is that a
plane cuts into the surface with an edge to remove slices of material, and an
abrader uses a flat surface to grind away material. Experimental research
9.2. Plane with handles that have hand wear on the proximal and distal edges.
Plane edge ground to shape with striations parallel to the edge. Use-wear dam-
age is perpendicular to the edge. (Adams 2000d: Figure 7.6.)
needs to be conducted similar to that of Bernard-Shaw (1990), who stripped

agave leaves with tabular tools. Comparisons with experimental tools used
to plane wood would help recognize different use-wear patterns on pre-
historic tools. Additional research is also needed to understand the tem-
poral and geographic distributions of planes, and how those compare to
distributions of abraders and chisels. It may be possible to recognize tool-
kits used in woodworking or fiber extraction through such research.
TABULAR TOOLS
The term tabular tool is used to classify thin, tabular pieces of stone of
varying sizes, with one or more edges used in cutting, scraping, slicing, or
chopping motions. Some tools in this set are manufactured from tabular
material that fractures to naturally sharp edges; others have edges that
are either flaked or honed to the desired sharpness. The attributes that
serve to distinguish strategically from expediently designed tabular tools
include handles or other modifications for holding or attaching a handle.
The configuration and number of edges vary considerably among tabular
tools. Straight, convex, concave, and serrated edges are known to occur,
sometimes in various combinations on the same tool. Experimental and
comparative use-wear analyses might distinguish some attributes that could
be used to refine the classification of tabular tools and answer questions
about edge configurations. Why are some edges convex, and others con-
cave or straight? Are these configurations manufactured or worn? Are
they related to what was processed, or to differences in the techniques of
processing?
Archaeologists most often associate tabular tools with agave produc-
tion and call them “agave knives” (Bernard-Shaw 1990:190; Castetter and
Bell 1937; Greenwald 1988:173–186), but other uses may include the work-
ing of wood, bone, or hide (Adams 1994b). Tabular tools are generally
distinguishable from saws and files by their finer texture and the selection
of naturally thin lithic material for their manufacture. An interesting minute
feature was recognized on some tabular tools recovered from the Tonto
Basin (Adams 2000c): a barely visible notch through one or more edges on
more than 20 tools was probably associated with a secondary use. The use
wear in the bottom of the notches is consistent with the contact of a soft,
thin surface. Such a pattern of wear was interpreted to have been the re-
sult of working a sinew or fiber strand through the notch, perhaps to
straighten, stretch, or work out imperfections in the strand. The notches
are positioned so that they did not interfere with the use of the tabular
tool edge in other activities, thus the secondary and primary uses were
concomitant.
No concerted effort has been made to trace the distribution patterns
for tabular tools through time or across space. They generally have been
lumped as a class without much attention to details of their morphology
and use-wear patterns. This is a category that needs further attention to
determine the range of activities in which the tools were used, and if there
were technological developments associated with their use.
Chapter 10
Paraphernalia
T he items in the category labeled paraphernalia include personal and

group ritual equipment, gaming devices, weights, ornaments, repre-
sentations (morphic, geometric, and abstract), and items whose specific
functions are unknown. All are articles that were manipulated either ac-
tively or passively. Active manipulation involves movement, such as that
of a ball in a game, or a medicine stone in certain rituals. Passive manipu-
lation involves the transfer of information through representation or dec-
oration. Most of what is known about the items in this category derives
from ethnography. Some items, such as lightning stones, bell stones, balls,
and a few others are classifiable because they are still used in traditional
contexts. Others we can describe by their shape, but we can only specu-
late about their use because no one uses them anymore, or they are used
in ritual contexts that have dubious ties to the distant past.
This grouping has more information about technology than is obvious
at first glance. Even though we may never know the specific functions,
exact meanings, or associated symbolism of some items, their position in
the archaeological record might impart information about their inherent
ritualness. Additionally, there is much to learn about how they were made,
and perhaps something of their intrinsic value is discernible by the level
of effort required to procure materials and manufacture them, and in their
distribution and discard patterns. If nothing else, we can learn something
about the tools and techniques used in their manufacture.
ATLATL PARTS
The intersection of grinding technology with that of propulsion technol-

ogy is not a common archaeological topic. However, it must be discussed
because there are examples of ground or polished stones attached to atlatls.
Questions have been raised asking whether their attachment is functional
or ritual. Those identified as atlatl weights are considered important to the
191
heft and arc of the spearthrower, and those identified as charmstones, to

the tool’s success. Evidence for distinguishing either type is tenuous. They
are identified on the basis of relatively few specimens, most recovered
from Sonora and Arizona. Ferg (1998:546–550, 569–572) comprehensively
summarizes what little is known about stone atlatl parts, such as finger-
loops, weights, and charmstones.
An atlatl is a composite tool that includes a wooden spear tipped with
a stone point and a wooden spearthrower. Ground stone items are some-
times attached to the spearthrower. A few complete spearthrowers that
have been recovered from caves in the Four Corners area of Utah and north-
ern Arizona and from Tularosa Cave in New Mexico have had stones at-
tached (see, for example, Guernsey and Kidder 1921:81–83, Plate 33; Kid-
der and Guernsey 1919:180, Plates 82, 83; Woodbury and Zubrow 1979:54,
Figure 2). Recognizing weights and charmstones when they are not at-
tached to a spearthrower is not an easy task. Any shaped stone of suitable
size could have been attached, but those with suspension grooves are more
likely candidates than those without grooves. One possible weight from a
site in the Tucson Basin measures 5.1 cm long, 3 cm wide, and 1.6 cm
thick (Ferg 1998: Figure 14.5). Most of those attached to the spearthrowers
found in the caves are loaf-shaped or rectangular, and were lashed snug-
gly against the wood. It has been postulated that the unusual cruciforms
discussed below were charmstones attached to spearthrowers (Ferg 1991,
1998:569–570). Their method of attachment is a matter of speculation.
Spearthrowers recovered from the greater Southwest have leather or
twine finger-loops. Two examples of spearthrowers with possible shell fin-
ger-loops are known from those recovered in Mexico during the Spanish
conquest and housed abroad (Ferg 1998:546–550, note 2). Thus, there is
very little evidence for identifying stone finger-loops as attachments to at-
latls (Figure 10.1).The individual pieces found archaeologically and inter-
preted as such are generally manufactured to a crescent shape from smooth,
cryptocrystalline stone and are polished, drilled, and sometimes notched
to accommodate their attachment to something. Some have been identi-
fied as personal ornaments because of their similarity to a shell ornaments
worn by the Cocopah (Ferg 1998:546–550, Figure 14.3).
For the most part, stone atlatl attachments, weights, charmstones, and
finger-loops are rare finds (Ferg 1998:554), but if they are correctly identi-
fied, they signal the existence of a very old technology that was slowly be-
ing replaced by bow-and-arrow technology perhaps as early as the Late
Archaic/Early Ceramic time periods (approximately 800 b.c.–150 a.d.)
(Sliva 1999). Until there is stronger evidence to identify unattached pieces
as specific to atlatl technology, it is perhaps best to avoid classifying them
10. PARAPHERNALIA 193
biconical hole
biconical hole
10.1. A possible atlatl finger-loop. Note the biconical holes drilled into either
end, one of which is broken. (Drafted by Rob Ciaccio.)
as finger-loops, weights, and charmstones, and instead classify them as

shaped items, describe that their specific functions remain uncertain, and
list these as possible uses.
BALLS
Most of what we know about stone balls is derived from ethnographic

references and archaeological associations. Any roughly spherical piece of
stone can be classified as a ball, but distinguishing those that have specific
functions requires analysis. Naturally spherical pieces (expediently de-
signed) that are lightly used or used in activities that cause little use wear
may be difficult to recognize compared to those that are ground to shape
(strategically designed) or extensively used. In ethnographic studies of
Puebloan and non-Puebloan peoples, stone balls have been identified
most commonly as gaming pieces, club heads, noise-making stones, or
racing stones (Adams 1979:90; Culin 1975:671, 675, 678–679; Di Peso et al.
1974:284–289; Russell 1975:172–173,179; Stephen 1936:271–280; Underhill
1939:146–150; Woodbury 1954:173). Woodbury (1954:172) discusses the
possibility that leather-wrapped stones were mounted on sticks as war
clubs by the Hopi. Balls used as gaming pieces or wrapped in leather for
use as war clubs are harder to identify than those used more actively in
races or as noisemakers. Those wrapped in leather should have a sheen
from contact with a soft surface. Gaming pieces may not become obviously
damaged after manufacture. Wear from handling and incidental abrasion
damage may be the only recognizable attributes.
Racing Stones
Racing stones, also called kickballs, are identifiable by impact fractures that
cover most of their circumference. Those recovered from archaeological
contexts range in size from 5 cm to 7 cm in diameter (Adams 1979:90, 1998,
1999; Woodbury 1954:172). Parsons (1939:818) describes historic Puebloan
kickballs as being 11⁄2 inches in diameter, and Russell (1975:172–173) meas-
ured Piman kickballs at 6 cm in diameter. Some have flattened surfaces or
slightly concave sides where pitch was applied to make the stone adhere
to the foot, and to allow the racer to heft the ball into the air for more dis-
tance (Adams 1979:90; Woodbury 1954:172). Ethnographic descriptions of
kickball races explain that the stone is hefted on top of the toes and tossed
ahead of the runner (Parsons 1939:818; Russell 1975:172–173; Titiev 1972:
321–323). Races were run on tracks prepared by the Pimans at Sacaton
Flats and Casa Blanca, as well as through open country (Bahr 1983: Figure
8; Russell 1975: Figure 88).
Thunder Stones
In historic contexts, stone balls were rolled across wood- or stone-covered
pits to simulate the sound of thunder (Parsons 1939:378; Woodbury 1954:
172). Balls used as thunder stones are larger (7 cm to 10 cm in diameter)
and generally more spherical than kickballs. Use-wear damage patterns
are also less obvious on thunder stones. Any impact fractures that occur,
as well as any natural asperities, should be rounded from rolling across a
wooden surface, but this is a speculation that needs to be experimentally
tested.
Gaming Pieces
Small stones, 3 to 4 cm in diameter, were described by Russell (1975:179)
as part of a jacks-like game played by Piman women. Such small sizes
would be more functional than larger stones for games such as the hidden
ball game (Woodbury 1954:173), which is ritual in nature and serves to
call for rain and abundant crops. Some spherical concretions might also
serve as gaming pieces without further modification. Roberts (1940:126,

Plate 47) noted that Zuni workmen considered the small balls recovered
from sites in the White Water District as similar to those they used in the
hidden ball game.
Archaeological Distributions
Balls of varying sizes have been consistently found in Late Archaic (Cienega
phase, 800 b.c.–a.d. 150) and Early Ceramic (Agua Caliente phase, a.d.
150–550) contexts at sites along the Santa Cruz River in the Tucson Basin
(Adams 1998:388, 2000b:51). Those found in the early contexts at Stone
Pipe are of the correct size for use as kickballs and thunder stones, but are
lacking any distinctive use wear that would help identify how they func-
tioned (Adams 1998:388). Those found at Los Pozos are small and similar
in size to those described for gaming pieces, but there is evidence that
they had been wrapped and perhaps attached to a handle (Adams 2000b)
(Figure 10.2). Balls of assorted sizes were recovered from the Badger House
Community (Hayes and Lancaster 1975:162, Figure 208), but there is no
mention of balls other than concretions from later Mesa Verde contexts.
Descriptions of balls also seem relatively rare in reports from New
Mexico, although they have been reported from settlements in the Bande-
lier area (Wood 1982:209, Figure 148). Both pithouse and Puebloan settle-
ments dating from a.d. 400 to a.d. 1350 in the Point of Pines area have
stone balls (Adams 1994a). Di Peso et al. (1974:284–285) describe those re-
covered from Casas Grandes and review various ethnographic and archae-
ological identifications of balls. Perhaps the largest balls are from Casas
Grandes, measuring 11 to 20 cm in diameter, and seem larger than any de-
scribed in ethnographic reports. Woodbury (1954:171) succinctly describes
the problems with discussing the distributions of balls through time and
across space. He asserts that jargon differences, misidentifications, and lack
of reporting makes it unsafe to assume that balls are absent from some sites
just because the term is not used in reports. However, the same is proba-
bly true for other ground stone types as well. Given the data available on
balls of all sorts, perhaps it is time to take on the task of sorting them into
useable categories that can be defined through time and across space.
BELL STONES
Bell stones are recognizable in the archaeological record only because they
were used in historic pueblos, such as Taos, where they were seen and
described by archaeologists (Lambert 1954:132, Plate 28). Suspended by a
wrapping
impressions
length
width thickness
10.2. Balls with evidence of being wrapped, perhaps for hafting (each about
2 cm in diameter). (Drafted by Chip Colwell.) (Adams 1996: Figure 11, 2000b:
Figure 5.2.)
leather thong, the long stones were struck in a way that caused them to
reverberate loudly enough to summon the men to a meeting (Lambert
1954:132). They are described as being 1 foot long, 2 inches wide, and 1
inch thick (Lambert 1954:133). Some have a depression on one side that
perhaps enhanced their tonal quality (Hayes et al. 1981:131), and some are
grooved to facilitate suspension. They range from naturally shaped to those
more carefully manufactured with pecking and grinding to achieve a de-
sired shape. Lambert (1954:132) describes their distribution among late
prehistoric pueblos such as Paa-ko, Rito de los Frijoles, Puye, Cuyamungue,
Puaray, and Kua’ua. In the description of those from Gran Quivera, Hayes
et al. (1981:131–132, Figure 173, called “ringing stones”) note that they
tend to be found in groups. A cache of 23 were recovered from Cuya-
mungue (Lambert 1954:132). The recognition of bell stones at New Mexi-
can settlements may be suggestive of a limited spatial and perhaps tem-
poral distribution, although stones classified as “gongs” (larger stones)
and “ringing stones” (smaller stones) were found at Casas Grandes (Di
Peso et al. 1974:283–284). It is also possible that throughout the South-
west some unidentified pieces of stone, perhaps those identified as plum-
mets, were bell stones, and that their distribution is much wider than cur-
rently recognized.
CRUCIFORMS
Cruciforms are uniquely cross- or X-shaped stones that have been variously
interpreted as functional or representational. They are strategically de-
signed, some polished to a lustrous sheen that obscures any other rem-
nants of their manufacture. The cruciform depicted in Figure 10.3 is an ex-
tine
length
width
thickness
10.3. This cruciform may have been flaked in the initial shaping stages of manu-
facture, but it has been so highly polished that no flake scars are visible. (Drafted
by Rob Ciaccio.)
ample of a very symmetrical, highly polished piece with all four tines ori-
ented at angles of 90 degrees. Other cruciforms are not so finely made
nor so precisely symmetrical. Some that were perhaps not completed have
remnants of flake scars from the initial manufacture stages. There are no
consistent use-wear patterns, but some have striations from rubbing against
something more asperite. Such wear could result from scuffing on the
ground, which may be evidence of their use as gaming pieces, as has been
postulated for some. One or two have been described as having notches
that should be associated with some type of use, but there is no substan-
tial evidence to support anything more than conjectures (Adams 2000b:
51; Di Peso et al. 1974:289). The purposeful destruction of cruciforms has
been identified among those recovered from Las Capas, a San Pedro phase
settlement in the Tucson Basin. The tines are damaged with fractures
caused by the forceful impact of a hammerstone that flaked them from the
body of the cruciform (Adams 2001b).
There are no ethnographic analogies for the function of cruciforms
except perhaps for some reference to their similarity to symbols that rep-
resent stars in rock art and other depictions (Ferg 1998:570; Johnson 1971).
Ferg (1998:560–572) summarizes research on cruciforms. Some of their
possible uses include representations of stars, charms, dice games, medi-
cine man’s tools, and atlatl weights (Di Peso et al. 1974:592; Hemmings
1967:162). The strongest temporal associations of cruciforms in the U.S.
Southwest are with the Cienega phase, Early Agricultural period, and pre-
ceramic contexts, dating earlier than a.d. 800 (Ferg 1998:572). Their geo-
graphic distribution seems to be limited to the southern U.S. Southwest,
extending into northern Sonora and Chihuahua (Ferg 1998:560).
CYLINDERS AND MEDICINE STONES
If the term cylinder is used in the Flagstaff area or in southern Arizona, it

refers to pieces of scoria or vesicular basalt that are completely different
from the highly polished pieces of fine-grain material referred to as cylin-
ders in the Four Corners area. The main problem, however, is that no mat-
ter how these shaped stones are classified, there is no clear understanding
of how they were used or if the regional differences in material selection
are related to different uses or different material availability. To further
confuse the issue, cylindrical stones are sometimes lumped with pieces
that are not quite cylindrical, and no matter where they were found or
what they were made from, they are often called “medicine stones.”
It is difficult to describe a single item as typical of all medicine stones.
The term is applied to items that are carefully manufactured to unusual
shapes and are interpreted as having some ritual significance. Many have
been found in contexts such as burials and caches that seem to support
such an interpretation. There are no obvious use-wear patterns to help
identify how they were used, and thus their function remains unknown.
Scoria and vesicular basalt cylinders range in size from 4 to 10 cm long,
and 2 to 3 cm in diameter. The sorting of conical or other shapes into the
cylinder category makes it difficult to make comparisons with other as-
semblages. This confusion is a product of classification and not a property

of fuzzy sets, for which we know how the tools functioned but their at-
tributes overlap with other sets.
Kamp (1995; Kamp and Whittaker 1999:113–115) working with collec-
tions from Lizard Man concluded that at least some scoria or vesicular
basalt cylinders with residues and use-wear patterns consistent with ex-
perimental tools served as pottery production tools. Further support for
this conclusion can be seen among the collections of vesicular cylinders at
the Museum of Northern Arizona from Pittsburg Fort and Nalakihu that
also have residues suggestive of their use in pottery production. Another
possible use was recognized during the analysis of cylindrical and conical
stones from Point of Pines settlements. One vesicular piece from a burned
room has minute burned agave fibers (Lynn Teague, personal communi-
cation, 1993); these are deeply embedded in the vesicles and visible only
under magnification. Perhaps these stones were used somehow in straight-
ening or cleaning fibers, and were preserved on this one only because of
the fire. Scoria or vesicular basalt pieces similar to cylinders have been
recognized as plugs and are described in the Chapter 11, “Containers and
Container Closures.”
Vesicular cylinders are scarce in Hohokam assemblages. None were
described from Snaketown, although Haury (1976: Figure 14.32) illus-
trates an ungrooved plummet that looks like a carefully shaped cylinder.
As the exception to the rule, numerous vesicular cylinders were recovered
from contexts that date to the Sacaton phase (a.d. 1070–1140) at Las Colinas
in the Phoenix Basin (Euler and Gregory 1988:310–311). Euler and Gre-
gory (1988:312) postulate that the clusters of cylinders found at Las Coli-
nas are the product of an enclave of non-Hohokam people. They also see
a similar pattern with the medicine stones and speculate that the two
types may be related (Euler and Gregory 1988:312). At about the same
time, cylinders are common in the Flagstaff area (Adams 2001a). Colton
(1946:288) postulates that vesicular basalt cylinders were at Cohonina set-
tlements (a.d. 900–1050) before they were at Sinagua settlements (after
a.d. 1070). McGregor (1941:178) identifies cylinders (his “type A”) as one
of the most distinctive Sinagua artifacts beginning with the Padre phase
(about a.d. 1100). The fact that he laments that they are rare or missing
from most Winona-phase contexts—a phase that was, in his view, heavily
influenced by Hohokam—must mean that he considered the vesicular
cylinders to be of Hohokam derivation. Vesicular cylinders were found at
settlements along U.S. 89 from contexts that date as early as a.d. 800–
1000, but they were most common among contexts dating post-a.d. 1065/
1075 (Adams 2001a).
During the later occupations at Wupatki especially, and among Elden-

phase (about a.d. 1150) Sinagua contexts, cylinders were quite common
(Colton 1946:288). Cylinders have been recovered from several contexts at
Elden Pueblo (Peter J. Pilles Jr., personal communication, 1999). Within
the collections at the Museum of Northern Arizona there are cylinders
from Wupatki, Grand Falls, Deadman’s Fort, Medicine Fort, Ridge Ruin,
Juniper Terrace, Winona Village, NA11,237, and NA11,535. There are also
reports of their location at Walnut Canyon (Stanislawski 1963:117), Lizard
Man (Kamp 1995; Kamp and Whittaker 1999:113–114), Big Hawk Valley
settlements, Stone Axe, Tuzigoot, Los Muertos, Canyon Creek, SU (Smith
1952:127–129), and Point of Pines (Adams 1994a:142, called “plugs”).
Woodbury (1954:181) notes the presence of vesicular cylinders at Pueblo
II–III Jeddito settlements, their declining frequency at Awatovi between
Pueblo III and Pueblo V, and their Pueblo IV presence at Stone Axe. From
this cursory recounting of what is known about the temporal and spatial
distribution of vesicular cylinders, it seems that they were recognized as a
technological development to be applied to a variety of tasks sometime
after a.d. 800 by folks living in central and northern Arizona.
There are also cylindrical stones made from a wide variety of fine-
grain or cryptocrystalline material, and these are usually more carefully
ground and shaped than cylinders of scoria. They have been variously
identified as medicine stones or possible ornaments such as nose plugs
(Wheeler 1980:293; Woodbury 1954:183). They range in size from 2 to 4 cm
in length, and .5 to 1 cm in diameter. No systematic study has been done
of their distribution through time or across space, but they seem more
commonly found in Basketmaker and Puebloan contexts in northern Ari-
zona, Colorado, and New Mexico than elsewhere (Dick 1965b:189; Hayes et
al. 1981:137, Figure 191; Kidder 1932:92–93, Figure 68; Morris 1980: Figure
43q; Rinaldo 1964:81; Stubbs and Stallings 1953:122, Figure 121f; Wheeler
1980:293, Figure 344; Woodbury 1954:183–184, Figure 28o–t). Because
these stones are so different than the scoria or vesicular basalt cylinders,
there should be some terminological distinction; however, suggestion of a
term should await a systematic study of their distinctive attributes and
their distributions through time and across space.
DISKS
This category includes a wide variety of items that share the common at-
tribute of having a general disk shape. There are several designs that vary
in diameter, thickness, material, manufacturing techniques, and presence
or absence of perforations. Thin disks without perforations are generally
manufactured from a fine-grain or cryptocrystalline material through var-

ious combinations of flaking, grinding, and polishing. How they func-
tioned is unknown, but their use as gaming pieces is the most common
speculation. Culin (1975:726–727) describes a Zuni game in which multi-
ple disks are used to knock down a standing corn cob. These disks range
in diameter from 5 to 13 cm. Some are crudely shaped, and others more
carefully manufactured and decorated with incised designs (Culin 1975:
Figures 951–953). A few items classified as disks from Early Ceramic con-
texts in the Tucson Basin have a facet worn from a possible secondary use as
polishing stones (Adams 1998:384–385). Some reports include lids in the
general category of disks. In this technological scheme, lids are sorted into
the category of containers and container closures, discussed in Chapter 11.
Perforated Disks
Perforated disks are a subset of the larger disk category. Some are obvi-
ously recognizable as whorls, or the fly-wheels that maintain the rotation
of a spinning shaft. Such whorl/shaft composite tools are used for spin-
ning fibers, drilling holes, and starting fires. See Chapter 8 for more dis-
cussion about whorls. Disks with two small holes may be toys, identified
as buzzers (Culin 1975:751; Hough 1918:290, Plate 47) or noisemakers called
whizzers (Adams 1979:95; Stephen 1936:106). Whizzers make noise by
means of a twisted string looped through the holes so that it can be alter-
nately loosened and tightened to spin the disk and create a hum. Such
humming was an important part of certain Hopi rituals (Stephen 1936:
106). There is some confusion in the use of the terms “whizzers” and
“buzzers.” Fewkes (1894:33) uses the term “whizzer” for a wooden noise-
maker that is twirled on a long string above the head or beside the body
and commonly referred to as a “bullroarer” by others. Stephen (1936:106)
and Adams (1979:95) use the term “whizzer” for the item identified as
“buzzers” by Culin (1975:751) and Hough (1918:290, Plate 47).
Stone Rings and Donut Stones
Some large perforated disks have been referred to as “stone rings” (Haury
1976:290–291), and others have been described as “doughnut shaped” or
called donut stones (Figure 10.4) (Di Peso et al. 1974:32, 307; Haury 1945:
141, Plate 63; Kelly 1978:96, Figure 6.14). The term ring implies a circular
band where the diameter of the hole is larger than half the diameter of the
entire piece, whereas the term donut implies a smaller hole. These rings
should not be confused with finger rings, which are much smaller and
outside hole diameter margin
inside hole
diameter
thickness perimeter
biconical hole
length
width
10.4. Perforated stone disk or donut stone. The hole was drilled from both sides,
making it biconical. (Drafted by Chip Colwell.) (Adams 1996: Figure 14.)
more delicate. Some stone rings have been called “pulley type” rings be-
cause they have an encircling groove on the perimeter like a modern pul-
ley. The margins surrounding the holes vary in configuration, with rings
usually having convex margins, and donut stones having convex, flat, or
even concave margins. Stone rings and donut stones range in diameter
from 5 to 18 cm; those with biconical holes have outer diameters of 2 to 6
cm, and inner diameters of 2 to 4 cm.
Suggestions regarding their possible uses include as tools for shelling
corn, as weights on digging sticks, or as gaming stones, such as the chun-
key stones identified in ethnographic records (Culin 1975:490, 521; Haury
1976:290). That some might have been used as digging-stick weights seems
more plausible in light of such a use by the Chumash in California (Hud-
son and Blackburn 1982:247–251). A wide variety of Chumash weights
are illustrated and classified in the report by Hudson and Blackburn (1982:
247–251), but it would be useful to know the diameter of the holes and the
use-wear patterns for comparison with archaeological specimens. A hole
only 2 cm in diameter such as they describe for some seems a little small
for a useful digging stick.
Rarely, archaeological context provides associations that may reflect
how some disks were used. One donut stone was found on the floor of a
Maverick Mountain–phase structure at Point of Pines Pueblo along with
an axe and other axe-shaping tools. The donut stone hole is 2.6 cm in di-
ameter, and the groove width on the axe is 2.2 cm. Use wear in the hole is
similar to that on experimental tools used to work wooden surfaces. The
hole could have served as an abrader to smooth and possibly bend a han-
dle for the axe. However, the damage patterns on all donut stones are not
the same, and their use remains unknown. Some donut stones are either
incompletely perforated or have shallow basins. See the Chapter 5 discus-
sion of mortars for more about the possible uses of these disks.
Use-wear analyses should help identify possible uses of the various
perforated disks and rings. Manufacture damage includes impact frac-
tures and striations from finishing and drilling the holes. If they were used
as digging weights or abraders, there should be evidence of contact with
wooden surfaces in the holes overlying the impact fractures and stria-
tions from manufacture. If the disks were used as chunkey stones, there
should be impact fractures along the perimeter where they were rolled
along the ground—perhaps similar to the impact fractures on kickballs.
Wear in the perimeter grooves on pulley-like rings should help determine
whether something was wrapped in the groove, or if they served as finger
grooves or in some other capacity. The rings and donut stones that I have
looked at have a wide range of use-wear patterns. Some have no evidence
of use, while others have evidence of multiple uses, remnants of pigment
from secondary processing activities, or wear in the hole that is different

than wear on the margins. Some margins have evidence of their use as
anvils or lapstones.
There has been no systematic study of the distribution through time
or across space of stone rings or donut stones. They seem to occur prima-
rily in contexts in the southern Southwest that date as early as, if not ear-
lier than, 400 b.c. (Adams 1998, 2000b:112; Huckell 1995:66, Figure 4.11).
In the Phoenix Basin they have been found among Pioneer and Sedentary
deposits at Snaketown (Haury 1976:290–291). The donut stones recovered
from Hodges Ruin in the Tucson Basin came from Rillito- and Tanque
Verde–phase contexts (Kelly 1978:96). Those recovered from the Point of
Pines area were in Puebloan settlements dating post-a.d. 1100 to 1350
(Adams 1994a:108–109). Rings and donut stones are not commonly found
in sites in the northern U.S. Southwest. Should a systematic study be done
in the future, it is imperative that rings be sorted from donut stones, and
that a comparison be made of use wear, material types, and configurations.
LIGHTNING STONES
A few items have been identified as lightning stones based on their simi-
larity to those observed in historic contexts (Dick 1965b:188–189; Jeancon
1923:68; Kidder 1932:93, Figure 69). They are usually described as white
quartzite pebbles that have been rubbed or struck against a similar stone
to generate friction heat and sparks. As such, they are expediently designed
and should have impact fractures and striations from use. A lightning
stone set illustrated by Kidder (1932: Figure 69) has a grooved stone within
which a cylindrical stone was rubbed. Lightning stones seem to be recog-
nized among New Mexican assemblages more often than elsewhere (Kid-
der 1932:94). For example, they have been described at Forked Lightning
(Kidder 1932:93–94, Figure 69), in the Chama area (Jeancon 1923:68, Plate
59b, called “firestone”), Paa-ko (Lambert 1954:130, Plate XXVII), Gran Quiv-
era, and Las Humanas (Hayes et al. 1981:132, Figure 132). Kidder (1932:
94) discusses how he worked the set recovered from Forked Lightning
until it glowed. Jeancon (1923:68) mentions that the use of such stones to
simulate lightning at the historic pueblos of Santa Clara and San Ilde-
fonso was confirmed by men from these villages.
PLUMMETS AND WEIGHTS
If the items classified as plummets were, in fact, used in the same manner
as modern plummets, they were designed to hang straight down, thereby
defining a precise vertical line. Such a precise line is needed for measur-
ing, engineering, and building purposes. Thus, the identification of plum-

mets in the archaeological record has technological implications beyond
grinding technology. Di Peso et al. (1974:237, called “plumb bobs”) specu-
late that plummets were used as architectural instruments in the laying
out of Casas Grandes (Paquimé). If this is truly how they were used, then
perhaps those from Hohokam contexts were also used for aligning canals
or building mounds. Could plummets have been useful in making celes-
tial observations?
The items considered plummets are generally pecked and ground to
conical or cylindrical shapes, and grooved, notched, or perforated for sus-
pension. They are considered expediently designed only if they are other-
wise unmodified. There is fuzziness between the categories of plummets,
cylinders, and weights. Features designed for suspension may be the only
recognizable attributes to distinguish plummets from cylinders. Direct ev-
idence for the use of tools designed for suspension as weights is minimal
in the U.S. Southwest. See, for example, the small grooved stone, possibly
a weight tied with a fiber twine, that was recovered from northern Ari-
zona (Kidder and Guernsey 1919: Plate 51g). Hayes and Lancaster (1975:
160) speculated that small stones grooved for suspension at the Badger
House Communities were used to tether birds. Their use as net weights
and in weaving activities has also been postulated.
PIPES AND TUBES
Pipes are tubes designed with bores to hold burning tobacco. Tubes may
have functioned in other capacities, or they may be unused pipes. Both
pipes and tubes are usually strategically designed, with pecking and grind-
ing used to shape the exterior. Unfinished pipes and tubes provide evidence
of how they were manufactured. Kidder (1932:84, Figure 60) illustrates
different stages of tubular pipe manufacture. The exterior is roughly shaped
before drilling the bore from both ends of the tube, thereby creating a bi-
conical hole. Once the bore is completely through the tube, the exterior is
finished. Several broken vesicular basalt pipes from Las Capas in the Tuc-
son Basin are evidence of how difficult it can be to successfully complete a
bore. Flaked drills were used to make the bores. Guernsey and Kidder
(1921:95, Plate 35) illustrate a hafted flaked tool that has a bit compatible
in size and use wear to have been used to make the pipes found in the
Basketmaker caves in northeastern Arizona. Other tools probably used in
pipe manufacture were recognized among the hammerstones and lap-
stones found at Las Capas.
Pipes and tubes can be categorized by their design and hole configu-
ration. Most pipes in the U.S. Southwest are either conical or cylindrical.
Cylindrical pipes have a more or less uniform exterior diameter and a

rod shape distinct from that of conical pipes (Figure 10.5). Conical pipes
are commonly referred to as “cloud blowers” because of the ethnographic
associations mentioned below. Globular pipes are wider in the middle
than on either end (Figure 10.6). Bores are drilled to either a biconical or
cylindrical configuration. The narrower interior hole of a biconical bore
holds the tobacco at some distance from the lips. Some pipes are manu-
factured with a socket for the attachment of a stem, which also keeps the
tobacco away from the lips (see, for example, Ferg 1998: Figure 15.15).
Others have wood, reed, or bone stems inserted in their narrowest open-
ings, which are sealed with a resinous substance (Figure 10.6). A few elab-
orately designed pipes called “elbow pipes” have been recovered from
Pecos Pueblo (Kidder 1932:85, Figure 61) and recognized among the pipes
from Hopi villages (Woodbury 1954:175).
Descriptions of Piman pipe use include both blowing and sucking mo-
tions (Russell 1975:112). Piman pipes were also used without tobacco for
sucking illness out of patients (Russell 1975:112). Perhaps some of the un-
burned pipes and tubes recovered from archaeological sites had similar
uses, although it would be impossible to distinguish a sucking tube from
a newly manufactured pipe. Ethnographic accounts of Puebloan pipe smok-
ing describe men socially sharing the pipe or blowing smoke over objects
or people in ritual observances (Stephen 1936:683). Cloud blowers are
used by placing the lips on the large end and the small end over an object.
The smoker puffs the smoke through the cloud blower and out over the
object (Fewkes 1894:32; Parsons 1939:683). With other pipes, including those
that have stems, the smoke is inhaled through the pipe and the smoke
held in the mouth until it is blown over the object (Fewkes 1894:32). Voth
(1903:310) describes a cone-shaped “cloud producer” used in conjunction
with cigarettes. Pipes are also components of certain Hopi altars (Hough
1915:137).
Woodbury (1954:175) summarizes the distribution of pipes as it was
understood in the 1950s. He mentions Rinaldo’s suggestion that thick tu-
bular pipes with a stem of bone or wood were characteristic only of eighth-
century Mogollon (Woodbury 1954:175). It is unclear if the shape or the
presence of a stem is the attribute with the limited distribution. No pipes
with stems were recovered from Jeddito settlements. He goes on to men-
tion that the more-slender pipes recovered in the Jeddito area generally
date to later contexts and have a wider distribution than those discussed
by Rinaldo for the Mogollon (Woodbury 1954:175). Wheat (1955:124) also
10.5. Conical pipe with a biconical bore. (Drafted by Rob Ciaccio.) (Adams
2001a: Figure 15.)
associates stemmed pipes with Mogollon technology: they have been re-
covered from SU (Rinaldo 1943: Figure 71) and from the pithouse occupa-
tion in the Kiatuthlanna area (Roberts 1931:159, Plate 640).
The evidence for time and space distributions of pipes with bone or
wood stems has increased since 1950. It now seems that they occur much
earlier and have wider geographic distributions than previously thought.
10.6. Pipe with a bone stem in the bore. (Photographed by Scott Van Keuren.)
For example, excavations at Las Capas, along the Santa Cruz River in Tuc-
son, uncovered several pipes from contexts dating earlier than 800 b.c.,
including at least two with bone stems (Figure 10.6). Stemmed pipes have
been recovered from New Mexico in both Basketmaker contexts in the
Prayer Rock District (Morris 1980: Figure 40, called “bits”), and in late
Pueblo contexts at Gran Quivera (Hayes et al. 1981:130–131). In work more
recent than his dissertation, Rinaldo (1964:81) suggests that the tubular
pipes reflect Zuni ancestry and the cloud blowers, Hopi ancestry; and that
vesicular basalt pipes are common to Mogollon, but rare in Anasazi. Stone
pipes of any sort are also considered rare in Hohokam technology (Euler
and Gregory 1988:312; Haury 1950:332). Considering the data from the
Archaic deposits in Ventana Cave (Haury 1950:329, Figure 79), and the
fact that stone pipes are still in use among Native American groups such
as the Hopi and Zuni, it is clear that smoking technology is long-lived,
and that stone pipes are among the oldest and most enduring ground
stone tools. Ferg (1998:595–606) summarizes the antiquity and distribu-
tion of various types of pipes in the greater Southwest.
SHAPED STONES
This is a rather artificial set of items that are grouped only because there is
no other category into which they fit. Some shaped stones were manufac-
tured into abstract shapes, but their specific use is unknown. They are of-
ten identified as ritual items and perhaps endowed with symbolic mean-
ing because they occasionally show up in contexts that seem ritual, and
because we know of no other function for them. Most shaped stones have
been altered only through manufacturing, but some have evidence of
handling.
Some large tabular stones are not obviously tools or architectural pieces.
In the classification scheme presented here, such slabs are considered
netherstones if there is use wear (see, for example, Hayes and Lancaster
1975:159, Figure 200). Those small enough to be handheld are classified as
tablets if there is no use wear, and lapstones if there is. Rohn (1971:241)
defines tablets as flat slabs that have been extensively ground to shape.
Some of the tablets included in this category have also been called “sandal
lasts.” For discussions about this controversial identification, see Kidder
and Guernsey (1919:105–106), Morris (1939:131–132, Plate 144), and Wheeler
(1980:289–291, Figures 339, 340).
Other items that also can be included in the shaped stone category
are called “corn mounds” (Martin et al. 1956:84), “corn goddess symbols,”
“tiponi,” “cones,” “conical fetishes,” “phallic symbols,” and other names
(see also Dorsey and Voth 1902:181; Fewkes 1911:67; Hayes and Lancaster
1975:163, Figure 209; O’Bryan 1950:85; Roberts 1932:61, 143; Smith 1952:
233; Voth 1912:52–53).
PIGMENTS
Traditional technologies grind pigment and mix the resulting powder with
an adhesive binder. Nothing more than water is necessary if the pigments
are clay based or have some other natural adhesive property. Otherwise,
vegetal gums or animal fats are needed to bind the powder. A dispersion
or paint is created by diluting the pigments with a liquid vehicle such as
water (Odegaard and Crawford 2000:9.2).
Evidence for pigment production is found among prehistoric sites in
the form of raw material, processed pigment powders and cakes, and the
tools used in the various stages of pigment processing and application.
See, for example, the artifacts and pigment cakes in Chaco Canyon from
Pueblo del Arroyo (Judd 1959:140). In the Tucson Basin there is evidence
for the use of ceramic dishes to standardize the size of pigment cakes
made at the Tortolita-phase (a.d. 450–700) settlement of Valencia Vieja
(Adams 2000a). Handstones, lapstones, and netherstones covered with
pigment from this site may have been devoted exclusively to processing
pigments. The several manos and metates with evidence of pigment were
secondarily used for pigment processing. The same suite of pigment-
processing tools was recovered from much earlier contexts (approxi-

mately 800 b.c.–a.d. 100) in the Tucson Basin at sites such as Santa Cruz
Bend, Las Capas, and Los Pozos (Adams 1998, 2000b). It is not unusual to
find evidence at prehistoric sites of pigment production, probably one of
the oldest technologies in the world. Unfortunately, no systematic studies
have been done of the ground stone tools involved in pigment produc-
tion. A comparative compositional analysis of the pigment residues on
the tools would provide valuable information useful for tracking techno-
logical developments through time and across space.
Stephen (1936:1190–1195) describes in detail several techniques used
by the Hopi in the late 1890s for creating various colors. Historically, the
Hopi have traded with the Havasupai for red ochre procured from a cov-
eted source in the Grand Canyon, and they, in turn, traded it to the Rio
Grande Pueblos (Colton 1948:125). Pigments are used by the Hopi in a va-
riety of coloring tasks, both secular and ritual. It seems, however, that
they have not used paint as personal body decoration as much as the na-
tive groups farther south. Some methods of preparation involved elabo-
rate methods of cooking, straining, and mixing with a variety of sub-
stances. The following is a simplification of the methods described by
Stephen (1936:112, 1190–1195) and serves as a model for pigment process-
ing in general. Turning rocks and minerals into pigment and paint involves
several basic steps: (1) the raw materials—such as iron-based stones and
clays, copper ore derivatives such as malachite and azurite, white stones
such as gypsum and white kaolin-based clay—are reduced to powdered
pigment between a handstone and netherstone, or pestle and mortar; (2)
some pigments are stored in powder form and others molded into cakes;
(3) cakes are created by boiling the pigment with a mixture of plant gum
and water until it reduces enough to hold a shape; (4) as needed, the dried
cakes are reground into powdered pigment; (5) paint is created by adding
a vehicle such as water or saliva generated by chewing appropriate seeds
(for example, squash, melon, sunflower).
McGee (1898:164–169) notes that in the late 1800s Seri women painted
their faces more often than men. Paint was made from ochre (pink, red,
and brown), gypsum (white), and dumortierite (blue), and applied with a
horsehair brush (McGee 1898:166). Face painting had a sacred role in mor-
tuary customs, combined with an importance to individuals as personal
decoration. Paints and painting equipment were highly prized personal
possessions (McGee 1898:166). Russell (1975:160–161) describes Pimans
mixing red ochre with mother’s milk to paint a baby’s face after birth to
improve skin. Other body paints mixed with grease were used by both
men and women; men used more black, and women more red (Russell
1975:161). White or red was used to color hair. Willow or mesquite char-
coal was worked into pricked skin to create tattoos.
NATURAL STONES
Many items were collected prehistorically, probably because of their un-
usual natural shape, the significance of which is knowable only to the
original collector (Figure 10.7). Some are reminiscent of human or animal
figures, and it is easier to postulate why these might have been collected
than others (see, for example, those illustrated by Hayes and Lancaster
1975:164–165, Figures 212–213). Natural stones include concretions, fossils,
meteorites, petrified wood, mica, selenite, gypsum, asbestos, crystals, and
other rocks and minerals. For the most part, they have not been altered,
although handling or moving them causes wear that sometimes can be
recognized. Wheeler (1980:295–298, Figures 349–353) extensively catego-
rizes and describes the many natural stones from Long House. They are
also briefly mentioned in almost any report on stone from all parts of the
U.S. Southwest. However, there is as yet no comprehensive description of
their distribution through time and across space.

Archaeologists often assign natural stones some sort of ritual significance
because they have been found in kivas, caches, and graves (Di Peso et al.
1974:335; Hayes et al. 1981:136–138; Kidder 1932:104–110; Woodbury 1954:
188), and because of frequent ethnographic references to their addition to
altars and rituals. For example, long pieces of petrified wood have been
identified as possible tinklers worn in certain Puebloan ceremonies or
placed as ritual objects in shrines and altars (Fewkes 1898:586; Hayes et al.
1981:129; Parsons 1939:308, 330; Smith et al. 1966; Woodbury 1954:193;
Voth 1903:337). Stephen (1936:400) notes that petrified wood and other
stones were used by Hopi as pendants to stabilize lightweight masks worn
during dances. Hopi who visit their ancestral villages at Homol’ovi set-
tlements where petrified wood pieces are abundant still collect them for
their personal use.
Hopi altars described by Stephen (1936) have crystals, fossils, pebbles,
and interestingly shaped stones in specific positions. Woodbury (1954:
189) reviews accounts of historic Pueblo beliefs that concretions are im-
bued with magical powers making them important ritual objects. Some
have more personal importance to the individual who found them. Roberts
(1940:154) cites several ethnographic references to infer the use of crystals
in curing ceremonies associated with the sun, and Parsons (1939:330) notes
length
width
10.7. Natural shape. This item was perhaps collected for its unusual shape. It is
weathered so that the different textured layers stand out. (Drafted by Chip
the placement of concretions and other stones into shrines associated with
sun and clouds. Certain Hopi rituals use crystals to direct sunlight (Fewkes
1898:586; Parsons 1939:308, 986).
PERSONAL ORNAMENTS
The large set of ornaments includes several subsets: beads, pendants (in-
cluding earrings), mosaic tesserae, nose plugs, bracelets, rings, figurines,
crescents, and other geometrics (Di Peso et al. 1974; Gladwin et al. 1975:
126–130; Jernigan 1978; Judd 1954). The classification presented here is a
simplified version of that used by Jernigan (1978) for ornaments of vari-
ous materials, including stone, bone, and shell. He details temporal and
spatial distributions of many ornament styles and should be consulted as
the primary reference on personal ornaments in the U.S. Southwest (Jerni-
gan 1978). The technological knowledge for making ornaments can be ex-
tensive and goes beyond grinding and polishing. The most complex orna-
mental pieces, composed of tiny tesserae of stone and other materials, are
mosaics. Mosaics require technological knowledge associated with adhe-
sive manufacture, and etched pieces require technological knowledge about
making and using acids. Thus, several technological strategies are involved
in manufacturing personal ornaments.
The tools used in ornament manufacture include netherstones, lap-
stones, abraders, saws, files, and reamers. Netherstones and lapstones pro-
vide a working surface upon which ornaments can be modified by an-
other tool: chipped, drilled, and polished. Their texture also can be used
to abrade or polish the ornament. Grooved abraders are useful in the final
stages of bead manufacture (Cosgrove and Cosgrove 1932:62; Haury 1985a:
119–121; Jernigan 1978:20; Judd 1954:86). Judd suggests that a groove on the
narrow edge rather than across the flat surface was distinctive to grooved
abraders used in bead manufacture (Judd 1954:86). Haury (1945:143, Fig-
ure 86) describes an abrader with a narrow groove as a “bead rasp” used
for manufacturing beads. A flat abrader is illustrated by Judd (1954: Fig-
ure 18) as used to shape turquoise ornaments.
The tools and techniques of ornament manufacture are known from
archaeological, ethnographic, and experimental sources. For example, some
tools are known about because of a tool cache associated with finished
and unfinished beads recovered from Swartz Ruin (Cosgrove and Cosgrove
1932:62, Figure 12). The netherstones used in their manufacture are flat
with irregular basins worn in them. The mass production of beads was
recognized among the manufacturing remains recovered from Tla kii and
other ruins in east-central Arizona (Haury 1985a:119, Figure 59). Tabular
pieces of steatite were scored with a saw or file and snapped into tiny
squares, creating bead blanks. The blanks were drilled, strung, and then
worked against an abrader for the final shaping.
Drill bit shape determines the configuration of the ornament hole
(Judd 1954:87; Haury 1931b: Figure 1). A cylindrical hole can be drilled with
the use of a cylindrical drill tip, such as a cactus spine, grass stem, or
other shaft bit. Cactus spines and grass stems were used to replicate the pos-
sible manufacturing techniques for tiny disk beads (Haury 1931b). Haury
(1931b:86–87) demonstrates that a cactus spine can be worked to create a
cylindrical hole, and that the procedure is enhanced with the addition of
sand and water. The hole created with a stone-tipped drill is conical. As
the drill penetrates the ornament, the wider drill body cuts a larger bore
than the narrow tip. If the hole is drilled from both sides of the ornament,
a biconical hole is created. If drilled from only one side, the hole remains
conical. A biconical or conical hole can be modified into a cylindrical hole
with additional work.
How ornaments were used can be inferred from where they have been
found archaeologically and modeled after historic examples. For exam-
ple, at Casas Grandes, pendants were found under support posts and in
viga holes, perhaps placed there as offerings during construction (Di Peso
et al. 1974:249). Parsons (1939:300) discusses the Puebloan perspective on
the power of various beads, both in strands and individually, and gives a
range of possible uses in addition to personal ornamentation. Specific col-
ors are selected for rituals surrounding burial and hunting, and are laid
down as foundation deposits during house construction. Certain powers
are associated with war at Zuni, and with the giving of beads to the war
gods. Red beads are associated with women at Isleta Pueblo.
Ornament Subsets
Ornament subsets have fuzzy boundaries, and analysts probably sort mi-
nor variations differently. The main goal is to be consistent. For example,
there is a fuzzy boundary between some beads and pendants. Beads are
perforated with a suspension hole that extends approximately through
the middle from side to side so that the edge or perimeter of the ornament
is most visible when strung. Pendants are perforated with a suspension
hole positioned toward one end and strung so that the broad surface is
most visible. Some larger teardrop and bilobe beads may be indistinguish-
able from small pendants.
Disk beads measure thinner than their circumference (called “disc” by
Jernigan 1978:94). Cylindrical beads measure thicker than their circumfer-
ence and have straight sides. Barrel beads measure the same as cylindrical
beads but are distinguished by having convex sides. Biconvex beads, also
called lenticular beads, are disk beads with curved rather than flat surfaces;
plano-convex beads have one flat and one convex surface. Tube beads, like
barrel beads, measure thicker than their circumference but can be distin-
guished from barrel beads by their straight perimeter; barrel beads have a
slightly convex perimeter. Teardrop and bilobe beads (also called “Figure 8
beads” [Jernigan 1978:95]) are a little different in that their hole is slightly
off center. When they are strung, their perimeter is most visible. Spool beads
have a constriction encircling the bead perimeter. Cuboid beads have a rec-
tilinear rather than circular perimeter.
Figurine beads can be classified according to what they represent: an-
thropomorphs, zoomorphs, or geometric shapes. Within both the bead
and pendant categories are figurine types that can be either two- or three-
dimensional. The positioning of the suspension hole, which determines
how the piece hangs, is what distinguishes a figurine bead from a figurine
pendant. The same is true for pieces of various geometric shapes: triangu-
lar, square, rectangular, and so on. See previous definition of beads and
pendants.
Finger rings are small, complete rings that encircle a finger or toe. Ring
bracelets are larger complete rings that encircle an arm or leg. These are
distinguished from C-rings and C-bracelets, which have an opening to fa-
cilitate hooking them over a nose, ear, or arm. Striations on their interiors are
more often remnant from their manufacture using abrasive reamers than
from use wear. Contact against skin creates a sheen and rounds asperities.
Mosaic tesserae are small, thin pieces of material cut into various shapes
and used to create a pattern of stone attached with a resin or tar to a thin
backing of shell, wood, bone, or other material (Di Peso et al. 1974:271).
Sometimes the flat tesserae surfaces are abraded or polished. If a saw or
file was used to cut the tessera material, and the piece was not polished
during the final stages of production, then striations from the saw or file
used to cut the material should be visible on the edges. Such scratches
should be oriented parallel to the broad surfaces. If striations are visible
perpendicular to the broad surfaces, they are damage from the use of an
abrader.
Other ornaments that are not very well documented are plugs and
toggles. Plugs are perhaps the most personal of ornaments in that they
pierce ears, cheeks, noses, and lips. They are manufactured in various sizes
and shapes. Toggles may be difficult to distinguish from plugs, but they
are designed to be fasteners. Toggles are usually cylindrical, with an encir-
cling groove for fastening the toggle to another surface. A toggle fastens
one surface to another when a loop is placed over the toggle, or when the
toggle is pushed through a hole and turned so that it cannot slip back
through. Toggles and plugs are usually abraded to shape and sometimes
polished.
Buttons are small disk pieces drilled with one or more holes for at-
tachment as a fastener. Those with one hole may be hard to distinguish
from beads or pendants. For the most part, buttons have larger diameters
than beads, and more centrally located holes than pendants. They are gen-
erally ground on the broad surfaces and the perimeter to shape. Distin-
10.8. Figurine. Incision, carving, and drilling were used to fashion this animal.
(Drafted by Rob Ciaccio.) (Adams 2000d: Figure 7.11b.)
guishing buttons from whizzers (described in Chapter 10 as noisemakers)

may be a bit of a challenge. Generally, whizzers are larger and probably
have more use wear in their holes than buttons. Both are shaped through
abrasion.
FIGURINES
Stone figurines depict a variety of beings ranging from realistic forms to

abstractions, including humans, animals (Figure 10.8), birds, and insects.
Unfinished figurines recovered from Casas Grandes provide evidence of
manufacturing techniques (Di Peso et al. 1974:293). Pecking, sawing, grind-
ing, drilling, and polishing are all used to shape the forms. Grooves, holes,
engravings, and carvings are used to embellish them. Ethnographic descrip-
tions of figurine use generally center around rituals. They are placed on
altars at Hopi, Zuni, Acoma, Laguna, and Zia, according to Cushing (1883)
and Stevenson (1904:378). Prehistorically they seem most common to
Tularosa-phase contexts in the upper Gila drainage and the White Moun-
tains at sites such as Pinedale, Showlow, and Kinishba (Rinaldo 1956:124).
The figurine depicted in Figure 10.8 was recovered from Sunset Mesa, a
Middle Rincon–phase (about a.d. 1000) settlement in the Tucson Basin
(Adams 2000d).
Chapter 11
Containers and
Container Closures
A ground stone container is a hollow utensil in which material is held or

carried, and the category includes bowls, trays, palettes, and censers.
Stone container closures include lids, plugs, and caps. A container can
function without a closure, but a closure cannot function without a con-
tainer. However, stone closures are not used only with stone containers.
Small stone closures have been found sealing ceramic or gourd containers,
and large ones sealing pits. Figure 11.1 diagrams the relationships among
containers, closures, and mortars. A fuzzy boundary exists between con-
tainers and mortars because it can be difficult to determine if a particular
vessel is an unused mortar or a container. A container does not need an-
other tool to serve its function and is therefore damaged through use dif-
ferently than a mortar. A container rim may become chipped or the basin
burned, but it does not become damaged through the use of a pestle. The
distinctive use-wear patterns created by the use of a pestle in a mortar
basin (described in Chapter 5) are the variable most useful for distin-
guishing between containers and mortars. A fuzzy boundary also exists
between palettes and lapstones, which is discussed in more detail in
Chapter 5.
BOWLS
Bowls are containers with hemispherical basins (Figure 11.2). They are
strategically designed if the exterior walls have been shaped, or their bases
flattened so that they rest securely on the ground. Their use as containers
does not necessarily create recognizable use-wear patterns, and unless
residues remain in the bowl, it may be difficult to determine their specific
use. If the contents were occasionally stirred or mixed, the resulting use-
wear can be confused with that occurring in mortars.
218
11. C ONTAINERS AND C ONTAINER C LOSURES 219
11.1. Diagram of containers and mortars expressed as sets and subsets.


lip
side basin depth

thickness
base
length
width
11.2. Plain bowl with a flat base and no use wear in the basin. (Drafted by Chip
Bowls are found throughout the U.S. Southwest but are traditionally
considered more common to the southern portion, especially decorated
bowls, but this view may be changing. Plain and knobbed bowls, called
“vessels,” have been recovered from early sites such as Cave Creek (Sayles
1945: Plate 9a,b), SU (Martin and Rinaldo 1947:112–113), and Bluff (Haury
and Sayles 1985: Figure 28a,b,e). Undecorated but carefully shaped bowls
have been recovered from Pueblo I pithouses in the Kiatuthlanna area
(Roberts 1931:156–157, Figure 34). A few decorated bowls have been re-
covered in the Mimbres area at the Swartz and Galaz ruins (Anyon and
LeBlanc 1984:281; Cosgrove and Cosgrove 1932:32). Some of these were
noted to have contained paint. The bowls recovered from Snaketown have
elaborate designs (Haury 1976:289). The most intricate designs were on
bowls recovered from Paquimé and are considered ceremonial both be-
cause of their elaborateness and their archaeological context (Di Peso et al.
1974:318). Even the bowls that are not considered ceremonial from Casas
Grandes are still well shaped (Di Peso et al. 1974:206–207). Bowls, more
often than mortars, are decorated with incised or carved designs. Rarely,
one is painted. A stone bowl with a painted decoration was recovered
from Higgins Flat Pueblo (Rinaldo 1956:82).
Small mortars and other items with small basins are sometimes de-
scribed in combination with bowls (Anyon and LeBlanc 1984:281; Rinaldo
1956:82–83). It would be useful to sort out these different types and search
for patterns in their distributions through time and across space. It may
be possible to recognize technological developments that could be behav-
iorally important.
TRAYS
Ferg (1998:573–582) summarizes what little is known about the specially

designed containers commonly referred to as trays, highlighting those that
have protrusions variously referred to as “knobs,” “ears,” “bifurcate ends,”
“tongs,” “projections,” or “extensions.” He also notes similarly shaped
trays made of clay (Ferg 1998:580). Trays are generally broader, with lower
sides and flatter bases than found on bowls (Figure 11.3). Some trays are
mortars—a conclusion supported by those found in association with tri-
angular pestles (Ferg 1998: Figure 14.11) and by compatible use-wear pat-
terns. Such use wear was found both in the tray basin and on the distal
end of pestle from Los Pozos (Adams 2000b:121, Figure 5.9). The most
important task of the classifier is to maintain consistency in deciding
whether to sort these as mortars or containers that have been secondarily
basin depth
thickness
width
base
length
nob
11.3. Tray with bifurcate knobs. (Drafted by Chip Colwell.) (Adams 1996:
Figure 13.)
used as mortars. Whatever the decision, this is a classic example of a fuzzy

set. Refer also to the discussion of mortars in Chapter 5.
CENSERS
A subset of containers (Figure 11.1), censers are generally distinguished

from bowls by their cylindrical basins and much more elaborate engraved
or carved designs (11.4). This term has not been consistently used, how-
ever, and many containers identified as effigy vessels or carved bowls
probably fit the above definition (see, for example, Gladwin et al. 1975:
Plates LVII, LVIII, LXIV). The term “censer” was applied to two stone ves-
sels associated with a palette recovered during the 1960s Snaketown ex-
cavations (Haury 1976:288–289), and it seems to have become a common
label. There is evidence that the carved containers identified as censers
from Sunset Mesa in the Tucson Basin were intentionally destroyed (Adams
2000d:152, 159). The broken censers had no evidence of burning and were
associated with structures. Only one was found with a cremation. Clarify-
ing the stone bowl/censer dichotomy would be a good research topic.
Until there is a clear definition for censers, it seems best to classify them
as bowls and describe their embellishments.
CLOSURES
The closing devices designed to seal the narrow opening of pottery and
gourd containers are classified as plugs, caps, and lids (Figure 11.1). Plugs
and caps are more or less cylindrical and fit within the diameter of the
container’s neck (Adams 1994a:142; Haury 1945:138, Figure 82, called “jar
stoppers”). Caps are distinguishable from plugs by their wide tops that
rest on the rim of the container’s neck. Plugs differ from other cylinders
by tapering almost to a point on one end and having use wear indicative
of contact with a pliable surface on that end. Those that were recovered
from sites near Flagstaff are about 7 cm long and 3 cm in diameter. Roberts
(1932:141, Plate 53) describes formally shaped caps that he labels as “stop-
pers” with necks that fit into jars or cups that were recovered from the
Village of the Great Kivas in the Zuni area. An effigy pot closed with a
scoria plug was recovered from a burial near New Caves and is now in
the possession of the Coconino National Forest Service office in Flagstaff
(Peter J. Pilles Jr., personal communication, 1999). It is possible that some
of the scoria pieces identified as cylinders were also used as plugs, but
unless a use-wear analysis identifies the distinctive patterns, they remain
unrecognized.
diameter
thickness
border width basin width

perimeter
basin depth
base
11.4. Bowl typically classified as a censer. Note the incised border designs and
shallow, straight-sided basin. (Adams 2000d: Figure 7.10.)
Lids are thin slabs of stone that rest across the tops of containers (Hayes
and Lancaster 1975:159, Figure 202; Judd 1959:139, called “jar covers”; Kid-
der 1932:75; Kidder and Guernsey 1919:125, called “pot covers”; Wood-
bury 1954:179, called “jar lids”). They are easily distinguished from plugs
and caps because they are flat and do not extend into the neck of the ves-
sel. Lids are of expedient design if appropriately sized tabular pieces are
available. Strategically designed lids are carefully ground to a specifically
sized disk. Those recovered from Badger House, Pueblo Bonito, and Pueblo
del Arroyo range in size from 5 to 18 cm in diameter and are .6 to 2 cm
thick (Hayes and Lancaster 1975:159; Judd 1954:127–128). Roberts (1940:
151) comments that thin slab lids are common throughout the San Juan
area and are usually chipped or ground on their edges. Examples of both
expediently and strategically designed lids were recovered from atop cor-
rugated vessels that were sunk into the ground. See, for example, those
from Mug House (Rohn 1971:197–198, Figures 233, 234) and Pecos Pueblo
(Kidder 1932:76). Rinaldo (1964:78–79) postulates that lids are more com-
mon among Anasazi than Mogollon assemblages. Recent excavations have
found them at non-Anasazi settlements elsewhere in Arizona (Adams
2001a; Simon and Rice 1996:563). Lids in a variety of sizes were also re-
covered from Gran Quivera, in central New Mexico (Hayes et al. 1981:126).
Lids associated with storage pits at Carter Ranch Pueblo were classi-
fied as “disks” (Rinaldo 1964:78–79). Where the association with pits is
clear, the term lid is preferable so as not to cause confusion with other items
classified as disks. Pit lids are larger and thicker than most pot lids. One
associated with a pit at a settlement near Flagstaff is 35 cm long, 32 cm
wide, and 2 cm thick.
Chapter 12
Structural Stones
T he category of structural stones includes pieces of ground stone built

into structures, as well as the slabs used to construct small features
such as hearths, bins, and altars. They are of strategic design if they have
been pecked or ground to fit a specific space or achieve a specific shape.
Some, such as thresholds and floor slabs, become worn through foot traf-
fic. Lintels, bin slabs, and rings (used to line pit openings) may not have
macroscopically visible use-wear damage. Some architectural pieces have
been decorated with petroglyphs. See, for example, those recovered from
Long House (Wheeler 1980:294, Figures 346, 347) and from Gran Quivera
(Hayes et al. 1981:136, Figure 170). Occasionally, carved pieces are included
in the wall construction. See, for example, Hooper Ranch Pueblo (Martin
et al. 1961:53–56, Figures 33–35), Badger House, Balcony House (Hayes
and Lancaster 1975:166, Figure 215), and Long House (Cattanach 1980:
Figure 17). Generally, masonry blocks are not coded as artifacts, even if
they have been pecked or ground to shape. However, observations made
in the field about the attributes related to how the stones were formed
may make it possible to recognize differing masonry techniques.
Doors, ventilators, and other openings were sometimes sealed by crude
to elaborately shaped slabs such as those recovered at Hooper Ranch Pueblo
(Martin et al. 1961:101), Higgins Flat Pueblo (Martin et al. 1956:83), Pueblo
Bonito (Judd 1954:125–126), Gran Quivera (Hayes et al. 1981:126), Prieta
Vista (Bice and Sundt 1972: Figures 54 and 55), and at the Badger House
Community (Hayes and Lancaster 1975:158–159). Those with holes of vary-
ing sizes, called “rings,” are inferred to have been collars for storage pits
or hatchways (Mindeleff 1989:192–194; Smiley 1952:56–57). Martin et al.
(1961:101) speculate that they are limited in their distribution to the upper
Little Colorado River drainage, although they have been found in the mid-
dle Little Colorado River area at the Homol’ovi sites.
Painted slabs may have been parts of altars serving as ritual architec-
tural pieces. They are thought by Di Peso (1951:63) and Smith (1952:269)
226
12. S TRUCTURAL S TONES 227
to have had a predominately western and southern distribution during

Pueblo III and Pueblo IV, and to be ancestral to Hopi and Zuni (Martin et
al. 1956:74). Loomblocks, described below, have also been recognized as
supports for altar pieces (Fewkes 1898:626).
LOOMBLOCKS
Loomblocks are large stone blocks used to anchor the wooden frame of a
loom (Woodbury 1954:153–157). Those that have been strategically de-
signed are pecked or ground to a specific shape, usually rectangular (Fig-
ure 12.1). Expediently designed loomblocks are naturally shaped stones
with the necessary holes manufactured in them but no other preparation.
Loomblocks range in size from 25 to 45 cm long, 17 to 30 cm wide, and 9
to 25 cm thick (Adams 1979:156–157; Mindeleff 1989:132). Most have one
or more circular depressions that secure the loom frame. Some have larger
depressions that have been called “handgrips” (Woodbury 1954:153–157),
but which more likely served to anchor the bottom of large wooden frames.
Stephen (1936:1185) describes the use of four loomblocks in a Hopi kiva to
stabilize the loom while the warp is wound between two beams.
Fewkes (1898:626) calls blocks of stone with handles “kiva seats” and
notes their association with kivas. These were probably loomblocks. His-
torically at Hopi, weaving is a man’s activity, and frequently looms are set
up in kivas. Looms are also set up in outdoor work spaces and in rooms
with religious significance (E. C. Adams 1982:72). Loomblocks are not the
only stone evidence for loom weaving: holes drilled in kiva floors served
to secure the lower part of the frame, which was suspended from the roof
beams (Mindeleff 1989:132–133, 212).
Loomblocks have a limited distribution through time and across space.
They have been found at Homol’ovi II, Showlow Ruin, and Pinedale
(Haury 1931a: 23, 53, Plate 18). Woodbury (1954:156) notes that they may
have originated on the Mogollon Rim during the Pueblo IV period before
they became common in the Hopi area by Pueblo V. Those recovered from
the historic village of Walpi have V-shaped grooves on their upper sur-
faces from secondary use for sharpening other tools (Adams 1979:46).
COOKING STONES
Subsumed within the generic set of cooking stones are griddles, pikistones,
stone comals, firedogs, and fire-cracked rocks. The slab-like stones, grid-
dles, pikistones, and comals provide the cooking surfaces over fires. Fire-
dogs support pots as they are heated over the fire. Fire-cracked rocks are
thickness
loom loom
socket socket
length
loom
socket
width
12.1. Loomblock with sockets for anchoring the wooden frame of an upright
loom. Note the grooves worn through concomitant secondary use to sharpen
tools. (Drafted by Rob Ciaccio.)
those pieces heated and placed either in roasting pits for cooking or in
room hearths for space heating.
Griddles
Griddles are tabular pieces of stone placed over fireplaces and used for cook-
ing various types of tortillas or cakes. Woodbury (1954:176–177) lumps
all cooking slabs in one category, but a distinction is made here between
the more expediently designed griddles and the strategically designed
pikistones. In general, griddles are designed to be thinner than pikistones,
and there is less attention to the finishing of the cooking surface. Griddles
can be recognized by their tabular shape and smoke-blackening or oxida-
tion from use over a fire.
Pikistones
Pikistones are cooking slabs placed over a formal hearth. They can be dis-
tinguished from griddles by the formal preparation of the surface and the
distinctive penetration of oil visible .2 to .7 cm into the profile of the stone.
This is a rare example of a stone tool that can still be seen in use. Ethno-
graphic descriptions of piki making and pikistone preparation help us
understand how the stones were designed, maintained, and used. Piki is
a multiple-layered roll of paper-thin corn tortillas. A piki oven is con-
structed on the floor of a living room or special room designed specifi-
cally for cooking piki. The oven is a two- or three-sided hearth (Stevenson
1904:361). The open side facilitates the constant feeding of fuel, which
keeps the fire hot enough to rapidly cook the batter. Two piki hearths are
illustrated by Mindeleff (1989: Figures 69, 70). Cushing (1920:332–333) de-
scribes the cooking of piki at Zuni. A woman squats in front of the low
piki oven to begin the process. From a nearby bowl, she scoops a handful
of watery corn batter, which is swiped across the hot stone. The batter
quickly cooks into a brittle sheet that is then laid to the side. Another two
or three sheets are cooked, rolled together, and stacked with other rolls
like cordwood.
Pikistones are highly prized commodities of modern Pueblo women
(Adams 1979:23). The stones are carefully selected and prepared, and be-
come family heirlooms (Adams 1979:23–24; O’Kane 1950:40–46; Woodbury
1954:176–177). The quarries for pikistone material are few and well guarded,
and there are ritualized methods for procuring new material (Woodbury
1954:177). Once the proper material is procured (usually by a man), there
is a long preparation process that starts with the use of a stone (either by a
man or a woman) to grind the surface smooth, and finishes with repeated
oiling (always by a woman) to create a surface smooth enough to prevent
sticking of the piki batter to the stone (Adams 1979:23; O’Kane 1950:40–
46; Stephen 1936:1197; Stevenson 1904:361–362). Cushing (1920:331) de-
scribes how pikistones are sometimes made “by subscription,” whereby a
woman who cannot obtain one otherwise hires someone to make one for
her. Either way the behavior associated with pikistones and piki making
can be considered ritual in the sense that there are beliefs that bad behav-
ior will cause the stone to break either during manufacture or use (Min-
deleff 1989:175, called “guyave stone”; Underhill 1946:91).
Historically, pikistone surfaces have been prepared with the oil from
crushed sunflower or cottonseeds (O’Kane 1950:44; Udall 1969:112–113).
Zuni and Hopi chewed squash seeds and then spit on to the hot stone
(Cushing 1920:326; Stevenson 1904:31; Underhill 1946:91) or used piñon
gum when nothing else was available. Pikistones are often treated these
days with commercial corn oil or vegetable shortening. Modern pikistones
are large: 60 to 90 cm long by 40 to 75 cm wide, and 3 to 5 cm thick. How-
ever, repeated heating and cooling causes them to become friable, with
pieces sloughing off the side closest to the fire, making them progressively
thinner. Sometimes the only part recovered from archaeological contexts
is the oil-saturated surface.
Pikistones have been recovered from thirteenth-century contexts at
the ancestral Hopi village of Homol’ovi I, and from late Pueblo III/Pueblo
IV contexts at Puerco Ruin (Burton 1990:192–194, Figures 10.4, 10.5) in
Arizona, and from fourteenth-century contexts at Leaf Water in the Chama
Valley of New Mexico (Leubben 1953:25). No pikistones have been recov-
ered from Hohokam contexts. The cooking slabs recovered from Hohokam
contexts are most often called comals and are a post-Sedentary develop-
ment. Haury (1945:109–111) suggests that this means tortilla making was
not previously common. Although Russell (1975:68) describes tortillas as
a part of Piman diets, he does not mention how they were cooked.
The cooking slabs recovered from Prieta Vista in northern New Mexico
were both carefully shaped and not (Bice and Sundt 1972:88, Figure 48).
Hayes et al. (1981:125) use the term “comale” to describe all the cooking
slabs recovered from Gran Quivera. Some are nearly circular, and these,
probably more than the others, warrant the term. The others, especially
those described as having oil-polished surfaces are perhaps pikistones.
Hayes and Lancaster (1975:159) also refer to cooking slabs associated with
pithouse firepits as comals. This term usually refers to ceramic cooking
slabs, most frequently found in Mexico. More research is needed to under-
stand if there are stone comals, or if they should be subsumed under the
term griddle.
At this point it is not clear how cooking-slab technology developed.

In some contexts, especially in southern Arizona, it appears that clay grid-
dles preceded stone ones. However, morphologically they are not very
similar: cooking slabs are generally thicker and larger than clay griddles
(Woodbury 1954:177). The information has not yet been organized well
enough to confidently discuss the temporal and spatial distributions of
the various types of cooking slabs or the development of hearth-cooking
technology.
FIREDOGS AND TRIVETS
Other stones associated with cooking are firedogs, which are rocks placed
within the hearth to support vessels during cooking (called “firestones”
by Russell [1975:111]). They are smoke-blackened and sometimes heat-
cracked. Trivets, in contrast, are rocks placed on the floor to support ves-
sels during use or storage, or to prop metates in a particular position. The
term “pot rest” seems to be used interchangeably to describe stones placed
in thermal features and those used on floors. Other terms, such as “comal
stone” or “comal supports,” are also occasionally used to categorize stones
found in thermal features (see, for example, Hayes et al. 1981:126).
Three stones found in place at Po-shu-oinga in the Chama Valley still
had the remains of a broken pot resting on them (Jeancon 1923:24, called
“andirons”). Neither firedogs nor trivets require alteration before use, and
sometimes only their positions identify them as artifacts. Strategically de-
signed firedogs and trivets are altered to make them more stable when
resting on the ground. Use wear on either may result from pots or metates
moved across them. Firedogs are thought to have been most commonly
associated with late Pueblo contexts, for example at Hooper Ranch Pueblo
(Martin et al. 1961:103), Foote Canyon (Rinaldo 1959:250–252), and Table
Rock Pueblo (Martin and Rinaldo 1960:256, Figure 74, called “pot rests”)
in Arizona, and at Gran Quivera in New Mexico (Hayes et al. 1981:126,
Figure 161). However, they are more recently recognized in other contexts;
for example, in pithouses near Flagstaff (Adams 2001a). There has not
been a systematic study of their distribution through time or across space.
FIRE-CRACKED ROCK
Fire-cracked rocks have only recently received much attention even though
they are sometimes the most numerous artifacts at a site. Not all fire-cracked
rocks were at one time ground stone, but many started out that way and
were recycled into thermal activities. Why this happened is only now be-
ginning to be researched. One conclusion is that no other stone was avail-
able for thermal activities, and thus ground stone items, especially large
metates and manos, were scavenged (Adams 1994b). One of the biggest
problems with sorting fire-cracked rocks is recognizing the attributes of
thermal alteration. Because there are many types of thermal activities, such
as stone boiling, roasting, and space heating, no single attribute distin-
guishes fire-cracked rocks.
Duncan and Doleman (1991) conducted a series of experiments to help
them learn how to recognize thermal alteration of the rocks from sites in
the Tularosa Basin of central New Mexico. They evaluated material in
terms of durability and effectiveness for thermal functions. Discolorations
and fracture patterns were the most distinctive attributes they found to
distinguish stone boiling from roasting. Their most important conclusion
was that different materials react differently to heat stress (Duncan and
Doleman 1991:342). They suggested more experimentation with the types
of materials found at specific sites. For more information, see the web
page WWW.mtsu.edu/~kesmith/TNARCHNET/pubs/fcr.html. There is
also an amazing amount of information that comes across the Internet
when searching the words “fire-cracked rock.”
It is fitting that this manual ends with a reference to the Internet: the
stone age meets the computer age. There are new and exciting tools for
building and testing models of how specific items might have been used.
Computer modeling is one tool that has yet to be applied. More work is
needed both with native groups and scientific experimentation. The exist-
ing database is huge and getting larger every day. More than ever before,
there is a need to compile these data into broad studies that document de-
velopments through time and across space, and it is hoped that this man-
ual will spark increasing interest in multiple areas of research.
Appendix A: General Artifact Form
This form records basic information about each artifact. Some artifacts
will be recorded only on this form, others have details that are recorded
on other forms.
ARTIFACT ATTRIBUTE EXPLANATIONS

Artifact Number (AN). Every artifact should be recorded on this form with a
unique artifact number.
Artifact Types (ART) and Subtypes (SUB) as defined in the manual. Codes can
be added for artifacts not described in the manual.
Artifact Condition (CND). An artifact is considered whole even if it has been

chipped. If the artifact is broken but the measurements for the whole can still
be obtained, it can be treated as a whole artifact in summary descriptions. If
fragments can be rejoined, they are counted as a single artifact.
Burned (BRN). A burned artifact may be detected by a color change created by

oxidation, smoke blackening, or the presence of carbonized residue. A heat-
cracked artifact may not be easy to identify without oxidation or blackening.
Shape (SHP) describes the general shape of the artifact. It is most useful for
remembering specific artifacts and has little analytical value.
Texture (TXT) can be standardized so that material with grains of less than 1
mm are considered fine, 1–2 mm are medium, 2–4 mm are coarse, and larger
than 4 mm are conglomerate. Material without macroscopically visible grains
has no texture. Small vesicles are less than 2 mm; large are greater than 2 mm.
Material Type (MAT) is identified by a geologist, as is the determination of the

rock type availability in relation to the site where it was recovered. The codes
for rock type are too numerous to be listed here.
Manufacturing (MAN) is an attempt to record the location and nature of

damage created by manufacturing techniques. Categories can be added as
needed. The purpose is to get the analyst to thoroughly examine the artifact
and decide if it is of expedient or strategic design.
233
234 A PPENDIX A
Design (DSN) records whether an artifact is expediently or strategically

designed. If the only manufacturing damage is on the use surface, it is
expediently designed. If there is additional shaping to make a tool comfortable
to hold, to create a shape not essential to its function, or to improve its
appearance, then it is strategically designed.
Use (USE) records primary and secondary use categories. Single-use artifacts
were used only in the activity for which they were designed. Reused tools
were used secondarily in the same general task but in a slightly different way
that did not impact tool design. Multiple-use tools were designed for use in
more than one activity at a time. Redesigned artifacts were redesigned for
secondary use and may not be usable in their primary function. Recycled
artifacts were removed from tool use altogether. More detail is provided on
these categories in the manual.
Second Use (SCN) uses the ART code to record the artifact’s second use.
Sequence (SEQ) refers to the sequence of secondary use. This is not applicable
to those items used only in the activity for which they were designed.
Sequential secondary use means the item was no longer usable in the activity
for which it was originally designed. Concomitant secondary use means that
the item was usable in both primary and secondary tasks.
Designed Activity (DCT) codes the activity for which the artifact was
designed. The artifacts described in the manual are sorted into activity
categories.
Actual Activity (ACT) codes the activity in which the artifact was actually
used. This determination involves use-wear analyses, residue analyses, and
motor-habit evaluations.
Measurements (L) (W) (T) (WT). Many of the illustrations in the manual show
the best places to take artifact measurements. Measurements should always be
taken from the same locations on each artifact type, even if on a particular
piece the greatest measurement is not the “length.”
Number of Used Surfaces (SNO) keeps track of the location and orientation of
used surfaces. This is not applicable to all artifact types.
Use Level (USL) standardizes the amount of wear the tool as a whole has
received.
Contact Type (COT) records the interpretation of the use-wear analysis. This
can be very general, such as hard, pliable, soft, etc., or more specific, such as
stone-against-stone, stone-against-hide, etc., depending on the level of
identification confidence and access to comparative collections.
Residues (RSD) helps keep track of pigment, clay, carbon, and other residues.
G ENERAL A RTIFACT F ORM 235
Color (CLR) uses the Munsell Soil Color chart to standardize color terms. Used
only to code pigment colors.
Comments (CMT) is a code that indicates whether or not there are additional
notes about the artifact. Because coding forms are incapable of covering all
possibilities, comments should be written about each artifact. Comments need
not duplicate the data recorded on the forms, but can explain minor variations
or expand on unusual attributes.
Table A.1 is an example of a general artifact form, and Table A.2 lists the
codes used on the form.
Table A.1. Ground Stone Analysis Form:
Date ____ /____ /____ Site Numbers __________ /__________
AN ART SUB CND BRN SHP TXT MAT MAN DSN USE SCN SEQ
General Artifact Form
Feature __________________________
DCT ACT L W T WT SNO USL COT RSD CLR CMT

238 A PPENDIX A
Table A.2
Coding for General Artifact Form
Artifact Number (AN)
Artifact Type (ART)
Abrading, Smoothing, Polishing Tools

5. abrader; 15. lapstone; 10. netherstone; 4. polishing stone.
Grinding and Pulverizing Tools

56. crusher/fergolith; 11. grinding slab; 1. handstone; 15. lapstone; 2. mano;
80. mano/metate; 12. metate; 13. mortar; 10. netherstone; 14. palette; 3. pestle.
Percussion Tools
16. anvil; 62. chopper; 18. cupped stone; 81. hammerstone; 9. pecking stone;
8. pottery anvil; 82. pulping stone.
Hafted Percussion Tools

28. adze; 20. axe; 22. hoe; 21. maul; 27. pick.
Spinning Tools
19. fire-drill hearth; 25. spindle base; 23. whorl.
Slicing, Scraping, and Perforating Tools

123. awl/pin; 111. file; 61. plane; 60. reamer; 110. saw; 6. tabular tool.
Paraphernalia
48. ball; 71. bell stone; 126. cruciform; 57. cylinder; 121. disk; 50. figurine;
2. geometric; 54. grooved stone; 70. lightning stone; 66. loop; 44. medicine
stone; 98. natural shape; 124. offering; 49. ornament; 14. palette; 55. pigment
source; 46. pipe; 45. plummet/weight; 122. shaped; 47. tube.
Containers and Container Parts

30. bowl; 118. cap; 31. censer; 14. palette; 119. pit lid; 120. pot lid; 125. plug;
32. tray; 34. vessel.
Structural
51. architectural; 105. firedog/trivet; 40. cooking slab; 41. griddle;
43. loomblock; 42. pikistone;
Other Codes
99. not an artifact; 100. fire-cracked rock; 75. unidentified.
Table A.2
Coding for General Artifact Form (continued)
Artifact Subtype (SUB)
Abraders
30. flat; 36. flat & single V; 37. flat & single U; 38. flat & multiple V; 39. flat &
multiple U; 40. flat & both V & U; 33. multiple V; 34. multiple U; 35. both 1V &
1U; 31. single V; 32. single U.
Architectural
152. bin stone; 150. building stone; 151. hearth stone; 154. lintel; 153. ring;
156. splash stone; 155. threshold.
Awls
43. flat; 46. hairpin; 42. needle (no head); 41. pin (headed); 4. indeterminate.
Axe Heads
87. blank; 81. full groove; 99. full-double bit; 86. incomplete groove;
80. notched; 82. 3/4 groove; 89. 3/4-double bit; 88. 3/4 and wedge groove;
83. 5/8 groove; 84. spiral groove; 85. regrooved; 4. indeterminate.
Balls
121. 1 flat side; 122. 2 flat sides; 123. irregular; 120. spherical.
Choppers
240. expedient; 241. hand axe.
Containers
92. effigy-flat bottom; 93. effigy-round bottom; 96. incised-flat bottom;
97. incised-round bottom 90. plain-flat bottom; 91. plain-round bottom;
94. shaped-flat bottom; 95. shaped-round bottom; 98. tray-plain; 99. tray-
bifurcate; 4. indeterminate.
Cylinders/Plummets/Medicine Stones
118. bilobed; 111. conical; 114. conical & groove; 117. conical & head;
110. cylindrical; 113. cylindrical & groove; 116. cylindrical & head;
119. geometric; 112. parabolic; 115. parabolic & groove.
Disks/Whorls
131. whorl-concave; 130. whorl-flat; 138. disk-unperforated; 134. donut-
biconcave; 136. donut-basin (incomplete perforation); 137. donut-biconvex;
133. donut-concave; 132. donut-flat; 135. donut-unperforated; 139 ring;
4. indeterminate.
Figurines
142. animal; 143. animal part; 140. human; 141. human part; 144. natural;
4. indeterminate.
240 A PPENDIX A
Table A.2
Handstones/Manos
1. basin; 5. blank; 3. flat; 7. flat/concave; 225. hide processing; 6. multiple;
226. polishing; 2. trough; 9. other.
Mortars
77. blank; 72. boulder; 73. bowl; 76. disk; 70. pebble; 71. rock; 74. shaped-
anthropomorphic; 78. shaped-zoomorphic; 401. tray-bifurcate; 400. tray-plain;
4. indeterminate.
Natural
142 animal; 120 spherical; 4. indeterminate.
Netherstones/Metates/Grinding Slabs/Lapstones
50. basin; 160. basin-open; 162. basin-3/4; 172. blank; 55. flat; 58. flat/concave;
53. trough-closed; 51. trough-open; 52. trough-3/4; 54. trough-Utah; 59. trough-
indeterminable; 4. indeterminate.
Palettes
62. anthropomorphic; 64. blank; 61. raised border; 60. flat border;
63. zoomorphic.
Personal Ornaments
304. bead-barrel; 308. bead-bilobe; 305. bead blank; 315. bead-convex;
309. bead-cuboid; 302. bead disk; 306. bead-irregular; 316. bead-plano-convex;
307. bead-teardrop; 300. bead-tube; 303. bead-zoomorphic; 350. blank;
325. bracelet-C; 320. bracelet-ring; 331. button; 340. figurine-2 dimensional;
332. plug; 341. figurine-3 dimensional; 34. geometric; 330. mosaic tesserae;
370. necklace; 313. pendant-blank; 314. pendant-inlay; 310. pendant-2-
dimensional; 311. pendant-3-dimensional; 326. ring-C; 321. ring-finger;
333. toggle; 360. whizzer; 312. zoomorphic inlay; 4. indeterminate.
Pestles
400. blank; 17. block; 16. cobble; 10. conical; 12. cylindrical; 19. natural;
11. parabolic; 14. pebble; 18. shaped; 13. triangular; 4. indeterminate.
Pipes/Tubes
104. conical-biconical hole; 100. conical-conical hole; 101. conical-cylindrical
hole; 105. cylindrical-biconical hole; 102. cylindrical-conical hole;
103. cylindrical-cylindrical hole; 107. socketed; 106. elbow; 108. globular;
109. other; 4. indeterminate.
Polishing Stones
26. disk; 27. faceted; 23. floor; 24. handstone; 20. pebble-surface; 21. pebble-
edge; 22. pebble; 4. indeterminate.
Table A.2
Pottery Anvils
250. grooved; 251. handled; 252. plain.
Planes
221. backed; 220. hafted.
Tabular Tools
207. edge and surface; 214. hafted; 200. 1 straight edge; 204. 1 concave
edge; 202. 1 convex edge; 209. 1 irregular edge; 203. > 1 convex edge;
205. > 1 concave edge; 210. > 1 irregular edge; 201. > 1 straight edge;
206. multiple edges; 213. multiple surfaces; 216. notched; 215. shaped but
not used; 212. too fragmentary; 211. unused material.
Artifact Condition (CND)

1. whole; 7. conjoined fragments; 2. > 1/2; 3. < 1/2; 4. indeterminate; 8. reused
fragment; 6. measurable; 5. reconstruct whole.
Burned (BRN)
4. after use total; 5. after use partial; 3. before use; 7. before and after use;
8. before second use; 2. from use; 6. heat cracked; 9. indeterminate; 1. no.
Shape (SHP)
11. bilobe; 17. broken; 7. conical; 15. crescent; 6. cylindrical; 13. diamond;
8. disk; 9. donut; 16. irregular; 18. morphic; 3. ovoid; 12. pebble/cobble;
1. rectangular; 10. ring; 19. round; 14. semicircular; 20. slab; 4. spherical;
2. square; 5. triangular.
Texture (TXT)
2. coarse; 6. coarse and fine; 5. coarse and medium; 1. conglomerate; 4. fine;
7. fine and medium; 3. medium; 11. no texture; 8. vesicles-large; 9. vesicles-
small; 10. vesicles-large and small.
Manufacturing (MAN)
10. carved; 16. chipped; 15. chipped and ground; 11. chipped for hafting;
3. ground; 17. ground and incised; 19. ground edge only; 5. ground perimeter;
24. ground surface only; 1. natural; 8. pecked, ground, and polished; 2. pecked
to hold; 4. pecked and ground; 2. pecked edge only; 23. ground for stability;
20. pecked perimeter; 18. pecked for stability; 14. pecked to hold; 13. pecked
surface only; 21. pecked surface and to hold; 7. pecked and polished;
6. polished; 12. indeterminate.
Design (DSN)
1. expedient; 2. strategic; 3. indeterminate; 4. not applicable; 5. incomplete.
Use (USE)
8. destroyed; 3. multiple use; 9. offering; 2. reused; 5. recycled; 4. redesigned;
1. single; 7. unused; 6. indeterminate.
242 A PPENDIX A
Table A.2
Second Use (SCN)

Insert artifact type code.
Sequence (SEQ)
1. sequential; 2. concomitant; 4. both; 3. not applicable; 5. indeterminate.
Designed Activity (DCT)

1. food processing; 2. general processing; 3 percussion; 4. abrading; 5. polishing/
abrading; 6. cutting/scraping/perforating; 7. ornamentation; 8. representation;
9. paraphernalia; 10. structural; 11. specimens (samples); 12. indeterminate.
Actual Activity (ACT)

1. food processing; 2. general processing-pigment; 3. general processing-other;
4. manufacture-pottery; 5. manufacture-stone; 6. manufacture-wood/bone;
7. manufacture-other; 8. procurement; 9. ritual/symbolic/decorative;
10. multiple; 11. other; 12. unused. 13. indeterminate.
Length (L)
Width (W)
Thickness (T)
Weight (WT)
Number of Used Surfaces (SNO)

1. 1; 2. 2 opposite; 3. 2 adjacent; 4. 3 opposite and adjacent; 5. 4–2 adjacent
each side; 6. multiple surfaces; 11. corner; 9. 1 edge; 10. edge and corner;
12. multiple edges; 8. indeterminate; 7. not applicable.
Use Level (USL)

1. light; 2. moderate; 3. heavy; 4. nearly worn out; 5. worn out; 6. not
applicable; 7. indeterminate; 8. unused.
Contact Type (COT)

3. bone; 11. hard; 4. hide; 8. indeterminate; 7. multiple; 6. other; 9. pliable;
5. pottery; 12. resilient; 13. soil; 10. soft; 1. stone; 2. wood.
Residues (RSD)
7. caliche; 6. carbon; 2. clay; 4. indeterminate; 5. none; 1. organic; 11. other;
3. pigment; 8. pigment and caliche.
Color (CLR)
1. 10R4/6; 2. 10R4/8; etc.
Comments (CMT)
1. yes; 2. no.
Appendix B: Handstone Form
This form records details about handstones, including manos, abraders,

polishing stones, and any stone that is held in the hand during operation.
Table B.1 is an example of a form for recording handstone attributes. Table
B.2 lists the codes used on the form.

Artifact Number (AN) from the General Artifact Form.
Artifact Type (ART) and Artifact Subtype (SUB) from the General Artifact
Form.
Surface Number (SNUM) keeps track of which surface or edge is being

recorded. The largest or most worn surface or edge is recorded as number 1.
The written comments kept about each artifact can help keep track of the
locations of surface or edge numbers.
Grips/Grooves (GR) keeps track of the nature and location of finger grips,
grooves, notches, handles, and other methods of holding artifacts. Grips are
roughened areas, grooves have depth, noteches are not as long as grooves.
Wear (SWEAR) records the condition of wear on each surface or edge. A

separate line on each form should be filled out for each used surface or edge
used. If wear is barely visible on the surface, it is light. If it is easy to see but
does not alter artifact shape, it is moderate. If the wear alters the artifact shape,
it is heavy.
Surface Configuration (SCON) records the general shape of the surface or

edge. For example, some surfaces are flat from end-to-end and convex edge-to-
edge. Abrader grooves may have been worn in a uniform manner or they may
have been used in such a way that they are deeper toward the end. Some edges
are serrated.
Surface Texture (STEXT) records the nature of the use surface. A coarse-grain
material can be worn smooth and then resharpened by pecking to restore the
roughness of the stone. This attribute helps the analyst decide the nature of the
contact surface and assess the use of wear-management strategies.
243
Table B.1. Ground Stone Analysis Form:
Date ____ /____ /____ Site Numbers __________ /__________
AN ART SUB SNUM GR SWEAR SCON STEXT

Handstone Form
Feature __________________________
WRLV WRTP COT STRK COMPAT RSD SL SW

246 A PPENDIX B
Wear Level (WRLV) is an assessment combining both macroscopic and

microscopic observations. Like surface texture, this attribute helps the analysts
evaluate the nature of the contact surface. The Chapter 2 discussion of use-
wear analysis explains why these are important attributes to observe.
Wear Type (WRTP) is another assessment that relies on both macroscopic and
microscopic observations. It records the nature of the damage from the
mechanisms described in Chapter 2.
Contact Type (COT) is the same code as on the General Artifact Form if the
artifact was only used in one activity. If the artifact was secondarily used, this
code will be different when it is recording the surface or edge that was used in
the secondary activity.
Stroke (STRK) records the type of motor habits used with the artifact.
Reciprocal strokes move the tool primarily in back-and-forth motions. Circular
strokes move the tool in rotational motions around a surface. Flat strokes
maintain the tool in full contact with the opposing surface at all times. Rocking
strokes lift either the proximal or distal edge away from the opposing surface
at some point in the motion. Crushing motions use pressure and the weight of
the stone; pounding motions are more forceful, involving more muscle and lift
of the stone. Pecking strokes are at more of an angle than crushing or
pounding strokes.
Compatible (COMPAT) records the AN of any other artifact that might have
been used with the artifact being recorded. For example, some manos are
compatible with certain metates and not others. Compatible means that the
size, configuration, and use wear match on both tools.
Residue (RSD) records the nature of visible residues. Some residues are
probably remnant from burial. Others may be related to use, and this code
allows for tracking items that may require further investigation. This variable
is also recorded on the General Artifact Form, but refers here to the specific
surface.
Length (SL) and Width (SW) are two measurements taken on the surface so
that we can understand how much was actually involved in wear.
H ANDSTONE F ORM 247
Table B.2
Coding for Handstone Form
Unique Artifact Number (AN) from the General Artifact Form.
Artifact Type (ART) and Subtype (SUB) from the General Artifact Form.
Surface Number (SNUM)

Number of the surface for which attributes are being recorded.
Grips/Grooves (GR)
1. no; 3. grip-1 edge; 7. grip-2 edges; 2. groove -1 edge; 13. groove-2 edges;
11. groove-encircling; 12. ground to fit hand; 8. handle; 9. notched; 4. too worn;
10. wear only; 5. not applicable; 6. indeterminate.
Surface Wear (SWEAR)

1. light; 2. moderate; 3. heavy; 4. unused; 5. indeterminate; 6. not applicable.
Surface Configuration (SCON)

15. basin; 9. concave all over; 8. concave-end-to-end, flat-edge-to-edge;
4. convex all over; 13. edge; 14. edge-serrated; 1. flat all over; 3. flat-edge-to-
edge and convex-end-to-end; 2. flat-end-to-end, convex-edge-to-edge;
11. groove-not to end; 5. groove-uniform; 6. groove-worn end; 7. irregular;
10. variable; 12. indeterminate.
Surface Texture (STEXT)

4. coarse; 2. fine; 3. medium; 5. mixed; 6. resharpened; 7. resharpened-worn;
10. rough from manufacture; 1. smooth; 8. not applicable; 9. indeterminate.
Wear Level (WRLV)

1. highs only; 2. highs and lows; 3. smooth spots; 4. smooth all over;
5. indeterminate; 6. unused.
Wear Type (WRTP)

1. abrasion; 5. abrasion and impact fractures; 15. abrasion, impact fractures, and
chips; 22. abrasion and rounding; 7. abrasion and sheen; 3. chips; 8.chips and
impact fractures; 23. chips and rounding; 10. chips and sheen; 2. impact
fractures; 9. impact fractures and sheen; 17. impact fractures and rounding;
12. manufacture only; 11. multiple; 13. rounding; 18. rounding and sheen;
4. sheen; 19. indeterminate.
Contact Type (COT)

7. multiple; 3. bone; 11. hard; 4. hide; 6. other; 9. pliable; 5. pottery;
12. resilient; 10. soft; 13. soil; 1. stone; 2. wood; 8. indeterminate.
248 A PPENDIX B
Table B.2
Coding for Handstone Form (continued)
Stroke (STRK)
3. circular-flat; 4. circular-rocking; 1. reciprocal-flat; 2. reciprocal-rocking;
5. combination-flat; 6. combination-rocking; 13. chopping; 16. crushing;
19. cutting; 14. multiple; 18. scraping; 9. shaving/slicing; 15. stirring; 17. stirring
and crushing; 7. pecking; 8. pounding; 12. not applicable; 11. indeterminate.
Compatible (COMPAT)
Artifact number of any compatible artifact.
Residues (RSD)
7. caliche; 6. carbon; 2. clay; 6. multiple; 5. none; 1. organic; 3. pigment;
8. pigment and indeterminate; 4. indeterminate.
Surface Length (SL)

Surface Width (SW)
Appendix C: Netherstone Form
This form records details about netherstones, lapstones, metates, anvils,

bowls, mortars, and so on. Table C.1 is an example of a form for recording
netherstone attributes. Table C.2 lists the codes used on the form.

Surface Number (SNUM) records the number of the surface for which
attributes are being recorded. Generally the largest or most heavily used
surface is recorded first.
Surface Coverage (SCOV) records the extent and nature of the surface. In some
cases, this helps the analyst recognize the size and configuration of the
handstone or other artifact used with the netherstone.
Wear (SWEAR) records the wear on each surface. A separate line on each form
should be filled out for each used surface. If wear is barely visible on the
surface, it is light. If it is easy to see but does not alter artifact shape, it is
moderate. If the wear alters the artifact shape, it is heavy.
Surface Configuration (SCON) records the general shape of the surface. For
example, some surfaces are flat end-to-end and concave edge-to-edge. This
helps understand the nature of the handstone used against the netherstone.
Surface Texture (STEXT) records the nature of the use surface. A coarse-grain
material can be worn smooth and then resharpened by pecking to restore the
roughness of the stone. This attribute helps the analyst decide the nature of the
contact surface and assess the use of wear-management strategies.
Surface Manufacture (SMAN) records the nature of visible damage created by

the manufacturing techniques used to shape the surface. If use has obliterated
all evidence, then record as indeterminate.
Stroke (STRK) records the general nature of the motor habit used with the
netherstone. Reciprocal strokes would have moved across the surface in a
back-and-forth motion. Circular strokes would have moved around the stone.
249
Table C.1. Ground Stone Analysis:
Date ____ /____ /____ Site Number __________ /__________
AN ART SUB SNUM SCOV SWEAR SCON STEXT

Netherstone Form
Feature __________________________
SMAN STRK COMPAT RSD SL SW SD BW

252 A PPENDIX C
Compatible (COMPT) records the number of any other artifact that might have
been used with the artifact being recorded. For example, some manos are
compatible with certain metates and not others.
Residue (RSD) records the nature of visible residues. Some residues are
probably remnant from burial. Others may be related to use, and this code
allows for tracking items that may require further investigation. This variable
is also recorded on the General Artifact Form, but refers here to the specific
surface.
Surface Length (SL), Width (SW), and Depth (SD) record the dimensions of the
used surface.
Border or Rim Width (BW) is recorded if there is one. Select a spot that looks
typical.
N ETHERSTONE F ORM 253
Table C.2
Coding for Netherstones Form
Unique Artifact Number (AN) from the General Artifact Form.
Artifact Type (ART) and Artifact Subtype (SUB) from the General Artifact Form.
Surface number (SNUM).
Surface Coverage (SCOV)

2. border-flat; 3. border-raised; 1. complete; 6. incomplete; 4. indeterminate;
5. not applicable.
Wear (SWEAR)
1. light; 2. moderate; 3. heavy; 4. unused; 5. indeterminate; 6. not applicable;
7. killed.
Surface Configuration (SCON)

10. basin; 7. convex; 4. concave; 1. flat all over; 3. flat-edge-to-edge;
concave-end-to-end; 2. flat-end-to-end; concave-edge-to-edge; 5. irregular;
6. indeterminate.
Surface Texture (STEXT)

4. coarse; 2. fine; 3. medium; 5. mixed; 6. resharpened; 7. resharpened-worn;
10. rough from manufacture; 1. smooth; 8. not applicable; 9. indeterminate.
Surface Manufacture (SMAN)

6. natural; 5. ground to shape; 1. pecked to shape; 7. pecked and ground to
shape; 2. worn to shape; 9. worn and pecked to shape; 8. worn and ground to
shape; 3. combination; 6. incomplete; 4. indeterminate.
Stroke (STRK)
2. circular; 1. reciprocal; 7. pecking; 3. combination; 5. grinding and pecking;
6. not applicable; 4. indeterminate.
Compatible (COMPT)
Artifact number of compatible tools.
Residue (RSD)
7. caliche; 6. carbon; 2. clay; 1. organic; 3. pigment; 8. pigment and caliche;
4. indeterminate; 5. none.
Surface Length (SL)

Surface Width (SW)
Surface Depth (SD)
Border or Rim Width (BW)

Appendix D: Hafted Tool Form
This form records the details about all tools designed for hafting. Table D.1
is an example of a form for recording hafted tool attributes. Table D.2 lists
the codes used on the form.

Artifact Type (ARTP) and Subtype (SUBT) from the General Artifact Form.
Measurement Points: See Figure 7.3 for an illustration of where to take

measurements.
Bit Length (BITL)
Bit Width (BITW)
Bit Thickness (BITH)
Bit Edge Length (BEDL)
Bit Edge Width (BEDW)
Poll Length (POLL)
Poll Width (POLW)
Poll Thickness (POLTH)
Groove Width (GRVW)
Groove Depth (GRVDP)
The measurements should be taken in the same place on all axe heads. For
measurements on mauls or other tools that do not have a distinct bit and poll,
record the largest side of the hafting feature as the bit, and the smallest as the
poll. Substitute notch measurements for grooves with notched hafting features.
Bit Edge Shape (BITSH) records whether the bit edge has been resharpened or
not. A new, unresharpened bit has the profile of an isosceles triangle.
Resharpening usually creates an off-center edge.
Bit Edge Damage (BDAM) records the nature of the damage to the bit edge. Do
not confuse manufacturing damage with use wear.
255
Table D.1. Ground Stone Analysis:
Date ____ /____ /____ Site Number __________ /__________
AN ART SUB BITL BITW BITH BEDL BEDW

Hafted Tool Form
Feature __________________________
POLL POLW POLTH GRVW GRVDP BITSH BDAM BDSHR USBL

258 A PPENDIX D
Bit Edge Sharpness (BDSHRP) records the nature of the bit edge. The edge is
considered sharp if it is less than 2 mm thick, dull if it is 2 mm to 1 cm thick,
rounded if thicker than 1 cm, and flattened if there is no curvature to the edge.
Usable (USBL) is an interpretation of how usable the axe is currently. Usable

means that the edge is sharp and has not been resharpened (similar to Figure
7.4a). Usable with resharpening means that the edge is dull and there is no
prior resharpening (similar to Figure 7.4b). Resharpened–usable means that
the bit has been resharpened and remains sharp (similar to Figure 7.4c).
Resharpened–dull means that the bit has been resharpened but is now dull
(similar to Figure 7.4d). Not usable means that the bit is no longer usable for
chopping and there is no evidence of resharpening (similar to Figure 7.4e).
Resharpened–not usable means the bit has been resharpened but is no longer
useable. Axes become unusable when the bit is either worn or broken to the
point that it can not be resharpened.
H AFTED T OOL F ORM 259
Table D.2
Coding for Hafted Tool Form
Artifact Type (ART) and Artifact Subtype (SUB) from the General Artifact Form
Bit Length (BITL)
Bit Width (BITW)
Bit Thickness (BITH)
Poll Length (POLL)
Poll Width (POLW)
Poll Thickness (POLTH)
Groove Width (GRVW)
Groove Depth (GRVDP)
Bit Edge Length (BEDL)
Bit Edge Width (BEDW)
Bit Edge Shape (BITSH)

1. original; 2. resharpened; 3. indeterminate; 4. incomplete.
Bit Edge Damage (BDAM)

1. none; 4. abraded; 7. abraded and chipped; 9. abraded, chipped, and sheen;
8. chipped and sheen 3. battered; 6. battered and chipped; 5. multiple.
Bit Edge Sharpness (BDSHRP)

1. sharp; 2. dull; 3. rounded; 4. flattened; 5. broken; 6. unfinished.
Usable (USBL)
1. usable; 2. usable with resharpening; 3. resharpened-usable; 4. resharpened-
dull; 5. not usable; 6. resharpened-not usable.
Appendix E: Grooved Artifacts Form
This form records details about all grooved artifacts.
ARTIFACT ATTRIBUTES EXPLANATION

Embellishment (EMB) records the nature of any embellishment. Embellish-

ments include incised lines, raised ridges, and decorative grooves. These can
occur singly or in multiples, and can be mixed such that incised lines and
raised ridges both occur on one tool.
Groove Configuration (GRVCF) records whether the groove is uniform across

its length or if there is more wear toward the end.
Groove Orientation (GRVO) records whether the groove is oriented along the
length, width, or diagonally across the stone. Each groove should be recorded
on a separate line.
Groove Length (GRL)

Groove Width (GRW)
Groove Depth (GRDP)
Groove dimensions should be taken at their maximum.
Table E.1 is an example of a form for recording the attributes of grooved

artifacts. Table E.2 lists the codes used on the form.
261
Table E.1. Ground Stone Analysis: Grooved Artifact Form
Date ____ /____ /____ Site Number ________ /________ Feature _____________________
AN ART SUB EMB GRVCF GRVO GRVL GRVW GRVDP

G ROOVED A RTIFACT F ORM 263
Table E.2
Coding for Grooved Artifacts Form
Embellishment (EMB)
4. incised line; 3. incised lines perpendicular to grooves; 1. one ridge
perpendicular to grooves; 7. multiple ridges perpendicular to grooves;
2. multiple ridges and incised lines perpendicular to grooves; 6. other; 5. none.
Groove Configuration (GRVCF)

1. groove uniform; 2. groove worn one end; 3. groove worn both ends.
Groove Orientation (GRVO)

1. lengthwise; 2. widthwise; 3. diagonal; 4. mixed.
Groove Length (GRL)

Groove Width (GRW)
Groove Depth (GRDP)
Appendix F: Perforated Artifact Form
This form records details about all perforated items.

Inside Hole Diameter (IHD)
Outside Hole Diameter (OHD)
Measurements should be taken at their maximum.
Hole Type (HTYP) records the hole configuration. Biconical means the outside
hole diameters on both ends of the hole are larger than the inside hole
diameter. Conical means that one outside hole diameter is larger than the
inside and the other outside diameters. Cylindrical holes have the same
diameter throughout the hole.
Table F.1 is an example of a form for recording attributes of perforated ar-

tifacts. Table F.2 lists the codes used on the form.
265
Table F.1. Ground Stone Analysis: Perforated Artifact Form
Date ____ /____ /____ Site Number ________ /________ Feature _____________________
AN ART SUB IHD OHD HTYP

P ERFORATED A RTIFACT F ORRM 267
Table F.2
Coding for Perforated Artifacts Form
Inside Hole Diameter (IHD).
Outside Hole Diameter (OHD).
Hole Type (HTYP)

2. biconical; 6. broken; 1. conical; 7. conical-incomplete; 9. cylindrical
3. incomplete; 5. natural; 8. remodeled; 4. not applicable.
Glossary
This glossary defines the terms and concepts used throughout the manual.
Artifact types, however, are not defined here because they are described
in Part 2.
Abrasive wear is caused by the movement of a rougher, more durable surface

across a smoother, softer surface. Movement of the harder, rougher surface digs
into the smoother material of the opposing surface, displacing the softer mate-
rial in the direction of the movement. The most obvious results are scratches
called striations.
Adhesive wear occurs when two surfaces come into contact, and even if there is
no movement, there are molecular interactions. These interactions create bonds
that are broken when there is movement of one surface across or away from the
other surface.
An analytical construct uses the language of description and quantification; for

example, the handstone is 15 cm long; weighs 340 grams; has multiple, unidirec-
tional striations; and has red pigment on the grinding surface. Contrast with be-
havioral construct.
Analyzing an artifact is the act of examining a complex item and, on the basis of
the relationship of its individual elements, deciding to which category or cate-
gories it belongs.
Asperity is a tribological concept used to discuss surface texture. For ground

stone surfaces, asperity is a combination of material granularity and surface tex-
ture, and is influenced by material durability. For example, the surfaces of tools
made from coarse-grain material naturally have more asperity, or are more as-
perite, than the surfaces of tools made from fine-grain material. If the coarse-
grain material is more durable than the fine-grain material, and they both be-
come worn smooth, the asperity of the coarse-grain surface will be less than
that of the fine-grain material. This is because the grains of the more durable
material remain in place and become worn level with the matrix, whereas the
wearing of the less durable material either removes grains from the matrix, or
the softer material is more intensely abraded. Either way, the surface asperity is
greater than that of the more durable material. The surface of the coarse-grain
tool can be made more asperite by pecking it to resharpen the surface texture.
269
270 G LOSSARY
An asperity is a single grain or a single projection from a surface.
A behavioral construct is inferential and uses the language of action and people;
for example, the stone was used to grind pigment with a reciprocal motion
against a rough netherstone. Contrast with analytical construct.
Blanks are items that made it through the initial stages of manufacture and have
some evidence of being shaped, but not enough to recognize the intended tool
type. Contrast with unused items.
Chopping describes the stroke used with tools that have sharp edges. The stroke
chips away material from the contact surface. Contrast with pecking.
A circular stroke arcs the tool across the contact surface, causing striations to
form in multiple directions on both surfaces. Contrast with reciprocal stroke.
Classifying artifacts is the process of creating categories and sorting items into
them.
Comfort features are manufactured elements that include specific shapes, handles,
grooves, or other attributes that make the tool comfortable to hold.
Concomitant secondary use allows an item to function in a second activity without

destroying the attributes of its primary use. Manos also used to grind pigment,
and polishing stones with shaft-straightening grooves on the opposite surface
are examples of concomitant secondary use. Contrast with sequential secondary
use.
Contact surface is the surface against which a tool is worked.
Crushing describes the stroke that uses the weight of the stone to reduce the in-
termediate substance. Contrast with pounding.
The distal edge is the one most distant from the user. Contrast with proximal edge.
The dorsal surface or side is the upper side of the tool. For manos and most hand-
stones, this is the side held in the hand. For grooved abraders, this is the side
with the grooves, and the ventral side is in the hand or on the ground. Contrast
with ventral surface or side.
Durability is the ability to withstand wear.
The concept of efficiency is best evaluated in quantifiable terms as the amount of

product that can be processed in a standardized amount of time, or the number
of hours spent at completing a particular task.
An expedient design is one where the natural shape of the rock was altered only
through use. Contrast with strategic design.
G LOSSARY 271
An extensively used tool was worked over a long period of time; for example, 50
years instead of 1 year. Contrast with intensively used tool.
Factors are the variables that can be independently changed in experimental sit-
uations, such as selecting either coarse-grain or fine-grain material for tools, or
selecting oily or dried seeds to grind. Contrast with response variables.
Fatigue wear occurs as pressure or the alternating stress of movement is applied

to contacting surfaces. The highest topographic elevations bear the weight and
mass of the load. If the load is more than is bearable, there is collapse and crush-
ing of the elevations.
A flat stroke maintains contact between tool and the contact surface at all times;
the edges are not lifted. Contrast with rocking stroke.
Granularity is a measurable analytical construct of the stone described as fine-

grain, coarse-grain, and so on. A stone’s granularity is determined when it is
formed. Contrast with texture.
Hard contact surfaces engage only the highest grains and topographic elevations.
For example, when two metamorphosed stones come into contact, their surfaces
are so hard that there is not enough give between them to allow material from
one to be pushed into the low parts of the other. Contrast with soft, resilient, and
pliable contact surfaces.
Heavy wear changes the natural or manufactured shape of the tool.
An intensively used tool was worked during tasks of long duration; for example,
five hours a day instead of one hour a day. Contrast with extensively used tool.
Interstices are the spaces between grains.
Light wear leaves so little evidence that it can barely be seen with the unaided
eye.
The term margin is used to identify both the edges between vesicles, and the
edges of basins and troughs. It refers to natural, unmodified portions of the
stone. The area around basins or troughs that have been modified are called
rims or borders.
Matrix is the stone material that holds the grains of sedimentary and metamor-
phic rocks in place.
Microtopography distinguishes the topographic variation visible under magnifi-

cation from the topography visible with the unaided eye.
Moderate wear is enough to leave obvious damage without altering the basic
shape of the tool.
272 G LOSSARY
A multiple-use tool is designed for a specific primary activity, but another area or
surface is also used in a second activity. Use in one activity does not inhibit use
in the other, even if the tool’s configuration is slightly altered.
Nearly worn out tools have been used so much that they are difficult to hold for
continued use, or the usable surface or edge is almost gone.
Pecking describes a stroke used with a tool that has a rounded or convex surface,
such as a hammerstone. It is a short-distance stroke used with less force than a
pounding stroke. This stroke alters the configuration of the contact surface by
creating small impact fractures that are also sometimes called pecking or pecked
features. Contrast with chopping and pounding.
Pliable surfaces are more rigid than soft ones and do not extend as deeply into
the lows and interstices. Contrast with hard, soft, and resilient contact surfaces.
Pounding describes a stroke that uses muscular force to propel the weight of a
stone tool against the contact surface. Contrast with crushing.
Primary use is that for which the item was originally designed. Contrast with
secondary use.
The proximal edge is closest to the user. Contrast with distal edge.
A reciprocal stroke moves the tool back and forth across the surface, resulting in
unidirectional striations on both surfaces. Contrast with circular stroke.
Recycled items are designed and used in one activity, but ultimately employed in
a completely different context that may or may not physically alter the item,
such as a metate used as a building stone.
Redesigned tools are designed for a primary activity and either remanufactured
for, or altered through use in, a second activity to the extent that the item no
longer functions in the first, such as a mano redesigned with a groove across the
working surface.
Resilient contact surfaces have some give and may extend into the topographic
lows, but they are not soft enough to extend deep into the interstices. For exam-
ple, when a stone comes into contact with dried pottery, the surface has enough
give to round the edges of grains and the margins of vesicles, but not enough to
reach between grains or into the vesicles. Contrast with hard, soft, and pliable
contact surfaces.
Response variables are those attributes that change in response to independent

factors; they are indicators of a process. They are the attributes that are meas-
ured, described, compared, and analyzed during experiments. Contrast with
factors.
G LOSSARY 273
A reused item is designed for a specific primary use, but becomes employed in a
second activity without altering the design, such as a food-processing mano
also used to process pigment.
A rocking stroke lifts the edges of the tool away from the contact surface. A circu-
lar rocking stroke lifts alternating edges as the tool arcs around the surface. A re-
ciprocal rocking stroke alternately lifts the proximal and distal edges as the tool
moves back and forth over the surface.
Secondary use is usually a later addition to that for which a tool was originally
designed. Contrast with primary use.
Sequential secondary use precludes an item’s ability to function in its primary use.
Tools recycled into roasting pits or redesigned into other tool types are examples
of sequential secondary use. Contrast with concomitant secondary use.
A single-use item is employed only in the activity for which it is designed and
seemingly has only one function.
A soft contact surface engages the stone’s working surface more three-dimension-
ally than a hard contact surface. The soft surface can move into the interstices
and topographic lows that are unreachable by hard contact surfaces. Hides and
hands are examples of soft contact surfaces. Contrast with pliable and resilient
contact surfaces.
A strategic design is one where modifications create a specific shape or make the
item easier to hold. Contrast with expedient design.
Striations are the scratches caused by abrasive wear. Deep striations are some-
times called gouges.
Texture is a relational construct that describes a tool’s surface. It is tactile and

perceptible as smooth, rough, and so on. It is independent of material granularity
such that a coarse-gain rock can become worn smooth, and the surface of a tool
made from fine-grain material can be roughened. Contrast with granularity.
Topography refers to the features or the elevational differences of a stone’s surface.

It is a useful concept for understanding how surfaces wear. Those with high
topographic relief wear on the highest parts first.
Tribochemical wear is an interactive process involving adhesive wear, abrasive

wear, and fatigue wear, all of which create an environment for chemical inter-
actions. These interactions produce reaction products, which are films and oxides
that build up and become visible on stone surfaces as sheen, sometimes referred
to as polish by technologists studying flaked stone tools.
Types are the categories into which entities are sorted. Compare with analyzing
and classifying.
274 G LOSSARY
Unused items are those that were manufactured with all the necessary attributes
to be specific tools but were never used. They may have damage to their surfaces
if, as part of the manufacture process, pecking or grinding was employed to
create the surface, but there is no damage from use. Contrast with blanks.
The ventral surface or side is the lower surface. For manos and other handstones,
this is the working surface. For netherstones and lapstones, this is the side that
rests on the ground or in the hand. Contrast with dorsal surface or side.
Vesicles are cavities in volcanic rock left by bubbles of air or gas that escaped as
the molten rock hardened.
Worn-out items are no longer usable in the activity for which they were designed.
References Cited
Adams, E. C.
1982 Walpi Archaeological Project: Synthesis and Interpretation. Museum of North-
ern Arizona, Flagstaff.
Adams, E. C., and D. Hull
1980 The Prehistoric and Historic Occupation of the Hopi Mesas. In Hopi
Kachina Spirit of Life, edited by D. K. Washburn, pp. 11–27. California Acad-
emy of Sciences, University of Washington Press, Seattle.
Adams, J. L.
1979 Groundstone from Walpi. Walpi Archaeological Project, Phase II, vol. 4,
Stone Artifacts from Walpi. Museum of Northern Arizona, Flagstaff.
1980 Perishable Artifacts from Walpi. Walpi Archaeological Project, Phase II,
vol. 5, Perishable Artifacts and Floral and Faunal Material from Walpi. Mu-
seum of Northern Arizona, Flagstaff.
1988 Use-Wear Analysis of Handstones Used to Grind Corn and Process Hides.
Journal of Field Archaeology 15(3):307–315.
1989a Experimental Replication of the Use of Ground Stone Tools. Kiva 54(3):
261–271.
1989b Methods for Improving Ground Stone Analysis: Experiments in Mano
Wear Patterns. In Experiments in Lithic Technology, edited by D. S. Amick
and R. P. Mauldin, pp. 259–276. BAR International Series No. 528. British
Archaeological Reports, Oxford.
1993a Technological Development of Manos and Metates on the Hopi Mesas.
Kiva 58(3):331–344.
1993b Mechanisms of Wear of Ground Stone Surfaces. Pacific Coast Archaeolog-
ical Society Quarterly 29(4):61–74.
1994a The Development of Prehistoric Grinding Technology at Point of Pines, East-
Central Arizona. Ph.D. dissertation. University Microfilms, Ann Arbor.
1994b Ground Stone Artifacts and Tabular Stone Tools from the Schuk Toak
Mitigation Project. In Archaeological Studies of the Avra Valley, Arizona: Ex-
cavations in the Schuk Toak District, vol. 2, Scientific Studies and Interpreta-
tions, edited by A. Dart, pp. 115–182. Anthropological Papers No. 16.
Center for Desert Archaeology, Tucson.
1995 The Ground Stone Assemblage: The Development of a Prehistoric Grind-
ing Technology in the Eastern Tonto Basin. In The Roosevelt Community
Development Study, vol. 1, Stone and Shell Artifacts, edited by M. D. Elson
and J. J. Clark, pp. 43–114. Anthropological Papers No. 14. Center for
Desert Archaeology, Tucson.
1996 Manual for a Technological Approach to Ground Stone Analysis. Center for
Desert Archaeology, Tucson.
275
276 R EFERENCES
1998 Ground Stone Artifacts. In Archaeological Investigations of Early Village Sites

in the Middle Santa Cruz Valley: Analyses and Synthesis, edited by J. B.
Mabry, pp. 357–422. Anthropological Papers No. 19. Center for Desert
Archaeology, Tucson.
1999 Refocusing the Role of Food-grinding Tools as Correlates for Subsistence
Strategies in the U.S. Southwest. American Antiquity 64(3):475–498.
2000a Tortolita Phase Grinding Technology. In Roots of Sedentism: Archaeological
Excavations at Valencia Vieja, a Founding Village in the Tucson Basin of Southern
Arizona (draft), edited by H. D. Wallace. Anthropological Papers No. 29.
Center for Desert Archaeology, Tucson.
2000b The Ground Stone Assemblage. In Excavations in the Santa Cruz River
Floodplain: The Early Agricultural Component at Los Pozos, edited by D. A.
Gregory, pp. 107–134. Anthropological Papers No. 21. Center for Desert
Archaeology, Tucson.
2000c Ground Stone Analysis. In Archaeological Investigations along Tonto Creek:
Artifact, Environmental, and Mortuary Analyses, vol. 2, Lithic, Ground Stone,
and Environmental Analyses, edited by J. J. Clark. Anthropological Papers
No. 23. Center for Desert Archaeology, Tucson.
2000d Middle Rincon Grinding Technology as Manifest at Sunset Mesa. In Ex-
cavations at Sunset Mesa, edited by M. W. Lindeman, pp. 135–162. Techni-
cal Report No. 2000–02. Desert Archaeology, Inc., Tucson.
2001a The Development of Grinding Technology in the Flagstaff Area. In The
U.S. 89 Archaeological Project: Stone and Shell Artifacts (draft), edited by M.
Elson. Anthropological Papers No. 31. Center for Desert Archaeology,
Tucson.
2001b The Development of Early Agricultural Grinding Technology. In Early
Agricultural Period Material Culture (draft), edited by R. J. Sliva. Anthropo-
logical Papers. Center for Desert Archaeology, Tucson.
Adams, W. Y., and E. W. Adams
1991 Archaeological Typology and Practical Reality: A Dialectical Approach to Arti-
fact Classification and Sorting. Cambridge University Press, Cambridge.
Anyon, R., and S. A. LeBlanc
1984 The Galaz Ruin: A Prehistoric Mimbres Village in Southwestern New Mexico.
University of New Mexico Press, Albuquerque.
Bahr, D. M.
1983 Pima and Papago Social Organization. In Southwest, edited by A. Ortiz,
pp. 178–192. Handbook of North American Indians, vol. 10. Smithsonian In-
stitution, Washington, D.C.
Barnett, F.
1969 Tonque Pueblo: A Report on a Partial Excavation of an Ancient Pueblo IV Indian
Ruin in New Mexico. Albuquerque Archaeological Society, Albuquerque.
Bartlett, K.
1933 Pueblo Milling Stones of the Flagstaff Region and Their Relation to Others in
the Southwest. Museum of Northern Arizona Bulletin No. 3. Museum of
Northern Arizona, Flagstaff.
Batkin, J.
1987 Pottery of the Pueblos of New Mexico, 1700–1940. The Taylor Museum of the
Colorado Fine Arts Center, Colorado Springs.
R EFERENCES 277
Belitz, B.
1979 Brain Tanning the Sioux Way. Pine Ridge Reservation Publication, Pine
Ridge, South Dakota.
Bernard-Shaw, M.
1990 Experimental Agave Fiber Extraction. In Rincon Seasonal Occupation in
the Northeastern Tucson Basin, by F. W. Huntington and M. Bernard-Shaw,
pp. 181–195. Technical Report No. 90–2. Institute for American Research,
Tucson.
Bice, R. A., and W. M. Sundt
1972 Prieta Vista, a Small Pueblo III Ruin in North-central New Mexico. Albu-
querque Archaeological Society, Albuquerque.
Blau, P. J.
1989 Friction and Wear: Transitions of Materials. Noyes Publications, Park Ridge,
New Jersey.
Bostwick, T. W., and J. H. Burton
1993 A Study in Sourcing Hohokam Basalt Ground Stone Implements. Kiva
58(3):357–372.
Breternitz, C. D.
1997 An Analysis of Axes and Mauls from Chaco Canyon, New Mexico. In
Ceramics, Lithics, and Ornaments of Chaco Canyon, edited by J. Mathien, 3:
977–1012. Publications in Archaeology 18G, Chaco Canyon Studies. Na-
tional Park Service, Santa Fe.
Brew, J. O.
1974 Archaeology of Alkali Ridge, Southeastern Utah. Reprinted. Kraus Reprint
Co., Millwood, New York. Originally published 1946, Papers of the Pea-
body Museum No. 21, Harvard University, Cambridge.
Bunzel, R.
1972 The Pueblo Potter: A Study of Creative Imagination in Primitive Art. Reprinted.
Dover Publications, New York. Originally published 1929, Columbia Uni-
versity Press, New York.
Burton, J. F.
1990 The Archaeology of Sivu’ovi: The Archaic to Basketmaker Transition at Petrified
Forest National Park. Publications in Anthropology No. 55. Western Arche-
ological and Conservation Center, National Park Service, Tucson.
Castetter, E. F., and W. H. Bell
1937 The Utilization of Mesquite and Screwbean by the Aborigines in the American
Southwest. Ethnobiological Studies in the American Southwest No. 5. Uni-
versity of New Mexico, Albuquerque.
1951 Yuman Indian Agriculture. University of New Mexico Press, Albuquerque.
Cattanach, G. S.
1980 Long House: Mesa Verde National Park, Colorado. Publications in Archaeol-
ogy 7H, Wetherill Mesa Studies. National Park Service, Washington,
D.C.
Colton, H. S.
1946 The Sinagua: A Summary of the Archaeology of the Region of Flagstaff, Arizona.
Museum of Northern Arizona Bulletin 22, Flagstaff.
1948 Indian Life—Past and Present. In The Inverted Mountains: Canyons of the
West, edited by R. Peattie, pp. 111–128. Vanguard Press, New York.
278 R EFERENCES
Connor, S.
1943 Excavations at Kinnikinnick, Arizona. American Antiquity 8(4):376–379.
Cosgrove, H. S., and C. B. Cosgrove
1932 The Swarts Ruin: A Typical Mimbres Site in Southwestern New Mexico. Papers
of the Peabody Museum of American Archaeology and Ethnology 15(1).
Harvard University, Cambridge, Massachusetts.
Cosner, A. J.
1951 Arrowshaft-Straightening with a Grooved Stone. American Antiquity
17(2):147–148.
Cowgill, George L.
1990 Why Pearson’s r Is Not a Good Similarity Coefficient for Comparing
Collections. American Antiquity 55(3):512–521.
Culin, S.
1975 Games of the North American Indians. Reprinted. Dover, New York. Origi-
nally published 1907, Annual Report of the Bureau of American Ethnol-
ogy, Smithsonian Institution, Washington, D.C.
Cushing, F. H.
1883 Zuni Fetishes. Annual Report of the Bureau of American Ethnology. Smith-
sonian Institution, Washington, D.C.
1920 Zuni Breadstuff. In Indian Notes and Monographs, vol. 8. Museum of the
American Indian, Heye Foundation, New York.
Czichos, H.
1978 Tribology: A Systems Approach to the Science and Technology of Friction, Lu-
brication, and Wear. Tribology Series No. 1. Elsevier Scientific Publishing
Co., New York.
Dart, A.
1995 Ground Stone Artifacts. In Archaeological Investigations at Los Morteros, a
Prehistoric Community in the Northern Tucson Basin, Part II, by H. D. Wal-
lace, pp. 503–566. Anthropological Papers No. 17. Center for Desert Ar-
chaeology, Tucson.
de Beaune, S. A., and R. White
1993 Ice Age Lamps. Scientific American (March):108–113.
Diamond, W. J.
1989 Practical Experiment Designs for Engineers and Scientists. 2nd ed. Van Nos-
trand Reinhold, New York.
Dick, H. W.
1965a Bat Cave, Catron County, New Mexico. Monograph No. 27. School of Amer-
ican Research, Santa Fe.
1965b Report on Salvage Operations in Picuris Pueblo, New Mexico. National
Park Service. Manuscript on file, Arizona State Museum Library, Tucson.
Diehl, M. W.
1996 The Intensity of Maize Processing and Production in Upland Mogollon
Pithouse Villages a.d. 200–1000. American Antiquity 61(1): 102–115.
Dillingham, R., and M. Elliot
1992 Acoma and Laguna Pottery. School of American Research, Santa Fe.
Di Peso, C. C.
1951 The Babocomari Village Site on the Babocomari River, Southeastern Arizona.
Amerind Foundation Series No. 5. Amerind Foundation, Dragoon,
Arizona.
R EFERENCES 279
Di Peso, C. C., J. B. Rinaldo, and G. J. Fenner

1974 Casas Grandes: A Fallen Trading Center of the Gran Chichimeca, vol. 7, Stone
and Metal. Amerind Foundation Series No. 9. Northland Press, Flagstaff.
Dobres, M., and C. R. Hoffman
1999a Introduction: A Context for the Present and Future of Technology Studies.
In The Social Dynamics of Technology: Practice, Politics, and World View,
edited by M. Dobres and C. R. Hoffman, pp. 1–19. Smithsonian Institu-
tion Press, Washington D.C.
Dobres, M., and C. R. Hoffman (editors)
1999b The Social Dynamics of Technology: Practice, Politics, and World View. Smith-
sonian Institution Press, Washington D.C.
Doelle, W. H.
1976 Desert Resources and Hohokam Subsistence: The Conoco Florence Project. Ar-
chaeological Series No. 13. Arizona State Museum, University of Arizona,
Tucson.
Dorsey, G. A., and H. R. Voth
1902 The Mishongnovi Ceremony of the Snake and Antelope Fraternities. Field Mu-
seum of Natural History, Anthropological Series Vol. 3, No. 3, pp. 159–261.
Dowson, D.
1979 History of Tribology. Longman, New York.
Doyel, D. E.
1985 Exchange and Interaction. In Hohokam Settlement and Economic Systems in
the Central New River Drainage, Arizona, edited by D. E. Doyel and M. D.
Elson, pp. 715–726. Publications in Archaeology No. 4. Soil Systems, Inc.,
Phoenix.
Duncan, R. B., and W. H. Doleman
1991 Fire-Cracked Rock Studies. In Landscape Archaeology in the Southern Tula-
rosa Basin, vol. 2, Testing, Excavation, and Analysis, edited by W. H. Dole-
man, R. C. Chapman, J. A. Schutt, M. K. Swift, and K. D. Morrison, pp.
317–344. Report No. 185–324E. Office of Contract Archaeology, Univer-
sity of New Mexico, Albuquerque.
Dunnell, R. C.
1971 Sabloff and Smith’s “The Importance of Both Analytic and Taxonomic
Classification in the Type-Variety System.” American Antiquity 36(1):115–
118.
1986 Methodological Issues in Americanist Artifact Classification. Advances
in Archaeological Method and Theory 9:149–207.
Etzkorn, M. C.
1993 Stone and Mineral Artifacts. In The Duckfoot Site, edited by R. R. Lightfoot
and M. C. Etzkorn, pp. 157–179. Occasional Papers No. 3. Crow Canyon
Archaeological Center, Cortez, Colorado.
Euler, R. C., and H. F. Dobyns
1983 The Ethnoarchaeology of Pai Milling Stones. In Collected Papers in Honor
of Charlie Steen, Jr., edited by N. L. Fox, pp. 253–267. Papers of the Archae-
ological Society of New Mexico No. 8. Albuquerque Archaeological Soci-
ety Press, Albuquerque.
Euler, R. T., and D. A. Gregory
1988 Pecked, Ground, and Polished Stone Artifacts. In The 1982–1984 Excava-
tions at Las Colinas, vol. 4, Material Culture, by D. R. Abbott, K. E. Beckwith,
280 R EFERENCES
P. L. Crown, R. T. Euler, D. A. Gregory, J. R. London, M. B. Saul, L. A.

Schwalbe, M. Bernard-Shaw, C. R. Szuter, and A. W. Vokes, pp. 299–317.
Archaeological Series No. 162. Arizona State Museum, University of Ari-
zona, Tucson.
Ewers, J. C.
1945 Blackfeet Crafts. U.S. Indian Service, Lawrence Kansas.
1958 The Blackfeet: Raiders on the Northwestern Plains. University of Oklahoma
Press, Norman.
Ferg, A.
1991 Hohokam T-shaped Stones. Archaeology in Tucson 5(3):1–3.
1998 Rare Stone, Fired Clay, Bone, and Shell Artifacts. In Archaeological Inves-
tigations of Early Village Sites in the Middle Santa Cruz Valley: Analyses and
Synthesis, Part II, edited by J. B. Mabry, pp. 545–654. Anthropological
Papers No. 19. Center for Desert Archaeology, Tucson.
Fewkes, J. W.
1894 The Snake Ceremonies at Walpi. Journal of American Ethnology and Archae-
ology, vol. 4.
1898 Archeological Expedition to Arizona in 1895. In Annual Report of the Bu-
reau of American Ethnology to the Secretary of the Smithsonian Institution,
1895–1896, vol. 17, pp. 519–742. U.S. Government Printing Office, Wash-
ington, D.C.
1909 Antiquities of the Mesa Verde National Park, Spruce Tree House. Bureau of
American Ethnology Bulletin 41. U.S. Government Printing Office, Wash-
ington, D.C.
1911 Antiquities of the Mesa Verde National Park, Cliff Palace. Bureau of American
Ethnology Bulletin 51. U.S. Government Printing Office, Washington
D.C.
1912 Casa Grande, Arizona. In Annual Report of the Bureau of American Ethnol-
ogy to the Secretary of the Smithsonian Institution, 1906–1907, vol. 28, pp.
25–179. U.S. Government Printing Office, Washington, D.C.
1920 Fire Worship of the Hopi Indians. In Annual Report of the Board of Regents
of the Smithsonian Insititution, pp. 589–610. Publication 2622. U.S. Govern-
ment Printing Office, Washington, D.C.
Fiero, D. C., R. W. Munson, M. T. McClain, S. M. Wilson, and A. H. Zier
1980 The Navajo Project: Archaeological Investigations, Page to Phoenix 500 KV
Southern Transmissions Line. Museum of Northern Arizona Research
Paper No. 11. Museum of Northern Arizona, Flagstaff.
Flenniken, J. J., and T. L. Ozbun
1988 Experimental Analysis of Plains Grooved Abraders. Plains Anthropologist
33(119):37–52.
Fontana, B. L.
1983 Pima and Papago: Introduction. In Southwest, edited by A. Ortiz, pp. 125–
136. Handbook of North American Indians, vol. 10. Smithsonian Institution,
Washington, D.C.
Fowler, D. D.
2000 A Laboratory of Anthropology, Science, and Romanticism in the Amer-
ican Southwest, 1846–1930. University of New Mexico Press, Albu-
querque.
R EFERENCES 281
Franklin, H. H.
1980 Excavations at Second Canyon Ruin, San Pedro Valley, Arizona. Arizona State
Museum Contributions to Highway Salvage Archaeology in Arizona No.
60. Arizona State Museum, Tucson.
Fratt, L.
1996a Ground Stone. In Archaeological Investigations at the SRI Locus, West
Branch Site (AZ AA:16:3 [ASM]), a Rincon Phase Village in the Tucson Basin
(draft), by J. H. Altschul, T. Bubemyre, K. M. Cairns, W. L. Deaver,
A. Della Croce, S. K. Fish, L. Fratt, K. G. Harry, J. P. Holmlund, G. Huck-
leberry, C. H. Miksicek, A. W. Vokes, S. M. Whittlesey, M. N. Zedeño,
and L. F. Ziady. Statistical Research, Inc., Tucson. Submitted to Pima
County Department of Transportation and Flood Control District,
Tucson.
1996b Ground Stone Tools and Other Stone Artifacts from the Gibbon Springs
Site. In Excavation of the Gibbon Springs Site: A Classic Period Village in the
Northeastern Tucson Basin, edited by M. C. Slaughter and H. Roberts, pp.
267–329. Archaeological Report No. 94–87. SWCA, Inc. Environmental
Consultants, Tucson. Submitted to Canoa Homes, Tucson.
Fratt, L., and M. Biancaniello
1993 Homol’ovi III Ground Stone in the Raw: A Study of the Local Sandstone
Used to Make Ground Stone Artifacts. Kiva 58(3):373–391.
Geib, P. R., and M. M. Callahan
1988 Clay Residue on Polishing Stone. The Kiva 53(4):357–362.
Gifford, J. C.
1980 Archaeological Explorations in Caves of the Point of Pines Region, Arizona.
Anthropological Papers of the University of Arizona 36. University of
Arizona Press, Tucson.
Gladwin, H. S., E. W. Haury, E. B. Sayles, and N. Gladwin
1975 Excavations at Snaketown: Material Culture. Reprinted. Originally published
1938, University of Arizona Press, Tucson.
Gould, R. A.
1977 Ethnoarchaeology; or, Where Do Models Come From? In Stone Tools as
Cultural Markers, pp. 162–168, edited by R. V. S. Wright. Prehistory and
Material Culture Series No. 12. Humanities Press, New Jersey.
Gould, R. A., D. A. Koster, and A. H. L. Sontz
1971 The Lithic Assemblage of the Western Desert Aboriginies of Australia.
American Antiquity 36(2):149–169.
Gould., R. A., and P. J. Watson
1982 A Dialogue on the Meaning and Use of Analogy in Ethnoarchaeological
Reasoning. Journal of Anthrolopogical Archaeology 1:355–381.
Greenwald, D. M.
1988 Ground Stone. In Hohokam Settlement along the Slopes of the Picacho Moun-
tains, vol. 4, Material Culture, edited by M. M. Callahan, pp. 127–220. Mu-
seum of Northern Arizona Research Paper No. 35. Museum of Northern
Arizona, Flagstaff.
1993 Ground Stone Artifacts from La Ciudad de los Hornos. In In the Shadow
of South Mountain: The Pre-Classic Hohokam of La Ciudad de Los Hornos, 1991–
1992 Excavations, Part I, edited by M. L. Chenault, R. V. N. Ahlstrom,
282 R EFERENCES
and T. N. Motsinger, pp. 317–358. Archaeological Report No. 93–30. SWCA

Environmental Consultants, Inc., Tucson.
Gregory, D. A., and J. B. Mabry
1998 Revised Research Design for the Archaeological Treatment Plan, Interstate 10
Corridor Improvement Project, Tangerine Road to the Interstate 19 Interchange.
Technical Report No. 97–19. Desert Archaeology, Inc., Tucson.
Guernsey, S. J., and A. V. Kidder
1921 Basketmaker Caves of Northeastern Arizona. Papers of the Peabody Museum
of American Archaeology and Ethnology No. 8(2). Harvard University,
Cambridge.
Hard, R., R. P. Mauldin, and G. R. Raymond
1996 Mano Size, Stable Isotope Ratios, and Macrobotanical Remains as Multi-
ple Lines of Evidence of Maize Dependence in the American Southwest.
Journal of Archaeological Method and Theory 3(4):253–318.
Haury, E. W.
1931a Showlow and Pinedale Ruins. In Recently Dated Pueblo Ruins in Arizona,
by E. W. Haury and L. L. Hargrave, pp. 4–79. Smithsonian Miscellaneous
Collections 82(11). Smithsonian Institution, Washington, D.C.
1931b Minute Beads from Prehistoric Pueblos. American Anthroplogist 33:80–87.
1934 The Canyon Creek Ruin and Cliff Dwellings of the Sierra Ancha. Medallion
Papers No. 14. Gila Pueblo, Globe, Arizona.
1945 The Excavation of Los Muertos and Neighboring Ruins in the Salt River Valley,
Southern Arizona. Papers of the Peabody Museum of American Archae-
ology and Ethnology No. 24(1). Harvard University, Cambridge, Massa-
chusetts.
1950 The Stratigraphy and Archaeology of Ventana Cave. University of Arizona
Press, Tucson.
1958 Evidence at Point of Pines for a Prehistoric Migration from Northern Ari-
zona. In Migrations in New World Culture History, edited by R. H. Thomp-
son, pp. 1–6. University of Arizona Bulletin No. 29(2), Social Science Bul-
letin 27. University of Arizona, Tucson.
1975 Stone Palettes and Ornaments. In Excavations at Snaketown: Material Cul-
ture, edited by H. S. Gladwin, E. W. Haury, E. B. Sayles, and N. Gladwin,
pp. 121–134. Reprinted. Originally published 1938, University of Arizona
Press, Tucson.
1976 The Hohokam: Desert Farmers and Craftsmen, Excavations at Snaketown, 1964–
1965. University of Arizona Press, Tucson.
1985a Tla Kii Ruin: Forestdale’s Oldest Pueblo. In Mogollon Culture in the Forest-
dale Valley, East-central Arizona, by E. W. Haury, pp. 7–133. University of
Arizona Press, Tucson.
1985b Excavations in the Forestdale Valley, East-Central Arizona. In Mogollon
Culture in the Forestdale Valley, East-Central Arizona, by E. W. Haury, pp.
137–454. Reprinted. University of Arizona Press, Tucson. Originally pub-
lished 1947, University of Arizona Bulletin, Vol. XI, No. 4, Social Sciences
Bulletin 12.
Haury, E. W., and E. B. Sayles
1985 An Early Pit House Village of the Mogollon Culture, Forestdale Valley,
Arizona. In Mogollon Culture in the Forestdale Valley, East-Central Arizona,
R EFERENCES 283
by E. W. Haury, pp. 281–454. Reprinted. University of Arizona Press,

Tucson. Originally published 1947, University of Arizona Bulletin, Vol.
XVIII, No. 4, Social Sciences Bulletin 16.
Hawley, F. G.
1975 Chemical Investigation of the Incrustation on Pottery Vessels and Palettes
from Snaketown: Appendix IV. In Excavations at Snaketown: Material Cul-
ture, by H. S. Gladwin, E. W. Haury, E. B. Sayles, and N. Gladwin. Re-
printed. University of Arizona Press, Tucson. Originally published 1938,
Medallion Paper No. 25, Gila Pueblo, Globe, Arizona.
Hayden, B.
1987a Traditional Metate Manufacturing in Guatemala Using Chipped Stone
Tools. In Lithic Studies among the Contemporary Highland Maya, edited by
B. Hayden, pp. 8–111. University of Arizona Press, Tucson.
Hayden, B. (editor)
1979 Lithic Use-Wear Analysis. Proceedings of the Conference on Lithic Use-
Wear Analysis, Simon Fraser University, Burnaby (Vancouver), British
Columbia, March 1977. Academic Press, New York.
1987b Lithic Studies among the Contemporary Highland Maya. University of Ari-
zona Press, Tucson.
Hayden, B., and J. Kamminga
1979 An Introduction to Use-Wear: The First CLUW. In Lithic Use-Wear Analy-
sis, Proceedings of the Conference on Lithic Use-Wear Analysis, Simon
Fraser University, Burnaby (Vancouver), British Columbia, March 1977,
edited by B. Hayden, pp. 1–13. Academic Press, New York.
Hayden, B., and M. Nelson
1981 The Use of Chipped Lithic Material in the Contemporary Maya High-
lands. American Antiquity 46(4):885–896.
Hayden, J.
1959 Notes on Pima Pottery Making. The Kiva 24(3):10–16.
Hayes, A. C., and J. A. Lancaster
1975 Badger House Community, Mesa Verde National Park, Colorado. Publications
in Archaeology 7E. Wetherill Mesa Studies. National Park Service, Wash-
ington, D.C.
Hayes, A. C., J. N. Young, and A. H. Warren
1981 Excavation of Mound 7, Gran Quivera National Monument, New Mexico.
Publications in Archaeology 16. National Park Service, Washington,
D.C.
Hemmings, E. T.
1967 Cruciform and Related Artifacts of Mexico and the Southwestern United
States. The Kiva 31(4):150–169.
Hibben, F. C.
1938 The Gallina Phase. American Antiquity 4(2):131–136.
Hill, J. N., and R. K. Evans
1972 A Model for Classification and Typology. In Models in Archaeology, edited
by D. L. Clarke, pp. 231–273. Metheun, London.
Hill, W. W.
1982 An Ethnography of Santa Clara Pueblo, New Mexico. Edited and anno-
tated by C. H. Lange. University of New Mexico Press, Albuquerque.
284 R EFERENCES
Hoebel, E. A.
1979 Zia Pueblo. In Southwest, edited by A. Ortiz, pp. 407–417. Handbook of
North American Indians, vol. 9. Smithsonian Institution, Washington, D.C.
Hoffman, C. R.
1999 Intentional Damage as Technological Agency: Breaking Metals in Late
Prehistoric Spain. In The Social Dynamics of Technology: Practice, Politics,
and World View, edited by M. A. Dobres and C. R. Hoffman, pp. 103–123.
Smithsonian Institution Press, Washington, D.C.
Horsfall, G. A.
1987 Design Theory and Grinding Stones. In Lithic Studies among the Contempo-
rary Highland Maya, edited by B. Hayden, pp. 332–377. University of Ari-
zona Press, Tucson.
Hough, W.
1890 Fire-Making Apparatus in the U.S. National Museum. In U.S. National
Museum Annual Report 1888, pp. 531–587. U.S. National Museum, Wash-
ington, D.C.
1907 Antiquities of the Upper Gila and Salt River Valleys in Arizona and New Mex-
ico. Bureau of American Ethnology Bulletin No. 35. U.S. Government
Printing Office, Washington, D.C.
1914 Culture of the Ancient Pueblos of the Upper Gila River Region, New Mexico
and Arizona. National Museum Bulletin 87. Smithsonian Institution, Wash-
ington, D.C.
1915 The Hopi Indians. The Torch Press, Cedar Rapids, Iowa.
1918 The Hopi Indian Collection in the United States National Museum. Pro-
ceedings of the National Museum 54(235):235–296.
Huckell, B. B.
1995 Of Marshes and Maize: Preceramic Agricultural Settlements in the Cienega
Valley, Southeastern Arizona. Anthropological Papers No. 59. University
of Arizona Press, Tucson.
1998 The Ground Stone Collection. In Early Farmers of the Sonoran Desert: Ar-
chaeological Investigations at the Houghton Road Site, Tucson, Arizona, edited
by R. Ciolek-Torrello, pp. 119–126. Technical Series 72. Statistical Research,
Inc., Tucson.
Hudson, T., and T. C. Blackburn
1982 The Material Culture of the Chumash Interaction Sphere. Vol. 1, Food Procure-
ment and Transportation. Ballena Press Anthropological Papers No. 25.
Santa Barbara Museum of Natural History, Los Altos, California.
Jackson, T. L.
1991 Pounding Acorn: Women’s Production as Social and Economic Focus. In
Engendering Archaeology: Women in Prehistory, edited by J. M. Gero and
M. W. Conkey, pp. 301–325. Basil Blackwell, Cambridge.
Jeançon, J. A.
1923 Excavations in the Chama Valley, New Mexico. Bureau of American Eth-
nology Bulletin No. 81. U.S. Government Printing Office, Washington,
D.C.
Jernigan, W. E.
1978 Jewelry of the Prehistoric Southwest. School of American Research, Univer-
sity of New Mexico Press, Albuquerque.
R EFERENCES 285
Johnson, A. E.
1965 The Development of Western Pueblo Culture. Unpublished Ph.D. disser-
tation, Department of Anthropology, University of Arizona, Tucson.
Johnson, A. S.
1971 Finger-Loops and Cruciform Objects. American Antiquity 36(2):188–194.
Judd, N. M.
1954 The Material Culture of Pueblo Bonito. Smithsonian Miscellaneous Collec-
tions No. 124. U.S. Government Printing Office, Washington, D.C.
1959 Pueblo del Arroyo, Chaco Canyon, New Mexico. Smithsonian Miscellaneous
Collections, Vol. 138, No. 1. U.S. Government Printing Office, Washing-
ton, D.C.
Kamp, K. A.
1995 A Use-Wear Analysis of the Function of Basalt Cylinders. Kiva 61(2):109–
119.
Kamp, K. A., and J. C. Whittaker
1990 Lizard Man Village: A Small Site Perspective on Northern Sinagua Social
Organization. Kiva 55(2):99–125.
1999 Surviving Adversity: The Sinaqua of Lizard Man Village. University of Utah
Press, Salt Lake City.
Kandel, A.
1982 Fuzzy Techniques in Pattern Recognition. John Wiley and Sons, New York.
Keeley, L.
1980 Experimental Determination of Stone Tools Uses: A Microwear Analysis.
University of Chicago Press.
Kelly, I. T.
1978 The Hodges Ruin, a Hohokam Community in the Tucson Basin. J. E. Officer
and E. W. Haury, collaborators. Anthropological Papers No. 30. University
of Arizona Press, Tucson.
Kewanwytewa, J., and K. Bartlett
1946 Hopi Moccasin Making. Plateau 19(1)21–28.
Kidder, A. V.
1932 The Artifacts of Pecos. Yale University Press, New Haven.
Kidder, A. V., and S. J. Guernsey
1919 Archaeological Explorations in Northeastern Arizona. Bureau of American
Ethnology Bulletin No. 65. U.S. Government Printing Office, Washington,
D.C.
Kintigh, Keith W.
1984 Measuring Archaeological Diversity by Comparison with Simulated As-
semblages. American Antiquity 49:44–54.
Kluckhohn, C. C., W. W. Hill, and L. W. Kluckhohn
1971 Navajo Material Culture. Belknap Press, Harvard University Press, Cam-
bridge.
Kragelsky, I. V., M. N. Dobychin, and V. S. Kombalov
1982 Friction and Wear: Calculation Methods. Pergamon Press, New York.
Kraybill, N.
1977 Pre-Agricultural Tools for the Preparation of Foods in the Old World. In
Orgins of Agriculture, edited by C. A. Reed, pp. 485–521. World Anthro-
pology. Mouton Publishers, The Hague.
286 R EFERENCES
Ladd, E. J.
1979 Zuni Economy. In Southwest, edited by A. Ortiz, pp. 492–498. Handbook
of North American Indians, vol. 9. Smithsonian Institution, Washington,
D.C.
Lakoff, G.
1987 Women, Fire, and Other Dangerous Things. University of Chicago Press,
Chicago.
Lambert, M. F.
1954 Paa-ko, Archaeological Chronicle of an Indian Village in North Central New
Mexico, Parts I–IV. School of American Research, Santa Fe.
Lancaster, J.
1984 Ground Stone Artifacts. In The Galaz Ruin: A Prehistoric Mimbres Village in
Southwestern New Mexico, edited by R. Anyon and S. A. LeBlanc, pp. 247–
262. University of New Mexico Press, Albuquerque.
Larralde, S. L., and S. Schlanger
1994 Anasazi Axes from the La Plata Valley and the Northern Southwest: Man-
ufacture, Use, and Discard. Paper presented at the 59th Annual Meeting
of the Society for American Archaeology, Anaheim.
Laubin, R., and G. Laubin
1957 The Indian Tipi. University of Oklahoma Press, Norman.
LeBlanc, S.
1983 The Mimbres People: Ancient Painters of the American Southwest. Thames
and Hudson.
Lechtman, H.
1999 Afterword. In The Social Dynamics of Technology: Practice, Politics, and World
View, edited by M. A. Dobres and C. R. Hoffman, pp. 223–232. Smithson-
ian Institution Press, Washington, D.C.
Lemonnier, P.
1992 Elements for an Anthropology of Technology. Museum of Anthropology, Uni-
versity of Michigan, Ann Arbor.
Leubben, R. A.
1953 Leaf Water Site. In Salvage Archaeology in the Chama Valley, New Mexico,
assembled by F. Wendorf, pp. 9–33. Monographs of the School of Amer-
ican Research No. 17. Santa Fe.
Liljeblad, S., and C. S. Fowler
1986 Owens Valley Paiute. In Great Basin, edited by W. L. D’Azevedo, pp. 412–
Washington, D.C.
Lindsay, A. J. Jr.
1987 Anasazi Population Movements to Southern Arizona. American Archaeol-
ogy 6(3):190–198.
Logan, E. N., and L. Fratt
1993 Pigment Processing at Homol’ovi III: A Preliminary Study. Kiva 58(3):
415–428.
Lowell, E. S.
1990 Paletas and Tabletas: Comparisons of Hohokam Stone Palettes with Snuff
Trays of South America. Kiva 55:321–355.
R EFERENCES 287
Lowell, J. C.
1991 Prehistoric Households at Turkey Creek Pueblo, Arizona. Anthropological
Papers No. 54. University of Arizona Press, Tucson.
McCartney, P.H., and M. F. Glass
1990 Simulation Models and the Interpretation of Archaeological Diversity.
McCartney, P.H., and A. W. Simon
1997 Ground-Stone Artifacts from the Uplands Complex Sites. In Classic Period
Settlement in the Uplands of Tonto Basin. Roosevelt Platform Mound Study.
Reports on the Uplands Complex, edited by T.J. Oliver, pp. 403–409.
Roosevelt Monograph Series, Anthropological Field Studies 34, Arizona
State University, Tempe.
McGee, W. J.
1898 The Seri Indians. In Annual Report of the Bureau of American Ethnology
1895–1896, vol. 17, pp. 1–344. U.S. Government Printing Office, Washing-
ton, D.C.
McGregor, J. C.
1941 Winona and Ridge Ruin, Part I: Architecture and Material Culture. Museum
of Northern Arizona Bulletin 18, Flagstaff.
Martin, P. S.
1943 The SU Site: Excavations at a Mogollon Village, Western New Mexico, Second
Season, 1941. Anthropological Series Vol. 32, No. 2. Publication 526. Field
Museum of Natural History, Chicago.
1979 Prehistory: Mogollon. In Southwest, edited by A. Ortiz, pp. 61–74. Hand-
book of North American Indians, vol. 9. Smithsonian Institution, Washing-
ton, D.C.
Martin P. S., and J. B. Rinaldo
1947 The SU Site: Excavavations at a Mogollon Village, Western New Mexico, Third
Season, 1946. Anthropology Series Vol. 32, No. 3. Field Museum of Nat-
ural History, Chicago.
1950a Turkey Foot Ridge Site: A Mogollon Village, Pine Lawn Valley, Western New
Mexico. Fieldiana: Anthropology 38:2. Field Museum of Natural History,
Chicago.
1950b Sites of the Reserve Phase, Pine Lawn Valley, Western New Mexico. Fieldiana:
Anthropology 38:3. Field Museum of Natural History, Chicago.
1960 Table Rock Pueblo, Arizona. Fieldiana: Anthropology 51:2. Field Museum
of Natural History, Chicago.
Martin, P. S., J. B. Rinaldo, and E. Antevs
1949 Cochise and Mogollon Sites, Pine Lawn Valley, Western New Mexico. Fieldiana:
Anthropology 38:1. Field Museum of Natural History, Chicago.
Martin, P. S., J. B. Rinaldo, and E. R. Barter
1957 Late Mogollon Communities: Four Sites of the Tularosa Phase, Western New
Mexico. Fieldiana: Anthropology 49:1. Field Museum of Natural History,
Chicago.
Martin, P. S., J. B. Rinaldo, E. A. Bluhm, and H. C. Cutler
1956 Higgins Flat Pueblo, Western New Mexico. Fieldiana: Anthropology 45. Field
288 R EFERENCES
Martin, P. S., J. B. Rinaldo, E. A. Bluhm, H. C. Cutler, and R. Grange Jr.

1952 Mogollon Cultural Continuity and Change: The Stratigraphic Analysis of Tula-
rosa and Cordova Caves. Fieldiana: Anthropology 40. Field Museum of
Natural History, Chicago.
Martin, P. S., J. B. Rinaldo, and W.A. Longacre
1961 Mineral Creek Site and Hooper Ranch Pueblo, Eastern Arizona. Fieldiana:
Anthropology 52. Field Museum of Natural History, Chicago.
Martin, P. S., J. B. Rinaldo, W. A. Longacre, L. G. Freeman Jr., J. A. Brown, R. H.
Hevly, and M. E. Cooley
1964 Chapters in the Prehistory of Eastern Arizona, II. Fieldiana: Anthropology
55. Field Museum of Natural History, Chicago.
Marwitt, J. D.
1986 Fremont Culture. In Great Basin, edited by W. L. D’Azevedo, pp. 161–
Washington, D.C.
Mauldin, R. P.
1993 The Relationship between Ground Stone and Agricultural Intensification
in Western New Mexico. Kiva 58:317–330.
Mills, B. J.
1995 Gender and the Reorganization of Historic Zuni Craft Production: Impli-
cations for Archaeological Interpretations. Journal of Anthropological Re-
search 51(2):149–172.
Mills, B. J., S. Van Keuren, S. L. Stinson, W. M. Graves, III, E. J. Kaldahl, and J. M.
Newcomb
1999 Excavations at Bailey Ruin. In Living on the Edge of the Rim: Excavations
and Analysis of the Silver Creek Archaeological Research Project, 1993–1998,
edited by B. J. Mills, S. A. Herr, and S. Van Keuren, pp. 149–242. Archaeo-
logical Series 192. Arizona State Museum, University of Arizona, Tucson.
Mills, P. R.
1993 An Axe to Grind: A Functional Analysis of Anasazi Stone Axes from Sand
Canyon Pueblo Ruin (5MT765), Southwestern Colorado. Kiva 58(3)393–413.
Mindeleff, V.
1989 A Study of Pueblo Architecture in Tusayan and Cibola. Reprinted. Smithson-
ian Institution, Washington, D.C. Originally published 1891, in Annual
Report of the Bureau of American Ethnology, 1886–1887, vol. 8, U.S. Govern-
ment Printing Office, Washington, D.C.
Montero, L. G.
1992 Ground and Worked Stone Artifacts from the Verde Bridge Sites. In Pre-
historic and Historic Occupation of the Lower Verde River Valley: The State
Route 87 Verde Bridge Project by Mark R. Hackbarth, pp. 251–282. North-
land Research, Inc., Flagstaff, Arizona.
Morris, D. H.
1990 Changes in Groundstone Following the Introduction of Maize into the
American Southwest. Journal of Anthropological Research 46:177–194.
Morris, E. A.
1980 Basketmaker Caves in the Prayer Rock District, Northeastern Arizona. Anthro-
pological Papers of the University of Arizona No. 35. University of Ari-
zona Press, Tucson.
R EFERENCES 289
Morris, E. H.
1919 The Aztec Ruin. Anthropological Papers of the American Museum of Natu-
ral History, vol. 26, pt. 1. American Museum of Natural History, New York.
1928 An Aboriginal Salt Mine at Camp Verde, Arizona. Anthropological Papers
of the American Museum of Natural History, vol. 30, pt. 3. American
Museum of Natural History, New York.
1939 Archaeological Studies in the La Plata District, Southwestern Colorado and
Northwestern New Mexico. Carnegie Institution, Washington, D.C.
Nelson, M. C., and H. Lippmeier
1993 Grinding-Tool Design as Conditioned by Land-Use Pattern. American An-
tiquity 58:286–305.
O’Bryan, D.
1950 Excavations in Mesa Verde National Park, 1947–1948. Gila Pueblo, Medal-
lion Papers No. 39. Globe, Arizona.
Odegaard, N., and M. Crawford
2000 Three Painted, Clay-lined Baskets from Feature 411 at Las Tortugas, Locus
1 (U:3:297/332). In Tonto Creek Archaeological Project: Life and Death along
Tonto Creek (draft), by J. J. Clark and P. D. Minturn, pp. 299–302. Anthro-
pological Papers No. 24. Center for Desert Archaeology, Tucson.
Odell, G., and F. Odell-Vereecken
1980 Verifying the Reliability of Lithic Use-Wear Assesssments by ‘Blind Tests’:
The Low Power Approach. Journal of Field Archaeology 7:87–120.
O’Kane, W. C.
1950 Sun in the Sky. University of Oklahoma Press, Norman.
Opler, M. E.
1983 Chiricahua Apache. In Southwest, edited by A. Ortiz, pp. 401–418. Handbook
of North American Indians, vol. 9. Smithsonian Institution, Washington, D.C.
Parsons, E. C.
1939 Pueblo Indian Religion. University of Chicago Press.
Plog, F. T.
1974 The Study of Prehistoric Change. Academic Press, New York.
Plog, S., and M. Hegmon
1993 The Sample Size–Richness Relation: The Relevance of Research Questions,
Sampling Strategies, and Behavioral Variation. American Antiquity 58(3):
489–496.
Pond, A. W.
1930 Primitive Methods of Working Stone Based on Experiments of Halvor
Skavlem. Logan Museum Bulletin 2(1) Beloit College, Beloit, Wisconsin.
Quinn, T. F. J.
1971 The Application of Modern Physical Techniques to Tribology. Newnes-Butter-
worths, London.
Rapp, J.
1995 Worked Stone. In The Prehistoric Archaeology of Heritage Square, edited by
T. K. Henderson, pp. 139–150. Pueblo Grande Museum Anthropological
Papers No. 3. City of Phoenix Parks, Recreation and Library Department,
Phoenix.
1998 Lithic Analysis. In Life Along Big Bug Creek in the Early Years, compiled
by W. R. Punzman, M. Grun, L. Aguila, and A. Phillips, pp. 273–316.
290 R EFERENCES
Cultural Resource Report No. 15. Archaeological Consulting Services,

Ltd., Tucson, Arizona.
Rhode, D.
1988 Measurements of Archaeological Diversity and the Sample–Size Effect.
Rice, P. M.
1987 Pottery Analysis: A Sourcebook. University of Chicago Press, Chicago.
Rinaldo, J. B.
1940 Artifacts. In The SU Site: Excavations at Mogollon Village, Western New
Mexico, 1939, by P. S. Martin, pp. 34–77. Anthropological Series
No. 32(1). Publication No. 476. Field Museum of Natural History,
Chicago.
1943 Artifacts. In The SU Site: Excavations at a Mogollon Village, Western New
Mexico, by P. S. Martin. Second Season, 1941. Fieldiana: Anthropology
32:2:171–249. Field Museum of Natural History, Chicago.
1952 Specimens of Stone, Bone, and Clay. In Mogollon Cultural Continuity and
Change: The Stratigraphic Analysis of Tularosa and Cordova Caves, by P. S.
Martin, J. B. Rinaldo, E. Bluhm, H. C. Cutler, and R. Grange Jr. Fieldiana:
Anthropology 40:102–204.
1956 The Artifacts of Higgins Flat Pueblo. In Higgins Flat Pueblo, Western New
Mexico, by P. S. Martin, J. B. Rinaldo, E. A. Bluhm, and H. C. Cutler, pp.
55–128. Fieldiana: Anthropology 45. Field Museum of Natural History,
Chicago.
1957 Artifacts. In Late Mogollon Communities, Four Sites of the Tularosa Phase,
Western New Mexico, by P. S. Martin, J. B. Rinaldo, and E. R. Barter, pp.
39–88. Fieldiana: Anthropology 49:1. Field Museum of Natural History,
Chicago.
1959 Foote Canyon Pueblo, Eastern Arizona. Fieldiana: Anthropology 49:2. Field
1964 Artifacts. In Chapters in the Prehistory of Eastern Arizona, II, by P. S. Martin,
J. B. Rinaldo, W. A. Longacre, L. G. Freeman Jr., J. A. Brown, R. H. Hevly,
and M. E. Cooley, pp. 63–109. Fieldiana: Anthropology 35. Field Museum
of Natural History, Chicago.
Roberts, F. H. H., Jr.
1929 Shabik’eshchee Village: A Late Basketmaker Site in Chaco Canyon, New Mexico.
Bureau of American Ethnology Bulletin No. 92. U.S. Government Print-
ing Office, Washington, D.C.
1931 The Ruins at Kiatuthlanna, Eastern Arizona. Bureau of American Ethnology
Bulletin 100. U.S. Government Printing Office, Washington, D.C.
1932 The Village of the Great Kivas on the Zuni Reservation, New Mexico. Bureau
of American Ethnology Bulletin 111. U.S. Government Printing Office,
Washington, D.C.
1940 Archaeological Remains in the Whitewater District, Eastern Arizona; Part I:
Artifacts and Burials. Bureau of American Ethnology Bulletin 126. U.S.
Government Printing Office, Washington, D.C.
Robinson, G. K.
2000 Practical Strategies for Experimenting. John Wiley and Sons, Chichester,
New York.
R EFERENCES 291
Rohn, A. H.
1971 Mug House, Mesa Verde National Park, Colorado (Weatherhill Mesa Excava-
tions). Archaeological Research Series No. 7-D. U.S. Department of the
Interior, National Park Service, Washington, D.C.
Rowe, R.
1995 Towards Cupstone Classification: An Experimental Approach. Ohio Ar-
chaeologist 45(3):11–17.
Russell, F.
1975 The Pima Indians. Reprinted. University of Arizona Press, Tucson. Origi-
nally published 1908, in Annual Report of the Bureau of American Ethnol-
ogy, vol. 26. U.S. Government Printing Office, Washington, D.C.
Sayles, E. B.
1945 The San Simon Branch, Excavations at Cave Creek and in the San Simon Valley,
I: Material Culture. Medallion Papers No. 34. Gila Pueblo, Globe, Arizona.
1983 The Cochise Cultural Sequence in Southeastern Arizona. Anthropological
Papers No. 42. University of Arizona Press, Tucson.
Sayles, E. B., and E. Antevs
1941 The Cochise Culture. Medallion Papers No. 29. Gila Pueblo, Globe, Arizona.
Sayles, T., and G. Sayles
1948 The Pottery of Ida Redbird. Arizona Highways (January):28–31.
Schelberg, J. D.
1997 The Metates of Chaco Canyon, New Mexico. In Ceramics, Lithics, and Orna-
ments of Chaco Canyon, vol. 3, edited by J. Mathien, pp. 1013–1117. Publi-
cations in Archaeology 18G, Chaco Canyon Studies. National Park Ser-
vice, Santa Fe.
Schiffer, M. B.
1987 Formation Processes of the Archaeological Record. University of New Mexico
Press, Albuquerque.
Schiffer, M. B., and J. M. Skibo
1987 Theory and Experiment in the Study of Technological Change. Current
Anthropology 28:595–622.
Schlanger, S. H.
1991 On Manos, Metates, and the History of Site Occupations. American Antiq-
uity 56(3):460–473.
Schneider, J. S.
1996 Quarrying and Production of Milling Implements of Antelope Hills, Ari-
zona. Journal of Field Archaeology 23:299–311.
Seibert, S.
1987 An Analysis of the Hoes. In The Second Season’s Survey of the Homol’ovi
Ruins State Park, Northeastern Arizona, prepared by R. C. Lange, M. T.
Stark, L. Fratt, L. C. Young, and S. Seibert, Appendix H. Manuscript on
file, Arizona State Museum, University of Arizona, Tucson.
Shepard, A. O.
1956 Ceramics for the Archaeologist. Publication No. 609. Carnegie Institution,
Washington, D.C.
Simon, A. W., and P. H. McCartney
1994 Ground-Stone Artifacts of the Livingston Sites. In Archaeology of the Salado
in the Livingston Area of Tonto Basin, Roosevelt Platform Mound Study, by
292 R EFERENCES
D. Jacobs, pp. 781–798. Roosevelt Monograph Series 3, Anthropological

Field Studies 32. Office of Cultural Resource Management, Arizona State
University, Tempe.
Simon, A. W., and G. E. Rice
1996 Ground-Stone Artifacts from the School House Point Mound, U:8:24/
13c. In The Place of the Storehouses, Roosevelt Platform Mound Study: Report
on Schoolhouse Point Mound, Pinto Creek Complex, by Owen Lindauer, pp.
549–582. Roosevelt Monograph Series 6, Anthropological Field Studies
35. Office of Cultural Resource Management, Department of Anthropol-
ogy, Arizona State University, Tempe.
Simpson, R. D.
1953 The Hopi Indians. Southwest Museum Leaflets 25. Southwest Museum,
Highland Park, Los Angeles.
Sliva, R. J.
1999 Cienega Points and Late Archaic Period Chronology in the Southern
Southwest. Kiva 64(3):339–367.
Smiley, T. L.
1952 Four Late Prehistoric Kivas at Point of Pines, Arizona. University of Arizona
Bulletin 23:3. Social Sciences Bulletin 21. University of Arizona, Tucson.
Smith, W.
1952 Excavations in Big Hawk Valley, Wupatki National Monument, Arizona. Bul-
letin 24. Museum of Northern Arizona, Flagstaff.
Smith, W., R. B. Woodbury, and N. F. Woodbury
1966 The Excavation of Hawikuh by Frederick Webb Hodge. Contributions from
the Museum of the American Indian Heye Foundation No. 20. Museum
of the American Indian Heye Foundation, New York.
Spier, L.
1928 Havasupai Ethnography. Anthropological Papers of the American Museum
of Natural History 29:81–392.
1933 Yuman Tribes of the Gila River. University of Chicago Press, Chicago.
Stahl, A. B.
1993 Concepts of Time and Approaches to Analogic Reasoning in Historical
Perspective. American Antiquity 58(2):235–260.
Stanislawski, M. B.
1963 Wupatki Pueblo: A Study in Cultural Fusion and Change in Sinagua and Hopi
Prehistory. Unpublished Ph.D. dissertation, Department of Anthropology,
University of Arizona, Tucson.
1978 If Pots Were Mortal. In Explorations in Ethnoarchaeology, edited by R. A.
Gould, pp. 201–227. School of American Research, University of New
Mexico Press, Albuquerque.
Stephen, A. M.
1936 Hopi Journal of Alexander M. Stephen, edited by E. C. Parson. Columbia
University Press, New York.
Stevenson, M. C.
1904 The Zuni Indians: Their Mythology, Esoteric Fraternities, and Ceremonies.
In Annual Report of the Bureau of American Ethnology, vol. 23, pp. 3–634.
U.S. Government Printing Office, Washington, D.C.
R EFERENCES 293
Stone, T.
1994 The Impact of Raw-Material Scarcity on Ground-Stone Manufacture and
Use: An Example for the Phoenix Basin Hohokam. American Antiquity
59:4:680–694.
Strong, P. T.
1979 San Felipe Pueblo. In Southwest, edited by A. Ortiz, pp. 390–406. Handbook
of North American Indians, vol. 9. Smithsonian Institution, Washington, D.C.
Stubbs, S. A., and W. S. Stallings Jr.
1953 The Excavation of Pindi Pueblo, New Mexico. Monographs of the School of
American Reserach and the Laboratory of Anthropology 18. Santa Fe.
Swannack, J. D., Jr.
1969 Big Juniper House: Mesa Verde National Park, Colorado. Archaeological Re-
search Series Number 7-C, Wetherill Mesa Excavations. National Park
Service, Washington, D.C.
Szeri, A. Z.
1980 Tribology: Friction, Lubrication, and Wear. Hemisphere Publishing, New
York.
Teague, G. A.
1981 The Nonflaked Stone Artifacts from Las Colinas. In The 1968 Excavations
at Mound 8, Las Colinas Ruins Group, Phoenix, Arizona, edited by L. C.
Hammack and A. P. Sullivan, pp. 201–247. Archaeological Series 154,
Arizona State Museum, Tucson.
Teague, L. S.
1998 Textiles in Southwestern Prehistory. University of New Mexico Press,
Albuquerque.
Teer, D. G., and R. D. Arnell
1975 Wear. In Principles of Tribology, edited by J. Halling, pp. 94–107. MacMillan
Press, London.
Titiev, M.
1972 The Hopi Indians of Old Oraibi: Change and Continuity. University of Michi-
gan Press, Ann Arbor.
Towner, R. H.
1994 Lithic Artifacts. In The Roosevelt Rural Sites Study, vol. 2, Prehistoric Rural
Settlements in the Tonto Basin, edited by R. Ciolek-Torrello, S. D. Shelley,
and S. Benaron, pp. 469–534. Technical Series No. 28. Statistical Research,
Inc., Tucson.
Tringham, R., G. Cooper, G. Odell, B. Voytek, and A. Whitman
1974 Experimentation in the Formation of Edge Damage: A New Approach to
Lithic Analysis. Journal of Field Archaeology 1:171–196.
Udall, L.
1969 Me and Mine: The Life Story of Helen Sekaquaptewa. The University of Ari-
zona Press, Tucson.
Underhill, R. M.
1939 Social Organization of the Papago Indians. Columbia University Press, New
York.
1944 Pueblo Crafts. Bureau of Indian Affairs, Branch of Education, Washington,
D.C.
294 R EFERENCES
1946 Work-a-day Life of the Pueblos. Indian Life and Customs, No. 4. U.S. Indian
Service, Education Division, Phoenix.
1979 Papago Woman. Waveland Press, Prospect Heights, Illinois.
Unger-Hamilton, R.
1984 The Formation of Use-wear Polish on Flint: Beyond the “Deposit Versus
Abrasion” Controversy. Journal of Archaeological Science 11:91–98.
Vaughan, P. C.
1985 Use-Wear Analysis of Flaked Stone Tools. University of Arizona Press, Tucson.
Voth, H. R.
1903 The Oraibi Summer Snake Ceremony. Field Museum of Natural History,
Anthropological Series, vol. 3, no. 4. Chicago.
1912 The Oraibi Marau Ceremony. Field Museum of Natural History Publica-
tion 156, Anthropological Series 11:1. Field Museum of Natural History,
Chicago.
Walker, W. H.
1995 Ceremonial Trash? In Expanding Archaeology, edited by J. M. Skibo, W. H.
Walker, and A. E. Nielson, pp. 67–79. University of Utah Press, Salt Lake
City.
Wallace, E., and E. A. Hoebel
1952 The Comanches: Lords of the South Plains. University of Oklahoma Press,
Norman.
Watson, S. T.
1997 Ground-Stone Artifacts from U:4:33/132: The Cline Terrace Mound. In A
Salado Platform Mound on Tonto Creek, Roosevelt Platform Mound Study; Re-
port on the Cline Terrace Mound, Cline Terrace Complex, by David Jacobs,
pp. 409–430. Roosevelt Monograph Series 7, Anthropological Field Stud-
ies 36. Office of Cultural Resource Managment, Department of Anthro-
pology, Arizona State University, Tempe.
Webb, G.
1959 A Pima Remembers. University of Arizona Press, Tucson.
Wendorf, F. A.
1950 A Report on the Excavation of a Small Ruin near Point of Pines, East Central
Arizona. Social Science Bulletin No. 19, University of Arizona Bulletin
No. 21(3). University of Arizona, Tucson.
1953a Archaeological Studies in the Petrified Forest National Monument. Bulletin
No. 27. Museum of Northern Arizona, Flagstaff.
1953b Excavations at Te’ewi. In Salvage Archaeology in the Chama Valley, New
Mexico, assembled by F. Wendorf, pp. 34–124. Monographs of the School
of American Research No. 17. Santa Fe.
Whallon, R., and J. A. Brown
1982 Essays on Archaeological Typology. Center for American Archaeology Press,
Evanston, Illinois.
Wheat, J. B.
1954 Crooked Ridge Village (Arizona W:10:15). University of Arizona Bulletin
25(3), Social Science Bulletin 24. University of Arizona Press, Tucson.
1955 Mogollon Culture Prior to A . D . 1000. Memoirs No. 82. American Anthro-
pological Association, Menasha, Wisconsin.
R EFERENCES 295
Wheeler, R. P.
1980 Stone Artifacts and Minerals. In Long House, Mesa Verde National Park,
Colorado, edited by G. S. Cattanack Jr., pp. 243–305. Publications in Ar-
chaeology 7H, Wetherill Mesa Studies. National Park Service, Washing-
ton, D.C.
Wilcox, D. R.
1987 Frank Midvale’s Investigation of the Site of La Ciudad. Anthropological Field
Studies No. 19(4). Arizona State University, Tempe.
Wills, W. H.
1988 Early Prehistoric Agriculture in the American Southwest. School of American
Research Press, Santa Fe.
Wood, N. E.
1982 Ground Stone Analysis. In Pool of Cochiti Lake New Mexico, edited by L.
Hubbell and D. Traylor, pp. 177–238. Southwest Cultural Resources Cen-
ter, National Park Service, Denver.
Woodbury, R. B.
1954 Prehistoric Stone Implements of Northeastern Arizona. Papers of the Peabody
Museum of American Archaeology and Ethnology No. 34. Harvard Uni-
versity, Cambridge.
Woodbury, R. B., and E. B. W. Zubrow
1979 Agricultural Beginnings, 2000 b.c.–a.d. 500. In Southwest, edited by A.
Ortiz, pp. 43–60. Handbook of North American Indians, vol. 9. Smithsonian
Institution, Washington, D.C.
Woosley, A. I., and A. J. McIntyre
1996 Mimbres Mogollon Archaeology. Archaeology Series 10, Amerind Founda-
tion, University of New Mexico Press, Albuquerque.
Wright, K.
1994 Ground-Stone Tools and Hunter-Gatherer Subsistence in Southwest Asia:
Implications for the Transition to Farming. American Antiquity 52:238–262.
Wright, M. K.
1993 Simulated Use of Experimental Maize Grinding Tools from Southwestern
Colorado. Kiva 58(3):345–355.
Wylie, A.
1982 An Analogy by Any Other Name Is Just As Analogical: A Commentary
of the Gould-Watson Dialogue. Journal of Anthropological Archaeology 1:
382–401.
1985 The Reaction Against Analogy. In Advances in Archaeological Method and
Theory, vol. 8, edited by M. B. Schiffer, pp. 63–111. Academic Press, New
York.
Yohe, R. M., M. E. Newman, and J. S. Schneider
1991 Immunological Identification of Small-Mammal Proteins on Aboriginal
Milling Equipment. American Antiquity 56(4):659–666.
Zimmerman, H.-J.
1996 Fuzzy Set Theory and Its Applications. Kluwer Academic Publishers, Boston.
Figures
1.1. Map of the U.S. Southwest indicating areas and districts men-
tioned throughout the volume. 4
2.1. Multiple-use tool, concomitant secondary use. 23
2.2. Multiple-use tool, sequential secondary use. 24
2.3. Schematic of an unaltered stone surface. 29
2.4. Schematic of a surface damaged through fatigue wear. 30
2.5. Schematic of a surface damaged through abrasive wear. 31
2.6. Schematic of a surface damaged through tribochemical wear. 32
2.7. Experimental abrader with V-shaped grooves used to sharpen
sheep bones. 34
2.8. Experimental flat abrader used to smooth a digging stick. 34
2.9. Experimental mano and metate used to grind dried feed
corn. 35
2.10. Photomicrograph of use wear caused by grinding dried feed
corn. 36
2.11. Photomicrograph of use wear caused by grinding sunflower
seeds. 37
2.12. Photomicrograph of a stone surface used against a pliable
surface. 38
2.13. Photomicrograph of an unused surface. 39
2.14. Photomicrograph of use wear caused by working a hide. 40
2.15. Axes and associated netherstones that were probably used in their
manufacture. 46
2.16. Bar chart illustrating the activities represented at three sites. 50
3.1. Experimental manos and metates used to grind dried feed
corn. 67
4.1. Diagram of abraders, smoothers, and polishers expressed as sets
and subsets. 78
4.2. Faceted flat abrader from the Point of Pines area. 82
4.3. Abrader with V-shaped grooves, expediently designed. 83
4.4. Grooved abrader or shaft straightener, strategically designed and
embellished with a ridge. 85
297
298 L IST OF F IGURES
4.5. Grooved abraders with stone tools that might have been shaped
in the grooves. 86
4.6. Diagram of handstone, netherstone, and lapstone types expressed
as sets and subsets. 90
4.7. Polisher, expediently designed. 92
4.8. Floor polisher. 95
5.1. Basin metate. 101
5.2. Schematic of a circular stroke used in a basin metate. 102
5.3. Schematic of a reciprocal stroke used in a basin metate. 103
5.4. Basin mano. 104
5.5. Flat/concave metate. 105
5.6. Flat/concave mano. 107
5.7. Schematic drawing of mano surface configurations. 108
5.8. Open-trough metate. 109
5.9. Trough mano. 111
5.10. Metates with compatible manos. 112
5.11. Open-trough and flat metates in multiple bins. 113
5.12. Mano profiles. 113
5.13. Adobe grinding receptacle. 117
5.14. Slab-lined mealing bins. 118
5.15. Relative time line illustrating the occurrence of various metate de-
signs in different areas around the U.S. Southwest. 122
5.16. Shaped mortar. 129
5.17. Rock mortar with a compatible pestle. 130
5.18. Shaped mortar. 131
5.19. Disk mortar. 133
5.20. A disk mortar and a mortar with a handle. 134
5.21. Eagle watering bowl. 136
5.22. Pestles from Santa Cruz Bend. 139
5.23. Pestle used in more than one activity. 141
5.24. Netherstone. 144
5.25. Palette. 147
5.26. Terms used to describe palette attributes. 149
6.1. Pottery anvil. 156
6.2. Lithic anvil. 158
7.1. Diagram of hafted tool types expressed in terms of sets and
subsets. 161
7.2. Terms used to describe hafted percussion tools. 162
7.3. Axe head with 3/4-groove design. 164
7.4. Profiles of bit edges in various stages of wear. 166
L IST OF F IGURES 299
7.5. Three types of handles attached to axe heads. 167

7.6 Hoe with notches for hafting. 178
8.1. Fire-drill hearth with cupule. 181
9.1. Stone perforating tools. 186
9.2. Plane with handles. 189
10.1. A possible atlatl finger-loop. 193
10.2. Balls with evidence of being wrapped, perhaps for hafting. 196
10.3. Cruciform. 197
10.4. Perforated stone disk or donut stone. 202
10.5. Conical pipe with a biconical bore. 207
10.6. Pipe with a bone stem in the bore. 208
10.7. Natural shape. 212
10.8. Figurine. 216
11.1. Diagram of containers and mortars expressed as sets and
subsets. 219
11.2. Plain bowl with a flat base. 220
11.3. Tray with bifurcate knobs. 222
11.4. Bowl typically classified as a censer. 224
12.1. Loomblock with sockets for anchoring an upright loom. 228
Tables
Table 3.1. Results of grinding experiments conducted with different

mano/metate designs. 69
Table 5.1. Statistical description of mano types. 106
300
Acknowledgments
A book such as this is the creation of an army of people. Various versions

of the manuscript were read by Sally Bird, Donna Breckenridge, Chip Col-
well, Otehlia Kiser, Lisa Piper, and Arthur Vokes, and their comments
were greatly appreciated. Other reviewers’ comments were extremely help-
ful, especially those from John Clark and Tammy Stone. The Arizona State
Museum librarians, Mary Graham and Dianna Thor, helped me in innu-
merable ways with references. Several artists rendered illustrations. Chip
Colwell produced most of them, with contributions by Ron Beckwith, Rob
Ciaccio, Doug Gann, Catherine Gilman, Devin White, and Mike Donald-
son. Scott Van Keuren performed much appreciated emergency photogra-
phy. Emilee Mead rescued me from problems with format and style.
When it came down to the end, Taylor and Jim Johnson helped me wade
through the details of checking references and figure citations, and Alexis
Mills performed the final editing. I would also like to thank Bill Doelle
and the Center for Desert Archaeology for publishing the original version,
and Jeff Grathwohl with the University of Utah Press for encouraging me
to publish a revised version.
301
Index
abrade (-ing), 1, 20, 39, 48, 74, 77, 79, 84, 87, Ash Flat Cliff Dwelling, 53
148, 153, 185, 187, 215 asperity, 28, 30, 32, 36, 38, 77, 86, 93–94
abrader, 1, 19, 32, 34, 38, 41, 54, 77–89, 91, atlatl, 191–193, 198
98, 140, 142, 145, 184–185, 188–189, 203, Awatovi, 4, 11, 126, 137–138, 148, 173, 179,
213: arrow-shaft, 88; bar-shaped, 80; 200
faceted, 81–82; flat, 33–34, 81–82, 88–89, awl, 33, 37, 53, 62, 82, 84, 184–186
213; grooved, 23–24, 33, 53–54, 80, 82, 84– axe, 1, 22, 26–27, 42, 45–46, 40, 53, 55–56,
89, 91, 213; irregular, 80; tabular, 80; T- 63–64, 81, 94, 160–167, 169–177, 203:
shaped, 91; U-shaped, 80, 82, 84, 86, 89, 3/4-, 165–172; double-bit, 168–169; full,
185; V-shaped, 80, 82–84, 87, 89, 227; 160, 167–168; hafted, 163, 170; notched,
wedge, 81 166, 171–172; prehistoric, 64; single-bit,
abrasion (-ive), 1, 23, 28, 30, 35, 38, 40–41, 168–169; spiral, 162, 168; stone, 61–62;
86, 98, 110, 138, 143, 145, 153, 187–188, wedge, 162, 168, 170
194, 215–216 Aztec, 4, 126, 173
Acoma Pueblo, 10, 216
adze, 160–161, 168–169, 176–177 Babocomari Village, 5, 154
agave, 42, 188–190, 199 Badger House, 4, 123, 138, 195, 205, 225–226
Agricultural period, Early, 57, 198 Bailey, 5, 116
agriculture, 120–121, 166 Balcony House, 226
Agua Caliente, 195 ball, 103, 191, 193–196
Agua Fria, 5, 157 Bandelier, 4, 195
Alkali Ridge, 4, 116, 123–124, 171, 175–176, basalt, 119, 140: vesicular, 66, 69–70, 182,
179 198–200, 205, 208
altar, 226–227 Basketmaker, 200, 205, 208: III, 171, 174
amaranth, 33, 37 Bat Cave, 5, 121
Anasazi, 11, 56, 74, 116, 125, 127, 148, 172, bead, 33, 143, 212–215
174–175, 178, 208, 225 Bear Ruin, 5, 123, 174
Angell phase, 171 bell stone, 195–196
anvil, 136, 155, 157, 159, 204: lithic, 137, 143, Big Hawk Valley, 4, 172, 200
157–159; nutting, 157; pottery, 52, 155–157 Big Juniper House, 4, 138, 163, 187
Apache, 10, 88 billet, 151
Apache Creek, 5, 125 bin, 74, 112–113, 116, 124–127, 226: slab-
Archaic, 208: Late, 152, 192, 195; Middle, lined, 118, 124
121 Blackfeet, 97
Arizona, 4–5, 11, 20, 87, 89, 115, 120–125, 135, blank, 20–21
138, 140, 154, 169–171, 173–174, 177, 179, Bluff, 5, 123, 221
182, 192, 198, 200, 205, 213, 225, 230–231 Boatyard site, 5
arrow, 54, 82, 84, 88: shaft, 33, 37, 87; bone, 31, 33–34, 37–39, 52, 80, 82, 91, 93, 190,
straightener, 53, 88 213, 215
303
304 I NDEX
bowl, 116, 125, 130, 135, 218–221, 223–224, comal, 227, 230–231
229: eagle watering, 7, 135–136; plain, 131 Comanche, 97
building stone, 22–23, 43, 163 construct: analytical, 5–6, 11, 19, 25, 56;
bullroarer, 60, 201 behavioral, 5–6, 11, 17, 19, 21, 27;
burial, 22, 39, 148, 150, 198, 214, 223 relational, 19
button, 215, 216 container, 1, 7, 15, 74, 130–132, 137, 199, 201,
buzzer, 60, 201 218–220, 223, 225: closure, 199, 201, 218
Cordova Cave, 5, 121
cache, 51, 142, 170, 198, 211, 213 corn, 33, 35–37, 181, 203, 229
California, 20, 88, 132, 135, 138, 203 cottonwood, 180
Canyon Creek, 5, 126, 200 cremation (-ed), 43, 148, 150, 223
cap, 218–219, 223, 225 Crooked Ridge Village, 5, 123, 142, 170, 176
Carter Ranch Pueblo, 5, 53, 172, 225 Crow, 96
Casa Blanca, 194 cruciform, 43, 94, 196–198
Casas Grandes, 5, 12, 89, 91, 122–123, 135, crush (-ed, -ing), 42, 50, 99, 127–128, 138,
154, 169–171, 177, 182–183, 195–196, 205, 140, 154
214, 216, 221 crusher, 154–155
Cave Creek, 221 cupped stone, 12, 98, 130, 136–137
censer, 218–219, 223–224 cupule, 128, 136–137, 157, 180–182
Ceramic period, Early, 192, 195, 201 Cuyamungue, 5, 196
Chaco, 4, 108, 120, 123–126, 172, 175, 209 cylinder, 198–200, 205, 223
Chama Valley, 4, 126, 204, 230–231
charmstone, 192–193 Deadman’s Fort, 4, 200
Chenopodium, 63 design, 2–3, 8–9, 11, 15, 18–19, 21–22, 25–27,
Chihuahua, 89, 198 42, 44, 47, 55–58, 60, 65–66, 70, 73–75, 79–
chisel, 153–154, 188–189 80, 89, 93, 100, 112, 121, 138, 143, 146,
chopper, 1, 151, 153 159–160, 163, 170, 174, 177, 184, 200–201,
chopping, 42, 160, 166 221, 223: expedient, 15, 21, 49, 51, 80, 82–
Chumash, 203 83, 91–94, 99, 127–128, 138, 142–143, 145,
Cienega phase, 89, 129–130, 140, 195, 198 151, 153–155, 157–158, 173, 175, 177, 182,
Circle Prairie phase, 170 185, 189, 193, 204–205, 225, 227, 229;
Classic period, 89, 124, 170–171, 177: early, strategic, 15, 21, 49, 80, 91, 94, 96, 99, 127–
87, 89, 124 128, 138, 142, 145, 151, 153–155, 157, 163,
classification (strategy), 1–2, 9, 11–15, 17, 43, 176, 182–184, 188–189, 193, 196, 205, 218,
73–74, 79, 89, 100, 124, 130, 133, 135, 143, 225–227, 229, 231; theory, 8–9, 54
146, 174, 177, 199, 213 digging stick, 33–34, 37, 53, 82, 203
clay, 1, 31, 33, 35, 37, 134, 138, 143, 155, 165, digging tool, 142, 176
183, 210, 221, 231 disk, 132, 183, 200–201, 203, 215, 225:
cliff dwelling, 163, 182–183 notched, 12; perforated, 132, 201
Cochise culture, 120–121 donut stone, 201–204
Cochiti, 5 Double Adobe, 5, 121
Coconino National Forest, 223 drill, 84, 181–183, 205, 214: bow, 137; fire,
Cocopah, 192 180; pump, 137
Cohonina, 157, 199 Duckfoot, 4, 64, 115, 123, 171
coil-and-scrape (method), 52, 94
Colonial period, 148, 175: early, 170 efficiency, 27, 36, 47, 117, 119–120, 153
Colorado, 4, 64, 115, 123, 138, 171, 200 Elden phase, 171, 200
Colorado Plateau, 4, 12, 123, 178 Elden Pueblo, 4, 171, 200
I NDEX 305
ethnography (-ic), 3, 9–11, 15, 18, 20, 50, 54– griddle, 50, 52, 227, 229–231
55, 58–64, 68, 75, 87–89, 93, 95–96, 99, 116, Griffin Wash, 5
132, 134–135, 138, 140, 142, 145, 150, 152– grind (-ing), 1, 3, 6, 20, 23, 25–26, 35–37, 39,
155, 169, 175, 179–180, 182, 184, 191, 193– 41, 48, 50, 54–55, 61–62, 66–67, 70, 74, 98–
195, 198, 203, 206, 211, 213, 216, 229 101, 116–121, 138–140, 143, 145–146, 148,
153–154, 176, 188, 191, 196, 201, 205, 209,
fergolith, 154 213, 216: bin, 60; slab, 32, 98, 145; station,
fibrolite, 173 115, 116, 127; movable, 115; multiple, 115;
figurine, 1, 216–217 permanent, 115–116; single, 115; stone, 27;
file, 62, 79–80, 184, 187–188, 190, 213, 215 surface, 5, 8, 27, 43; technology, 49, 52, 63
fire-cracked rock, 44, 50, 52, 227, 231–232 Guatemalan Highlands, 8
firedog, 50, 52, 227, 231
fire-drill hearth, 136–137, 180–182 hafted tools, 160, 162
Flagstaff, 4, 89, 110, 123–125, 157, 170–172, hairpin, 84
177, 182, 198–199, 223, 225, 231 hammer, 62, 165, 173–174: grooved, 174–175
flake scars, 20, 138, 1 51, 153–154, 175, 188, hammerstone, 1, 19–20, 42–43, 138, 142,
197 151–154, 157, 163, 173, 198, 205
flake (-ed, -ing), 1, 153, 176, 197, 201 handstone, 5–7, 13, 32–33, 40–41, 45, 79–80,
Flattop, 4, 121 89–91, 96–98, 100, 123, 138, 140, 142–146,
flute, 54 151–152, 155, 157, 209–210
Foote Canyon Pueblo, 5, 53, 100, 125, 163, Havasupai, 88, 115, 210
231 Hawikuh, 5, 126
Forestdale, 5 hearth, 180, 226, 229, 231
Forked Lightning, 5, 204 hide, 6–7, 31, 33, 39–41, 48, 96, 190
form, 6–8, 12, 15, 57, 60, 81, 120 hide-processing stone, 7, 96–98, 142
Four Corners, 4, 165, 168, 171, 175–176, 179, Higgins Flat Pueblo, 5, 43, 124–125, 163,
192, 198 172, 187, 221, 226
Fremont, San Rafael, 124 historic, 9–10, 61, 88–89, 115, 130, 132, 135,
140, 148, 150, 152, 155, 169, 173, 175, 179,
Galaz Ruin, 5, 121, 148, 150, 169, 171–172, 182–183, 194–195, 204, 210, 214, 227
221 Hodges Ruin, 5, 120, 154, 157, 187, 204
Gallina, 89 hoe, 50, 160–161, 165, 177–179: blade, 177;
gaming piece/stone, 33, 192, 194–195, 197, tabular, 184
201, 203 Hohokam, 74, 120, 135, 146, 148, 150, 155,
Gibbon Springs, 5, 124 157, 170, 172, 174, 199, 205, 208, 230
Gila Basin, 168, 217 Homol’ovi, 4, 20, 178, 211, 226; I, 230; II, 227
Glasscock, 4 Hooper Ranch Pueblo, 5, 91, 172, 226, 231
Gnat Haven, 4, 124 Hopi, 4, 7, 68, 89, 94, 96, 114, 135, 169, 173,
gong, 196 175, 178–181, 183, 194, 201, 206, 208, 210–
gouge, 31–32, 38 212, 216, 227, 230
grain, 66, 99, 114, 118–119, 121 hornstone, 179
Gran Quivera, 5, 84, 88–89, 176, 196, 204,
208, 225–226, 230–231 impact fracture, 20, 23, 30, 36, 41–43, 98,
Granary Row, 5 110, 120, 127–128, 135–139, 142–143, 145,
Grand Canyon, 210 147, 151–154, 157–159, 163, 165, 173, 175,
Grand Falls, 4, 200 188, 194, 203–204
granite, 66, 69, 119 interstices, 28–29, 33, 35–38, 40, 54, 96
Great Basin, 88 Isleta Pueblo, 214
306 I NDEX
Jeddito, 4, 88, 123, 200, 206 11; one-hand, 7, 13, 100, 106, 120, 123;
Valley, 11, 89 open-trough, 67–68, 74; profile, 112–114;
Julian Wash, 43, 148 rocker, 11, 13, 109; small, 100; trough, 23,
Juniper Terrace, 4, 200 25, 66, 70, 106, 109–112, 121; two-hand, 7,
13, 100, 120; Type I, 100; Type II, 100
Kiatuthlanna, 5, 110, 135, 207, 221 Maricopa, 10, 82, 88–89, 115, 135, 138
kickball, 194–195, 203 material culture, 11, 18, 54, 62
kinetics, 17–18, 41–42, 45, 56–57, 62–64, 66, mattock, 160–161, 177
86, 100, 143, 151 maul, 152–154, 160–163, 165, 169, 173–176:
Kinishba, 5, 217 3/4-, 174–176; full, 171–172, 175
Kinnikinnick, 4, 157 Maverick Mountain phase, 125–126, 172,
knife, 179, 184: agave, 184, 190 203
Kua’ua, 5, 196 Meddler Point, 5, 148
Medicine Fort, 4, 200
La Plata, 4, 115, 123, 179 medicine stone, 191, 198–200
Laguna Pueblo, 10, 216 Medio period, 89
lapstone, 45, 74, 79–80, 89–91, 98, 142–143, Mesa Verde, 4, 89, 123–125, 138, 172–173,
145–147, 151, 157, 204–205, 209, 213, 218 175, 179, 182–183, 187, 195
Las Acequias, 5, 170 mesquite, 211: beans, 135, 140; pods, 50,
Las Capas, 5, 43, 198, 205, 208, 210 135, 140
Las Colinas, 5, 124, 148, 199 metate, 1, 8, 15, 19–26, 32–33, 35, 41–43, 45,
Las Humanas, 5, 204 49–51, 54–57, 60, 62–68, 98–100, 102–103,
Leaf Water, 4, 230 105–108, 110–126, 135–136, 140, 142–143,
lid, 201, 218–219, 223, 225 152–153, 169, 209, 231–232: 3/4-, 100;
life history, 2–3, 18, 64, 73, 80, 100, 113–114 basin, 13, 51, 57, 66–70, 100–103, 108, 110,
lightning stone, 204 115, 120–121, 123; 3/4-, 100, 108, 123;
lintel, 226 open-, 100, 108; block, 100, 103; flat, 74,
Little Colorado River, 5, 123, 172, 175, 226 103, 112–113, 115–116, 118, 124–127; flat/
Lizard Man Village, 4, 157, 171, 199–200 concave, 51, 57, 66–70, 100, 103, 105–106,
Long House, 4, 89, 168, 179, 187, 211, 226 115, 121; slab, 100, 103; trough, 24–25,
loomblock, 227–228 66–70, 106, 108–110, 112, 115, 120–121,
Los Hermanos, 46 123–127; 3/4-, 110, 112, 121, 123–124, 126;
Los Morteros, 5, 148 closed-, 110; open-, 74, 100, 109–110, 113,
Los Muertos, 5, 89, 163, 169–170, 176–177, 117–118, 121, 123–126; Utah-, 110, 123–
187, 200 124; Type I, 100; Type II, 100
Los Pozos, 5, 195, 210, 221 Mexico, 169–171, 174, 192, 230
Los Tortugas, 5 microtopography, 29, 37
migration, 55, 126
maize, 13, 35, 61, 63, 65–66, 100, 116, 120– Mimbres, 5, 148, 150, 171, 221
121, 143 Mineral Creek Pueblo, 5
mallet, 173 Mogollon, 11, 74, 89, 120, 124–125, 127, 135,
mano, 1, 7, 13–14, 19, 22–26, 32–33, 35, 41– 137, 148, 170–172, 174, 206–208, 225
45, 49–51, 57, 62–68, 96, 98–100, 102–103, Mogollon Rim, 170, 227
106–110, 112–115, 117–120, 123, 135, 137– Mono, Western, 132, 134–135
138, 142–143, 153, 178, 209, 232: basin, 66– Montezuma’s Castle, 171
68, 70, 103–104, 106; cobble, 13; flat/ mortar, 1, 7, 32, 41, 45, 49–50, 98–99, 110,
concave, 67–68, 74, 103, 105–107, 110, 112, 127–128, 130, 132–141, 143, 145–146, 148,
143; food-grinding, 22; large, 14, 100; loaf, 159, 203, 210, 218–219, 221, 223: basin,
I NDEX 307
135; bedrock, 128, 130, 134–135, 140, 219; pendant, 213–215: figurine, 215
boulder, 128, 135, 219; disk-shaped, 132– percussion (-ing), 48, 74, 159, 173; tool, 160,
134, 219; food-processing, 135; pebble, 184
128, 130, 132, 135–136, 159, 219; portable, Perry Mesa, 5, 157
135; rock, 128, 130, 219; shaped, 110, 129– pestle, 1, 23, 32, 41, 49, 49–50, 98–99, 127–
132, 219; stationary, 128; wooden, 140, 142 132, 135, 137–142, 146, 154, 159, 176, 210,
mortuary, 24, 42, 146, 148, 150 218, 221: notched, 140; stone, 138, 142;
mosaic, 212–213, 215 tooth, 140–141; triangular, 132, 140
Mug House, 4, 138, 173, 225 Petrified Forest, 4, 121
petrographic analysis, 54
NA 682, 172 Phoenix, 124
NA 11,237, 4, 200 Phoenix Basin, 5, 20, 89, 121, 123, 148, 154,
NA 11,535, 4, 200 199, 204
Nalakihu, 4, 199 phyllite, 146
Native American, 10, 60–61, 208 pick, 160–161, 175–177
Navajo, 10, 96 Picuris, 5, 173
needle, 82 pigment, 1, 5–6, 22, 24, 44, 48, 54, 132, 134–
netherstone, 6, 28, 45–46, 74, 79–80, 84, 89– 135, 137, 140, 142, 144, 146–148, 175, 203,
91, 98–99, 137, 140, 142–145, 151, 154, 157, 209–210: processing, 52, 146
159, 163, 182, 209–210, 213 pikistone, 20, 50, 52, 227, 229–230
New Caves, 4, 223 Pima, 135, 140, 142, 148, 155, 183, 194, 206,
New Mexico, 4–5, 48, 84, 89, 120–126, 171– 210, 230
173, 175, 182, 192, 195–196, 200, 204, 208, Pindi Pueblo, 5, 94, 173
225, 230–232 Pine Flat Cave, 53
nixtamal, 66 Pinedale, 5, 217, 227
noisemaker, 60, 201, 216 Pinelawn, 5, 123–125
Pioneer period, 123, 170, 204
occupation strategy, 47, 49, 51 pipe, 1, 205–208: conical, 206–207;
O’odham, 10 cylindrical, 206; elbow, 206; globular, 206;
ornament, 15, 191, 200, 212–215: personal, 1, stemmed, 207–208
213 pit, 45, 115–116, 125, 136–138, 157, 225:
exterior, 43, 57; interior, 43; in-the-floor,
Paa-ko, 5, 173, 196, 204 57; roasting, 229; thermal, 52
paddle-and-anvil (technique), 52, 93, 155 pithouse, 57, 89, 115, 124, 135, 142, 148, 176,
Padre phase, 171, 199 195, 207, 221, 231
paho, 53 pitted stone, 98, 130, 136–138
Pai, 135 Pittsburg Fort, 4, 199
Paiute, 10, 88: Owens Valley, 88; Southern, plane, 176, 184, 188–189
154; Surprise Valley, 88 plug, 199–200, 215, 218–219, 223, 225
palette, 15, 43, 98, 145–150, 187, 218–219, 223 plummet, 196, 199, 204–205
Papago, 155 Point of Pines, 5, 53, 56–57, 81–82, 89, 91,
Paquimé, 5, 123, 205, 221 113, 117–118, 122–127, 142, 165, 169–170,
Patayan, 155 172, 174, 182, 195, 199–200, 203–204
peck (-ed, -ing), 25, 28, 30, 42, 94–95, 99, 110, polish (-ing), 1, 20, 31, 46, 48, 53, 61, 74, 77,
128, 130, 145, 196, 205, 216, 226–227 79, 84, 91, 96, 145, 155, 163, 165, 196–197,
pecking stone, 1, 19, 30, 42, 151–153 201, 213, 215–216
Pecos Pueblo, 5, 113, 126, 137, 171, 173, 175, polisher, 19, 78–79, 81, 89, 91–94, 98, 145:
187, 206, 225 floor, 94–95; pottery, 22, 91, 93; stone, 94
308 I NDEX
polishing stone, 1, 20, 22–24, 41, 52–53, 79, ringing stone, 196
138, 142, 201 Rio Grande, 210
Po-shu-oinga, 4, 231 Rio Grande Pueblo, 5, 10, 172
pot, 6, 52–53, 61, 93, 155, 231 Rito de los Frijoles, 5, 196
pottery, 15, 22, 48, 91, 94, 155 roasting rock, 23. See also fire-cracked rock
pound (-ing), 23, 42, 127, 139 rock art, 198
Prayer Rock, 4, 123, 208 Roosevelt Basin, 5, 163, 169, 182
prayer stick, 53, 84 rubbing stone, 79
preceramic, 121, 135, 198
pre-Classic period, 187 Sacaton Flats, 194
prehistoric, 1–2, 7, 15–18, 20, 24, 27, 42, 50, Sacaton phase, 185, 199
52, 55–57, 59, 61–63, 65, 71, 80, 93, 117, sage, 64
126–127, 132, 135, 140, 155, 166, 169–170, Salado, 74
179, 182, 196, 209–211 Salt-Gila area, 5, 177
Prickly Saguaro, 5 San Ildefonso, 204
Prieta Vista, 4, 226, 230 San Juan Basin, 171, 225
protohistoric, 173 San Juan River, 4, 122, 179
Puaray, 5, 196 San Pedro phase, 198
pueblo, 89, 125, 172, 196, 204 San Pedro Valley, 5, 124
Pueblo, 57, 89, 130, 208, 211, 229 Sand Canyon Pueblo, 64
Pueblo Bonito, 4, 124, 126, 187–188, 225–226 sandstone, 66, 69, 84, 148, 163, 180
Pueblo del Arroyo, 4, 60, 124–125, 172, 179, Sangre de Cristo Mountains, 173
209, 225 Santa Clara, 204
Pueblo period, 175: I, 170–171, 221; II, 170– Santa Cruz Bend, 5, 51, 57, 121, 129–130,
171, 176, 179, 200; III, 171, 179, 200, 227, 132, 139, 210
230; IV, 172, 179, 200, 227, 230; V, 179, 200, Santa Cruz River, 47, 195, 208
227; late, 231 saw, 62, 184, 187–188, 190, 213, 215
Pueblo Tonque, 126 schist, 146
Puebloan, 10, 43, 55, 61, 64, 88, 93, 96, 155, scoria, 198–200, 223
177, 179–180, 193–195, 200, 204, 206, 211, scrape (-ing), 48, 153, 176
214 scraper, 97, 184
Puerco Ruin, 4, 230 Second Canyon, 5, 124
pulverize (-ing), 74, 98–99, 120, 138 Sedentary period, 148, 204; post-, 230
Puye, 5, 196 seeds, 50, 66–69, 116, 120–121, 134:
amaranth, 35, 66, 69; sunflower, 33, 35–37,
quarry, 54, 138, 151–152, 154, 176–177, 229 39, 66, 69
Seri, 211
racing stone, 194 Shabik’eshchee, 4, 123, 175
rasp, 79 shaft, 82, 84, 86–89, 201: reed, 84; wooden, 84
reamer, 62, 184–185, 213, 215 shaft smoother, 87–89: elongated, 87; loaf-
receptacle, 116–117, 125, 127 shaped, 87; ridged, 87; transversely
Red Bow Cliff Dwelling, 53 grooved, 87
replication studies, 62–63 shaft straightener, 22, 82, 85–88, 140
Reserve phase, 124, 170 shaped stone, 208–209
Ridge Ruin, 4, 171, 200 sheen, 31–32, 36–37, 39–41, 52–53, 77, 79, 84,
Rillito phase, 157, 187, 204 86, 91, 93–94, 142–143, 145, 163, 182, 185,
Rincon phase, 148, 157, 168, 187 194, 196, 215
Middle, 217 shell, 82, 185, 192, 213, 215
I NDEX 309
Showlow, 5, 143, 217 Tonto Basin, 5, 46, 121, 124, 163, 169–170,
Showlow Ruin, 89, 126, 227 185, 190: lower, 89, 148, 157
Sinagua, 74, 157, 170, 199 Tonto Creek, 87, 89, 110, 112, 163
Sioux, 96 Tonto Creek Archaeological Project, 10
slab, 116–117, 125, 209, 226: cooking, 50, 52, Tortolita phase, 123, 169, 170, 209
231; painted, 226 tray, 132, 218–219, 221–222: snuff, 150
smoother, 19, 77–79, 89, 91, 98 Tres Huerfanos, 5
Snaketown, 5, 120, 123, 146–148, 157, 169– tribology (-ist), 18, 27, 58
170, 175, 177, 185, 187, 199, 204, 221, 223 trivet, 115, 231
Sonora, 192, 198 tube, 205
South America, 150 Tucson, 132, 208
Southwest (United States), 2–3, 9–10, 15, 55, Tucson Basin, 5, 43, 47, 89, 115, 121–124,
60–62, 73, 81–82, 88–89, 93, 120–122, 127, 129–130, 148, 157, 168, 170, 187, 192, 195,
132, 135, 152, 154, 168–171, 174, 177, 192, 198, 201, 204–205, 209–210, 217, 223
196, 198, 204–205, 208, 211, 213, 221 Tularosa Basin, 232
spearthrower, 192 Tularosa Cave, 5, 121, 192
spindle, 136–137, 180: base, 182–183; whorl, Tularose phase, 124, 170, 174, 217
137, 182; wooden, 84 Tule Tubs Cave, 53
spinning, 74, 84, 137, 182–183, 201 Turkey Creek Pueblo, 5, 81, 117, 125, 165, 172
Spruce Tree House, 4, 168 Turkey Foot Ridge, 5, 123
Starkweather Ruin, 5 Tuzigoot, 5, 200
Stone Axe, 4, 200 Twin Bridges, 5, 123
Stone Pipe, 5, 115, 195 typology, 12, 15, 73–74, 81, 87, 143, 152
stone ring, 201, 203–204
striations, 5, 31, 35–36, 41–42, 52, 65, 77, 79, University Ruin, 5, 154
81, 86, 93–94, 98, 102–103, 106, 110, 127– use, extensive, 26, 47, 118; intensive, 26, 47,
128, 135, 137, 139, 142, 148, 157–159, 165, 118; multiple, 22–24; primary, 2, 17, 21–
182, 185, 187–189, 197, 203–204, 215 22, 44, 48, 58, 137, 190; secondary, 2, 17,
SU, 5, 121, 200, 207, 221 21–22, 44, 48, 58, 74, 96, 137–138, 140, 154,
Sunset Mesa, 217, 223 174, 179, 190, 201, 227; concomitant, 21–
Swarts Ruin, 5, 169, 172, 187, 213, 221 23, 48, 80, 137, 190, 228; sequential, 21–22,
Sweetwater phase, 169 24, 48, 80, 137; single, 22, 48
use life, 42–43, 114, 119, 135, 165
Table Rock Pueblo, 5, 231 use-wear, 7, 22, 25, 28, 32, 36–38, 40–41, 44–
tablet, 209 45, 50, 57, 62, 73–74, 77, 79, 81, 87, 92, 94,
tabular tool, 184, 187, 189–190 100, 102, 108, 130–133, 136–137, 141–144,
Talus Village, 4, 121 152–154, 157, 159, 163, 169, 184–185, 187–
Tanque Verde phase, 168, 204 188, 190, 193, 195, 203–205, 209, 215–216,
Taos, 195 218, 220–221, 223, 231; analysis, 3, 7, 9, 20,
tchamahias, 50, 160–161, 179 22–23, 27–28, 44–45, 50, 52, 55, 64, 81–82,
Te’ewi, 4, 126 91, 93, 96, 108, 121, 123, 130, 137, 140, 142,
Tewa, 169 145, 155, 159, 168, 174, 176, 182, 187, 189,
Three Circle phase, 123 203, 223; damage, 67, 81, 112, 129, 136, 138,
thunder stone, 194–195 142, 144, 173, 189, 194, 226; patterns, 8, 10,
Tla kii, 5, 163, 168, 213 18, 23, 28–29, 33, 35, 37, 39, 41, 44–45, 52,
toggle, 215 63–64, 68, 77, 82, 91, 94, 96, 110, 127, 140,
Tohono O’odham, 169 142–143, 151, 153–154, 157, 165–166, 173,
Tonque, 5, 173 185, 189–190, 197–199, 203, 218, 221
310 I NDEX
Utah, 4, 110, 123–124, 171, 192 worn out, 11, 25; tribochemical, 27, 31–32,
Ute, 10, 96 35–37, 39–40, 77, 86, 91, 98; well used, 6
worn out, 25
Vahki phase, 123 weathering, 28, 130
Valencia Vieja, 5, 121, 123, 170, 209 wedge, 25, 154, 173
Ventana Cave, 5, 121, 135, 208 Wet Leggett, 5, 121, 123
Verde Valley, 176 whetstone, 79
vesicle, 20, 35–38, 40, 129, 199 White Dog Cave, 4, 121
Viejo period, 171 White Mountains, 217
Village of the Great Kivas, 5, 223 White Water District, 148, 195
White Water Draw, 5, 121
W:10:37, 5 whizzer, 60, 201, 216
W:10:51A,B, 5 whorl, 183, 201
Walapai, 10, 135 Williford, 4, 171
Walnut Canyon, 4, 200 willow, 211
Walpi, 4, 7, 96, 135, 173, 183, 227 Wind Mountain, 5, 123
Walpi Research Project, 10 Winona phase, 171, 199
war club, 169, 194 Winona Village, 4, 89, 200
wear, 19–20, 25–26, 28–29, 35–37, 39, 41, 45 Wupatki, 4, 171–172, 200
47–48, 58, 63, 65, 73–74, 79, 84, 89, 100,
102, 112–114, 119, 139, 190, 196, 203–204: Yavapai, 87–88
abrasive, 27, 30–32, 35–37, 77, 91; yucca, 188
adhesive, 27, 29–32, 35, 40, 77, 98; barely Yuman, 10, 82, 89, 96, 140, 155
used, 11; facet, 52, 102–104, 106, 114, 143;
fatigue, 27, 30–32, 35–37, 40, 98, 151; Zia, 5, 94, 96, 216
heavy, 25–26, 48; light, 25–26, 48; manage- Zuni, 5, 89, 123, 143, 179, 195, 201, 208, 214,
ment, 25–27; moderate, 25, 48; nearly 216, 223, 227, 229–230

Ground Stone Analysis A Technological AP

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ground Stone Analysis A Technological AP

Uploaded by

Copyright:

Available Formats

GROUND STONE ANALYSIS

Ground Stone Analysis

The University of Utah Press

Published in conjunction with the

Library of Congress Cataloging-in-Publication Data

Part 1. A Foundation for Research

2. Grinding Technology and Technological Analysis 17

3. Resources for Modeling Tool Use and Technological Behavior 59

Part 2. Artifact Descriptions 73

4. Abrading, Smoothing, and Polishing Tools 77

5. Grinding and Pulverizing Tools 98

6. Percussion Tools 151

7. Hafted Percussion Tools 160

8. Spinning Tools 180

9. Perforating, Cutting, and Scraping Tools 184

10. Paraphernalia 191

11. Containers and Container Closures 218

12. Structural Stones 226

Appendix A: General Artifact Form 233

References Cited 275

List of Figures 297

List of Tables 300

A Foundation for Research

B efore plunging into the heart of ground stone research, it is impera-

Four Corners Northern New Mexico Wupatki

East-central Arizona/ Starkweather Ruin University Ruin

The methods presented in this manual guide quantitative and quali-

distinguish American addresses from British or French addresses on the

LAYING A FOUNDATION FOR ANALYSIS

worked among Native American groups sometimes describe the tools

in contemporary research is to be mindful of the cultural and political

CLASSIFYING GROUND STONE

In his classic analysis of stone implements of northeastern Arizona, Wood-

unusual characteristics (for example, a cupped stone or a notched disk).

of recognizing and sorting patterns in nondichotomous, nonprobabilistic

of the typology presented here is to emphasize the technologically impor-

This introductory chapter has touched on a series of diverse topics rang-

Grinding Technology and

G rinding technology, as a subset of the larger realm of technology, is

express their technological traditions in recognizably distinct designs and

DESIGN AND MANUFACTURE

Understanding technological processes begins with considering how an

in the life-cycle of an object. Decisions made at the design stage generally

gorized by large, small, or a combination of vesicle sizes (vesicles are cavi-

settlements came from quarries considerable distances away (Bostwick

concomitant secondary-use tools compared to another that has a high per-

smooth and require replacement or reroughening. Two adjacent surfaces

Use-wear analysis is the examination of an item for macroscopic and mi-

In this sense the stone surface would have topography on a microscopic

2.5. Schematic of a surface damaged through abrasive wear. Striations and

Arnell 1975:106). These scratches are descriptively referred to as striations,

2.6. Schematic of a surface damaged through tribochemical wear. The sheen on

stantly occurring and are enhanced by frictional energy and mechanical

ate damage patterns very different than those of stone-against-stone con-

Using experimental manos and metates as a baseline for description, it is

matrix and grains worn level frosted appearance

2.10. Photomicrograph of use wear caused by grinding dried feed corn

however, fine powder or meal fills up the interstices or vesicles of the

sheen flattened grains

2.11. Photomicrograph of use-wear caused by grinding sunflower seeds

stone’s surfaces. It does not, however, extend into the microtopographic

Stone-against-Wood or Bone Contact

rounded grains rounded grains