INFORMATION TO USERS

INFORMATION TO USERS
INFORMATION TO USERS
This was produced from a copy of a document sent to us for microfilming. While the
most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the material
submitted.
The following explanation of techniques is provided to help you understand
markings or notations which may appear on this reproduction.
1.The sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting through an image and duplicating
adjacent pages to assure you of complete continuity.
2. When an image on the film is obliterated with a round black mark it is an
indication that the film inspector noticed either blurred copy because of
movement during exposure, or duplicate copy. Unless we meant to delete
copyrighted materials that should not have been filmed, you will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., is part of the material being photo­
graphed the photographer has followed a definite method in "sectioning"
the material. It is customary to begin filming at the upper left hand corner
of a large sheet and to continue from left to right in equal sections with
small overlaps. If necessary, sectioning is continued again—beginning
below the first row and continuing on until complete.
4. For any illustrations that cannot be reproduced satisfactorily by
xerography, photographic prints can be purchased at additional cost and
tipped into your xerographic copy. Requests can be made to our
Dissertations Customer Services Department.
5. Some pages in any document may have indistinct print. In all cases we
have filmed the best available copy.
University
Microfilms
International
300 N. ZEEB ROAD, ANN ARBOR, M l 48106
18 BEDFORD ROW, LONDON WC1R 4EJ, ENGLAND
791W»Z
ROSENTHAL, ELEANORE JANE
SURFACE CO NT EXT , CONTEMPORANEITY AND CULTURAL
T R A D I T I O N : C H I P P E D STONE TOOLS FROM THE
S I E R R A P I N A C A T E , SONORA, M E X I C O .
T H E U N I V E R S I T Y OF A R I Z O N A , P H . D . , 1 9 7 9
University
Microfilms
International
300 n.zeeb road,ann arbor, mi 48106
©
1979
ELEANORE JANE ROSENTHAL
ALL RIGHTS RESERVED
SURFACE CONTEXT, CONTEMPORANEITY AND CULTURAL
TRADITION: CHIPPED STONE TOOLS FROM THE
SIERRA PINACATE, SONORA, MEXICO
by
Eleanore Jane Rosenthal
A Dissertation Submitted to the Faculty of the
DEPARTMENT OF ANTHROPOLOGY
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
19 7 9
Copyright 1979 Eleanore Jane Rosenthal
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
I hereby reconsend that this dissertation prepared under my
Eleanore Jane Rosenthal
direction by
entitled
SURFACE CONTEXT, CONTEMPORANEITY AND CULTURAL TRADITION;
CHIPPED STONE TOOLS FROM THE SIERRA PINACATE, SONORA. MEXICO
be accepted as fulfilling the dissertation requirement for the
degree of
Doctor of Philosophy
^
(/JsC -C CU/
Dissertation Director
/•?
/? 7 f
Date
As members of the Final Examination Committee, we certify
that we have read this dissertation and agree that it may be
presented for final defense..
/Ci.
Final approval and acceptance of this dissertation is contingent
on the candidate's adequate performance and defense thereof at the
final oral examination.
STATEMENT BY AUTHOR
This
requirements
is deposited
rowers under
dissertation has been submitted in partial fulfillment of
for an advanced degree at The University of Arizona and
in the University Library to be made available to bor­
rules of the Library.
Brief quotations from this dissertation are allowable without
special permission, provided that accurate acknowledgment of source
is made. Requests for permission for extended quotation from or
reproduction of this manuscript in whole or in part may be granted by
the copyright holder.
SIGNED:^;
PREFACE
The archaeology of the Sierra Pinacate region is most complex
and this study is just a modest portion of a larger project directed
by Julian D. Hayden. Hayden first employed me as cataloger and later
encouraged my inquiry into chipped stone tool-making methods. I am
deeply grateful for this opportunity to study his material.
I wish to thank my doctoral committee at The University of
Arizona. Drs. Emil W. Haury and Arthur J. Jelinek initially recom­
mended me for training at the Summer Lithic Field School of the
National Science Foundation and then provided helpful criticism in
research and writing. Dr. William J. Robinson, with his personal
knowledge of the Pinacate sites, also contributed greatly to the de­
velopment of the project.
Additionally, Drs. Jeremiah Epstein and Olga F. Linares pro­
vided suggestions which helped to overcome several intellectual
barriers during analysis. I also appreciated Dr. Eleanor Bates' recom­
mendations on statistical matters.
I profited from discussions about stone tools with Don E.
Crabtree and Jeffrey Flenniken, who, along with Jerry Epstein, helped
me replicate percussion techniques on Pinacate basalt. Many of my
ideas about the tools were further clarified during discussions with
iii
iv
Dr. David Wilcox, Michael Waters, Dr. V. K. Pheriba Stacy, Dr. Cathe­
rine Ungar, Kay Simpson, and Roberta Hagaman.
I wish to thank those who assisted in manuscript preparation:
Barbara Sims who edited preliminary drafts; Sharon Urban who drew the
figures; John and Carl Whittaker who illustrated the tools; and Marge
Herrera and Hazel Gillie who typed the manuscript.
I thank my colleagues at the Western Archeological Center
and then California State University, Long Beach, for their encourage­
ment.
Finally, I thank my family for their support; my mother,
brothers, my "honorary" aunt, Marguerite S. Rubinow, and especially
my aunt, Eleanore L. Schafer.
TABLE OF CONTENTS
Page
LIST OF TABLES
vii
LIST OF ILLUSTRATIONS
viii
ABSTRACT
1.
ix
INTRODUCTION
1
Research Objectives
2
2. THE STUDY AREA
7
The Archaeological Survey
Regional Chronology
8
10
3. THEORETICAL BACKGROUND
13
Experimental Approaches
Mathematical Approaches
if.
15
1?
METHODS
21
Procedures
The Variables
......
5. THE CHIPPED STONE TOOL COLLECTION
Question 1: How is Raw Material Selected and Employed
in the Pinacate?
Question 2: What are the Principal Tool-making
Methods?
Question 3: What are the Attributes of the Pinacate
Tool Types?
Question hi What Patterns of Surface Alteration are
Present?
6. THE PINACATE SITE SAMPLES
23
28
37
^6
^8
^9
51
53
Tinaja Maria: HPS 1
Tinaja Badilla: HPS 2 and HPS 3
Chipped Stone Tools: HPS 2
Chipped Stone Tools: HPS 3
v
53
56
57
58
vi
TABLE OF CONTENTS—Continued
Page
Papago Tanks: HPS 4.
Tinaja del Tule: HPS 20
Chivos Tanks: HPS 23
Tinaja Emilia: HPS 31
Tinaja del Ojo: HPS 32
Tinaja del Cuervo: HPS 37
Tinaja de las Figuras: HPS *tO
Tinaja Doble: HPS lf8
Sitio Celaya: HPS 66
Sitio La Playa: HPS 68
......
60
63
66
&9
71
73
76
79
82
85
7. SITE SAMPLE COMPARISONS
89
8. SUMMARY OF STUDY RESULTS
132
The Traditions
Site Variation
Surface Alteration
Problems of Analysis
Tool Typology
133
1^6
1^9
151
15^
9. CONCLUSIONS
156
APPENDIX A:
CODE DESCRIPTION
159
APPENDIX B:
ANALYSIS OF DEBITAGE
163
GLOSSARY
166
SELECTED BIBLIOGRAPHY
169
LIST OF TABLES
Table
Page
1. The Variables
30
2. Crosstabulation of material by site
91
3. Crosstabulation of category by site
9^
Crosstabulation of percussion reduction by site
5. Crosstabulation of retouch by site
97
99
6. Crosstabulation of flake reduction stage by site number . 102
?. Crosstabulation of core or nodule by site
105
8. Crosstabulation of flake platform preparation by site .. 112
9. Crosstabulation of tool types by site .......... 11*+
10. Crosstabulation of use location by site
117
11. Crosstabulation of surface alteration by site
122
12. Crosstabulation of amount of alteration by site
125
13. Paired sites coefficient of agreement for major variables 130
l*t. Site comparisons
136
vii
LIST OF ILLUSTRATIONS
Figure
Page
1. The Archaeological Sites of the Sierra Pinacate,
Sonora, Mexico .......
....in pocket
2. Procedural chart
.........
22
.
38
......
ho
3. Absolute frequency of major materials
km
Flake reduction stage frequencies
5. Nodule and core platform preparation and directionality
4l
6. Major tool type frequencies
^3
...»
7. Use pattern frequencies
8. 95& confidence interval for means of lengths ......
108
9. 955^ confidence interval for means of widths ......
109
10. 9956 confidence interval for means of thickness .....
110
11*
119
955<o confidence interval for means of rake angle ...»
12. 95# confidence interval for means of work single ....
120
13. Tools of Tradition 1 from Sitio Celaya
138
14. Possible Tradition 1 tool kit from Tinaja Doble ....
139
15. Quartz tools of Tradition 2 .............a
l4l
16. Tools of Tradition 3
^3
17. Small tools of Tradition 4 from Chivos Tanks
1^5
viii
ABSTRACT
Archaeological survey in the Sierra Pinacate region of north­
western Sonora, Mexico has produced artifact inventories containing
1386 chipped stone tools from 53 of 72 identified sites. All artifacts
are from surface contexts, lack conventional archaeological associa­
tions, and additionally have few time sensitive attributes conducive
to cross-dating. This paper reports on a computer assisted study of
morphological attributes designed to interpret the cultural and tem­
poral affiliations of the tools.
It is suggested that cultural groups maintain distinct methods
of tool manufacture. Tool-making traditions, therefore, may be recog­
nized by studying the co-occurrences of attributes resulting from tool
fabrication. Identification of demonstrably different methods at a
single archaeological site is indicative of multiple cultural tradi­
tions having been present. Using the concept of tool-making traditions,
surface material from the Pinacate sites may be grouped.
Four tool-making traditions are interpreted as being present
in the collection based upon patterns of material rr. jction, method of
manufacture and final tool shape. Analysis of the tocV3 cnggests that
most Pinacate sites have several traditions represented in their
samples. Correlation of morphological attributes and forms of surface
alteration was undertaken. A relative chronology of the traditions is
ix
X
suggested based upon the prevalent form of surface alteration, given
continuous exposure.
CHAPTER 1
INTRODUCTION
Since 1958 Julian D. Hayden of Tucson, Arizona has located 72
surficial archaeological sites in the 1500 kilometer square Sierra
Pinacate region of northwestern Sonora, Mexico. Within the rugged,
volcanic lava flows, deflation and exposure have destroyed perishable
artifacts. One thousand, three hundred eighty-six chipped stone tools,
artifacts deliberately shaped or having macroscopic evidence of use
(Tringham et al. 197^:171-17*+), collected at 53 sites constitute the
primary source for chronological and cultural reconstruction. These
Sierra Pinacate chipped stone tools are the focus of this study.
The Sierra Pinacate (henceforth Pinacate) tools present special
analytic problems. Found on the surface, they lack conventional
archaeological associations. Further, the tools have few time sensi­
tive attributes conducive to cross-dating. Therefore, a computer
assisted descriptive analysis of each tool's attributes was undertaken
to culturally and temporally affiliate the artifacts.
After observing variation among tools, I hypothesized: (1)
that demonstrable differences in tool fabrication methods indicate
multiple cultural traditions; and (2) that a single component site is
characterized by a consistent tool manufacture pattern.
With the aid of The University of Arizona's CDC 6^00 computer,
these hypotheses were tested by using the Statistical Package for the
1
Social Sciences (SPSS) to analyze functional and morpological tool
attributes.
Ideally this technique should be applied in conjunction with
typological and seriation studies. However, since many Pinacate loci
have neither projectile points nor ceramics, such an approach could
not be employed. Instead a ranked variable of surface alternation was
used to indicate relative antiquity.
In the following pages I first define research problems and
objectives, then briefly review the study area, survey and regional
chronology before detailing theoretical background, methods, data and
study results.
Research Objectives
Identifying the cultural and chronological affiliations of the
Pinacate tool collection presents a research problem because tradi­
tional archaeological principles of stratigraphy (Wheeler 195^:^1) and
association (Hole and Heizer 1965*116) cannot be applied to surface
material. Numerous aboriginal visits may be recorded in one surface
assemblage. In addition, natural forces may redeposit or seriously
disturb associations restricting the use of survey and non-stratified
site data in behavioral explanations. A further research constraint
exists in the Pinacate where typical artifacts ("type fossils") useful
for cross-dating the collection with stratified material seldom occur.
This study attempts to overcome these difficulties by developing a
method for analyzing and interpreting surface remains. The Pinacate
stone tools are a "pilot" sample used to test the applicability of
computerized technological analysis.
3
Research parameters were determined first by the context of the
tools and second by the collecting techniques of the survey. Many
artifacts today are collected according to explicit sampling designs
which determine where and how to collect (Redman 197^sl). Where these
methods are applied, sophisticated, computer-assisted, multivariate
statistics can be employed to help develop tool taxonomies or examine
spatial relationships. The Pinacate material, in contrast, was sub­
jectively collected over a 20-year period providing a non-random
selection of observed tools. Therefore many methodological approaches
are inappropriate to these data. In the absence of stratification,
association and statistical sampling, artifact classes may still be
established and the tool itself may become the focus of analysis.
A tool and its debitage are the products of tool-making be­
havior, providing clues to past human activity (Swanson 1975s1). By
examining each artifact's attributes manufacturing methods may be
discerned. Tool fabrication methods, the stages of manufacture (New­
comer- 1975*97)» disclose an artisan's cultural tradition because toolmaking is learned within a particular society. An artifact may be
perceived as an indicator of cultural tradition and a tradition may
be defined as a distinct attribute pattern limited to a geographical
region during a discernible time period (Sackett 1975:320). By de­
scribing a tool's attributes and inferring its manufacture methods a
tool-making tradition may be identified and its cultural and chrono­
logical affiliation suggested.
What artifact attributes aid in isolating a cultural tradition?
Don E. Crabtree (1975s106) suggests,
k
To the casual observer, flakes and their counterparts may look
alike, apart from obvious differences in dimensions. But
actually flakes and their scars are very distinctive, and can
give clues to the manufacture technique, direction of force,
type of force applied, platform preparation, curvature of
flake, flake termination, stage of manufacture, type of tool
used to induce fracture, and the type of artifact being made.
Selecting attributes indicative of cultural traditions means
evaluating these manufacturing clues. If a consistent pattern of manu­
facture is identified by analyzing chipped stone artifact attributes,
then the presence of a single cultural component may be suggested.
Conversely, if several patterns of manufacture are present, either
diverse traditions or differing time periods may be indicated. These
suppositions may be examined by developing and testing questions about
patterned attribute occurrence and co-occurrence within the Pinacate
tool collection and site samples. By quantifying attributes and
applying elementary descriptive statistical tests, results may be
partially verified.
To define Pinacate tool-making traditions and to date site
material, the following research strategy was established.
1. Identify and describe attributes relevant to tool fabrication
methods.
2. Develop analytical methods to study the collection using
appropriate mathematical techniques.
3. Apply quantitative methods to the samples.
km
Order artifacts into groups by perceived relationships.
5. Interpret results to confirm or nullify the hypotheses.
6. Determine constraints on study results.
5
If cultural traditions can be identified by these procedures,
then the study's second problem, temporal affiliation, can be ap­
proached. To deal with this issue, Hayden's theory (1967, 1976) that
surface alteration reflects the relative age of tools was tested. Tool
exposure has produced characteristic weathering:
accumulations of
staining, desert varnish, oxidation, ferric and lichen adherents and
caliche deposition. If surface alteration is an indicator of time,
then certain methods of manufacttire should be associated with specific
forms of alteration. These phenomena were investigated by quantifying
the co-occurrence of morphological attributes and different surface
alteration forms.
Isolating cultural traditions and relating them to surface
alteration are the research goals of this study. The methods employed
identify the frequency of attribute occurrence; therefore tool attri­
butes are the investigation's raw data (Weinberg and Schumaker 197^^)•
Determining the major relations between morphological attributes and
methods of manufacture is difficult. The chance presence must be dis­
associated from the persistent appearance, when discussing a tradition.
I am concerned with how a tool sample is made, not just how one tool
is fabricated. Technological interpretation, not typology, is the
purpose of analysis. Research procedures are influenced by mathe­
matical studies which emphasize comparisons of sets of attributes
rather than by studies which replicate or type typical tools.
Because it emphasizes first the collection and then individual
site samples this presentation reverses standard archaeological re­
porting which details individual site components and then summarizes
6
the total picture. Many Pinacate sites have too modest a collection
to permit individual statistical analysis. Nevertheless they can
contribute to an overall understanding of methods, materials, and sur­
face alteration patterning. Therefore, the total collection is dis­
cussed first and then attribute distributions at 13 sites that have
sufficient numbers of tools to permit statistical analysis are pre­
sented. The collection and site sample discussions present the results
of questions asked about how frequently an attribute occurs and how
frequently pairs of attributes co-occur in the collection. Then indi­
vidual site distributions are summarized and compared. When possible
data are cited so that the reader may draw alternative inferences.
If the procedures devised to study the Pinacate tools success­
fully answer questions about tradition and temporal affiliation, then
they may be applied wherever tool contexts present similar problems.
The study's primary objective is, therefore, to increase the archae­
ologist's ability to interpret surface artifacts.
CHAPTER 2
THE STUDY AREA
The Sierra Pinacate region is a 1500 square kilometer expanse
of lava flows, cinder cones, 11 collapsed calderas, desert pavements,
and loess-covered plains lying in Sonora's northwestern corner (Gutraan
1972; May 1973)- Hayden (1976) has described the Pinacate environment
as unique among North American deserts because nonperishable artifacts
are found directly associated with topographic features undisturbed by
erosion. The explanation for this singular situation lies in the
volcanic origin of the geology.
The lava flows are extremely resistent to erosion, and have
not been altered since man's entry into the region; they stand,
therefore, in sharp and important contrast to the older
weathered granitic and metamorphic hills and ranges of the
Southwest desert in general. ... Because the Sierra
Pinacate has not been affected by such destructive processes,
it provides an ideal archaeological laboratory (Hayden 1976:
27*0.
Both the geological and hydrological structures are important
for the understanding of the Pinacate tools and their technological
attributes.
Basaltic flows and pavements provide the source of tool mate­
rial for the aboriginal tool maker. Other stones of better quality
(chert, obsidian, quartzite) must be imported from considerable dis­
tances. Further, the development of desert pavements incorporating
artifacts or providing surficial contexts for non-contemporaneous
7
8
artifacts create temporal and cultural interpretational problems
(Hayden 1976).
The regional drainage patterns often control site location as
water is found at temporary or nearly permanent tinajas (tanks) exist­
ing in drainage channels which cut the lava flows creating scour or
plunge pools. These tinajas, before erosion destroys them, attract
both animals and people.
The mixing of artifacts from long term occupation and their
placement on or within pavements means that relative chronologies are
difficult to develop. Weathering from exposure, however, creates an
opportunity of judging a tool's age by its surface alteration. The
necessity for using basalts that are often tough, vesicular, or porphyritic is a further constraint upon tool-making behavior. All these
environmental factors contribute to the uniqueness of the Pinacate
collection.
The Archaeological Survey
In 1952 Paul Ezell began surveying the north-central Papagueria,
testing Gifford's theory that the western Sonoran desert was occupied
by Yuman people. Seventy-two sites within Organ Pipe Cactus National
Monument, along with additional sites in surrounding areas were located
(U.S. Department of Interior 1976). Summarizing his finding, Ezell
(1955) contended that the Cabeza Prieta, Organ Pipe, and Pinacate
regions were Piman homelands and the Yuman frontier lay further west
along the Colorado River.
Encouraged by his initial research, Ezell returned to his
Sonoran study area accompanied by Julian Hayden in 1958 and revisited
Tinaja de los Papagos. Malcolm Rogers commenting on the resulting
collection noted that the Pinacate was the last undisturbed desert
archaeological area containing a complete cultural sequence (Hayden
1976:335)• As construction of Mexican Highway 2 threatened the.-untouched area, Hayden initiated a survey to locate sites prior to their
destruction by increased numbers of visitors. Focusing on locales
which were major aboriginal resource areas, Hayden first visited
regional tina.jas both functioning and extinct and checked their vicin­
ities for artifacts and structures. He often found scattered shells,
sherds and stones on adjacent terraces or pavements. Associated with
these artifacts were sleeping circles or clearings in the desert pave­
ment, cached ground stone tools, bedrock mortars and rare petroglyphs.
Trails led from and linked tinajas with craters, dune camps, and re­
source areas of dense vegetation. Following and mapping some 700 km
of trails, Hayden discovered broken pottery, shrines and large ground
figures. In this manner 72 sites were located by Hayden and others
familiar with his research objectives (Fig. 1, in pocket).
Early in the survey, Hayden realized that in the Pinacate all
archaeological remains lay upon natural surfaces or within topographic
features, such as pavements or dunes. This situation presented
sampling problems because collecting artifacts meant destroying archae
ological associations such as trail breakage or chipping areas. These
conditions constitute "fragile-pattern" areas (Hayden 1965s272) and
necessitate either total collection or complete recording.
Fragile pattern techniques were applied to Pinacate loci when
archaeological remains were minimal but collections were needed to
10
clarify artifact affiliations. More extensive sites were sampled by
gathering typical specimens within topographic areas or cultural
features. If unique associations were encountered, all observed mate­
rial was collected. The artifacts from some Pinacate sites represent
the entire cultural inventory. Other samples represent the surveyor's
selection of common objects from topographically specific situations;
pavements, dunes or terraces.
Regional Chronology
Malcolm Rogers, the pioneer of American desert archaeology,
initially identified Pinacate material and related it to his own
regional chronology. Since Roger's death, Hayden has expanded this
chronology, applying it more specifically to the Pinacate through
cross-dating, associating remains and landforms, and classifying a
tool's surface alteration. Hayden (1967i 1976) chronology has two
periods, San Dieguito and Amargosa, with several phases.
The San Dieguito period was recognized first by Malcolm Rogers
in 1919 and discussed in 1939. Rogers and Farmer (19^8:*0 believed
that this tool complex (they stressed the term "complex" noting the
absence of non-stone items) represented a "thinly scattered" hunting
people dispersed across the western deserts but having regional
aspects.
In the Pinacate, two phases of San Dieguito occupation were
identified by Hayden (1976:280):
the Malpais (basal) and Phase I
(San Dieguito I). These phase distinctions are based upon tool
typology and surface varnishing, oxidation and ground staining.
11
The time of initial colonization is unknown, but Hayden (1976:
286) hypothesizes a pre-19,000 B.P. date for Malpais and a San Dieguito
Phase I date of about 11,000 to 17,000 B.P. Hayden has noted simi­
larities between Pinacate Sain Dieguito I tools and the Ventana complex
material (Haury and Hayden in Haury 1975siv-v).
Post-Pleistocene warming and drying trends altered the Sonoran
desert environment (VanDevender 1977) producing an occupational hiatus
visible at Ventana Cave (Bryan in Haury 1975:71). Reoccupation of the
region apparently occurred after the establishment of modern desert
floral communities. Assemblages found at post-Altithermal sites are
distinctively different from both San Dieguito and Plains (Llano and
Piano complexes) Paleo-Ihdian. These regioneil artifact inventories have
been variously termed Amargosa, Chiricahua-Amargosa, Picosa aind Chiricahua Cochise.
Rogers originally identified several Amairgosan phases including
Pinto, Gypsum, and Lake Mohave. However, these designations have been
isolated from the original chraology and established as distinct
periods in recent years (Hester and Heizer 1973). The terms were taken
from point types typical of the period which are unfortunately not
unique in space or time. To prevent confusion by basing phases solely
on "type fossils," Hayden has retained Rogers' (1959) Amargosa termi­
nology and applied it to the Pinacate.
Amargosa I and II refer to the preceramic occupation, while
Amargosa III encompasses the ceramic to historic time-span. These
phases contain both Western Desert and eastern Arizona point styles.
New groundstone tools like basin and slab metates and gyratory crushers
12
for plant processing also characterize the period (Hayden 196?s
339).
The Amargosa III phase follows the introduction of Yuman
(Patayan) ceramics. The earliest cross-dated artifacts are associated
with Snaketown and Gila Butte phase Hohokam trade wares. By A.D. 700
(Yuman I of Rogers 1939*l8l), the use of imported non-indigenous
Patayan I ceramics is indicated (Waters in press). This use con­
tinued through Patayan IV times.
This is the Pinacate tool context and chronology as inter­
preted by Hayden, who alone has visited most of the sites.
As Rogers
and Farmer (19^8:*0 cogently stated, however, difficulties arise be­
cause although styles change many simple tools are unaltered from their
invention or introduction until their replacement by metal. Thus only
a careful study of the artifacts themselves and their commonly shared
attributes will define the Pinacate's cultural tradition and clarify
some of its chronological problems.
CHAPTER 3
THEORETICAL BACKGROUND
Understanding the differences among Pinacate stone tools is an
arduous task as shape and size reveal only final production steps.
Previous stages of manufacture must be inferred (Clay 1976:303)•
Knowledge of techniques of stone working must underlie any stone tool
study. Lacking a theoretical background, chipped stone tool discus­
sions often are arbitrary interpretations founded upon suppositions of
use rather than objective judgments enhanced by am understanding of
stone working technology (Sheets 1975?372). Experimentation and an
awareness of lithic technology should support interpretation of pre­
historic chipped stone artifact. In the past 20 years, tool replica­
tion research has produced stone working theories explaining attributes
resulting from manufacture techniques (Swanson 1975:1)•
L. Lewis Johnson (1978:3^6-^8) has recently summarized the
history of flint-knapping experimentation detailing developments from
the experimental reports before 1880, through Holmes' charting of manu­
facturing steps in the l890's, to the eolith controversy of the early
20th century. She has commented on the growing interest in the
physics of stone breakage occurring in the 1930's and of aboriginal
stone working in the 19^0's.
As Johnson notes, however, the incorporation of lithic tech­
nology into the mainstream of archaeological research is primarily the
13
14
result of investigations conducted by three individuals, Bordes, Crabtree and Semenov, who have developed the theoretical background under­
lying modern stone tool studies, including this analysis.
In 19^6-^7, French prehistorian, Francois Bordes, produced
several paleolithic tool studies which, although not widely dissemi­
nated to the non-European archaeological community, initiated modern
lithic analysis.
Almost simultaneously, Russian archaeologist S. A.
Semenov recorded his lithic research in the volume Prehistoric Tech­
nology, unfortunately not translated until 196*4-. Bordes and Semenov
spurred their colleagues' interest in tool fabrication and use.
Working independently in the 1950's, Don E. Crabtree redis­
covered many aboriginal flintworking techniques including the dis­
tinction between percussion work and the direct application of pressure.
Bordes and Crabtree met to demonstrate and discuss "flintworking" at
Les Eyzies,France in 1964 (Jelinek 1965s277-79)- The Les Eyzies con­
ference was attended by many archaeologists interested in lithic
research and resulted in numerous projects designed to replicate,
analyze and explain technologic variation among tools and their waste
material (Johnson 1978:351)«
Crabtree has evolved theories on fracture, flintworking tools,
stone quality, stone preparation (like heat treating) and behavioral'
analysis of manufacturing stages. Bordes, in contrast, using morpho­
logical attributes has developed analytic approaches based upon
inspection, classification and quantification. From Bordes' and Crabtree's work a modern emphasis on tool manufacturing processes, not
just the final product,has emerged. Lithic technologists de-emphasize
15
size and shape for chronological seriation and concentrate instead on
industries and their cultural, chronological and theoretical implica­
tions.
Two technological approaches are prevalent today:
experi­
mental replication of manufacturing processes and the application of
mathematical techniques to the study of tool attributes (Johnson 1975:
6k). Both approaches have influenced procedures employed in this
study.
Experimental Approaches
Krieger (19^:272) has argued that distinct patterns or tech­
nology or behavior are acquired during interaction within a cultural
group.
Lithic technologists applying Krieger's concept suggest that
tools exhibit culturally imposed techniques of manufacture.
Employing
this idea, researchers suggest that tool attributes reflect either tool
use (function) or culturally controlled manufacturing dcills (style)
(Sackett 1977:371).
Initially, experimental lithic analysis concentrated on single
tools and techniques.
Crabtree's .1966
replication of the Linden-
meier Folsom in 1966 and Mesoamerican polyhedral core analysis of 1968
exemplify this approach.
Investigations have expanded, however, to
cover studies of complete industries, such as Ranere's (197^» 1975)
replications of tropical forest tool kits.
Crabtree and Butler's(196*0
research on chert heat preparation initiated studies on Purdy and
Brooks' (1971) stone workability followed by heat treating stone in
controlled laboratory conditions.
Speth (1972) and Tsirk (197*0 have
likewise provided technical data on stone fracture physics. All these
studies have contributed to the understanding of manufacture attributes
such as flake scars, bulbs of force, platform preparation, stage of
manufacture and heat treating.
Most replicative studies have emphasized typological goals.
From the morphological attribute approach to classification employed
by Bordes, varied and sophisticated tool analyses have evolved. Em­
phasis on tool attribute analysis permits classifications that in­
corporate more observations than do systems based upon observed form
or presumed functions. It has the advantage of providing information
not only on tool form but also on manufacture tradition. As such
typologies attempt to communicate as much data as possible about a
collection, the usefulness of such classificatory systems is obvious.
Epstein (196^:158) summarized the technological approach to
typology:
Since chipped stone artifacts are, by definition, objects that
have been modified by chipping, it follows that any classi­
fication of artifacts should consider the extent of modifi­
cation as basic in sorting chipped stone.
Lithic technology has used experimentation to better under­
stand process as well as product so that the whole artifact assemblage
and not just the individual tools can be interpreted. Crabtree and
Bordes noted the importance of studying tool manufacturing processes
and drew attention to the cores and debitage. Epstein has commented
further that archaeologists do not study people or function but the
remains of a manufacture tradition and must, therefore, describe the
behavioral tradition.
17
Students of Bordes and Crabtree have investigated manufacturing
processes.
meaning.
Muto (1971) studied initial reduction styles and their
Sheets (1973, 1975) structurally analyzed debitage to infer
the work of specialists and to determine socio-economic ranking.
Collins (1975) presented a general model of tool-making which may
answer broader questions about human adaptation. Gunn (1975) has
identified the individual styles of artisans in his replicative and
computer studies.
Thus, experimental replication and technological
analysis have led to numerous and varied behavioral interpretations.
Schiffer (197*0 has recently suggested that by knowing manufacturing
techniques probability statements about tool-making can be developed.
Gummerman (1976:8-10), however, has criticized this noting that the
application of tool-making techniques is not always predictable, and
that Schiffer's "laws" only cover tool production physics.
Replication studies produced for classification or behavioral
interpretation have resulted in two assumptions which underlie modern
lithic studies, including my own.
First, manufacturing method and
use can be understood by replicating tools and comparing them with
artifactual material (Johnson 1978:351).
Second, the trained lithic
technologist can recognize fabrication methods documented in a tool's
attributes and by so doing can describe a culture or an individual's
technological tradition (Gunn 1975535)•
Mathematical Approaches
Francois Bordes not only pioneered technological description,
but also recognized the problem of quantifying observations.
He
38
combined frequencies into a chart called a "cumulative graph" and then
was able to compare either tool or attribute occurrences between strata
or industries (Doran and Hodson 1975:120).
Approaches such as Bordes'
which tabulate different states of variables (attributes) not only
allow standardized judgmental decisions about tools but also permit
quantification of artifacts too numerous to handle by visual inspec­
tion.
It was, however, A. C. Spaulding (1953:305) who first suggested
that artifact types could be considered as mathematical groups having
a consistent set of attributes when he employed the chi-square signifi­
cance test on paired attributes.
Spaulding's work was expanded and computerized by Clarke (1971:
539) who recorded both continuous variables and discrete variables
(quantitative and qualitative attributes).
The frequency of attribute
occurrence in the matrix was a measure of attribute association rather
than merely a relational statement as with Spaulding*s method.
Mathematical analysis requires numerous tedious operations;
therefore, it was not until computers were used that there was an
increasing interest in this approach.
Computer assisted mathematical approaches have primarily used
techniques of cluster, factor, multivariate and principal component
analysis to analyze continuous variables. Studies by L. R. Binford
and S. R. Binford (1966), Glover (1969), and Hodson (1977) all employed
numerical procedures to divide a given artifact set into homogeneous
subgroups, but these techniques recently have been criticized because
they assume each attribute to be equally important, a concept inconsis­
tent with the needs of most typologists (Christenson and Read 1977:17*0 •
19
The primary focus of mathematical approaches has been classi­
fying artifacts rather than establishing patterns of co-occurring
attributes. Exceptions to this concern are Bordes* (e.g., 1972, 1973)
isolation of cultural groups in the Mousterian assemblages and S. R.
Binford's and L. R. Binford's (1969)
L. R. Binford's (1973) hypoth­
eses on Mousterian tool-kits. Few computer assisted lithic studies
have followed Bordes' or Binfords' ideas, perhaps because of the dif­
ficulty of bridging the interpretational gap between statistical pat­
terns and behavioral inferences.
One exception is L. L. Johnson's
(1975) study employing similarity matrices to differentiate core
preparation systems at Chilean quarry sites.
Though her mathematics
differ from my own, our common goal is to identify and quantify attri­
butes and to develop a concept of the predominant manufacture method
(Johnson 1975?93)•
To accomplish this purpose, Johnson defined 20
attributes and then tabulated how often they occurred.
My own approach
duplicates this step. However, she proceeded to compare each attribute
on individual cores, one to another, while I have compared attribute
counts (see Chapter 4) for groups of tools.
Experimental and mathematical approaches to lithic analysis
have resulted in the recognition of tool-making techniques which pro­
duce artifact attributes.
By examining manufacturing attributes with
a trained eye, behavioral interpretations of stone tool populations can
be presented.
Mathematical approaches permit the handling of large
quantities of data and allow for grouping above the level of inspec­
tion (Thomas 1978:232).
2D
Despite two decades of lithic technology research, typologic
approaches are most prevalent.
Analysis of tool attributes as clues to
cultural traditions, therefore, is difficult because few researchers
have pursued similar interests. If methods employed here diverge from
previous studies it is because objectives are different. Not all
attributes are measurable and Hayden's collecting procedures do not
permit the use of many elaborate techniques of numerical taxonomy be­
cause the parameters of the techniques cannot be met. Nevertheless,
the research summarized has produced the tool-making theories that are
the foundation of my study and have influenced procedures used for in­
vestigating the Pinacate tool traditions.
CHAPTER b
METHODS
To evaluate the research hypothesis discussed in Chapter 2,
analytical methods for testing its validity had to be developed.
A multistage study was therefore implemented to describe,
analyze and interpret the Pinacate tool collection (Fig. 2).
This
chapter discusses the research procedures employed to determine if
patterns of tool attributes could be interpreted to indicate cultural
or temporal affiliations.
A review of various approaches to stone artifact study sug­
gested that computerized mathematical analysis of tool attributes
would be a suitable method for four reasons. First, handsorting and
grouping of nearly l*fOO tools would be impossible in the laboratory
space provided.
Second, computerization would permit the study of more
than one attribute at a time.
Third, use of a statistical approach
would force careful definition of observations and reduce the subjec­
tivity of comparison.
Finally, research methods would be replicable
and could be verified and applied by other interested researchers.
David Clarke (1971:1^2) has suggested that mathematical
approaches to artifactual material be confined to observing specific
attributes with demonstrable relevance to research goals.
Because
mathematical analysis only manipulates data and does not interpret
21
Separation of Tools
from Debitage
determination of attributes
(isolating variables)
record discrete variables
record continuous variables
•J*
«*
apply FREQUENCIES
apply CONDESCRIPTIVE
I
Combine minimally repre­
sented states
I
apply CROSSTABS
repeat procedure at
13 sites
I
^^
apply CROSSTABS
apply ONEWAY
Site X other variables
analysis of variance
'
J
inspection of results
I
coefficient of agreement
I
pairing of sites by attributes
Figure 2. Procedural chart
1
23
observations, the investigators should understand the process that
created any attribute before recording it.
The first research step, therefore, was determining which tool
attributes would aid in testing ideas about tool fabrication and cul­
tural traditions.
Procedures
The empirical data of this study are the observations made on
a tool's morphology and technology.
fied.
In step one, tools were identi­
Initially, tools were separated, from debitage (waste flakes)
and abandoned cores, and catalogued by Carol Weed and Catherine Ungar.
Later, all artifacts were rechecked for evidence of edge damage using
a 30 power microscope and edge microflaking criteria established by
Tringham et al. (197*0.
Most waste flakes were obsidian and were
collected from just a few sites.
Therefore, it appeared that debitage
could not contribute substantially to the answering of research ques­
tions as it had not been routinely collected.
A descriptive sorting
of debitage manufacture stage and raw material was completed, and is
included in Appendix B.
Next, the tools themselves were rechecked for indications of
use.
This review permitted identification of attributes for analysis.
The artifact inspection suggested that some features such as color,
were determined by other attributes such as raw material and separate
recording was unnecessary.
Additional attributes such as the degree
of surface alteration or the distinction of two edge angle measurements
were distinguished as being of relevance to project goals.
A
2k
literature review (Chap. 3) had suggested observations which previously
had produced significant results in other technological studies (such
as platform preparation or direction of flaking).
By combining these
sources, a list of attributes to be analyzed was developed.
The next stage of analysis was determining appropriate mathe­
matical techniques.
This meant deciding what unit to examine and how
to record group observations.
The approach of descriptive statistics
was selected because it entails specifying a population of interest
and then collecting measurements about all members of the population.
The Pinacate tool collection became the subject population with each
tool established as a "case" or the basic analytic unit (Klecka, Nie
and Hull 1975s13). For each case multiple observations or measurements
could be made.
A tool's attributes, because they can take on one or more
states (being either present or absent in one or more forms), can be
considered as variables observed for every tool (Spiegel 1961:204).
These variables were the subject of statistical analysis.
Translating tool attributes into mathematical variables was
the next step and was accomplished by assigning a numerical code:
symbol for the variable's character.
a
A variable such as material had
27 identified states ranging from basalt to ignimbrite, to obsidian,
and each was given a number.
Other variables like length, width,
thickness or edge angle had actual numerical values, so representa­
tive codes were unnecessary.
The rationale behind this approach lies in the use of The
Statistical Package for the Social Sciences (SPSS) which was selected
25
to perform the mathematical analysis because it is "specifically de­
signed to compute those statistics typically used by social scientists"
(Klecka, Nie and Hull 1975:2) (primarily non-parametric).
Its use,
however, requires processing data as described below, otherwise its
underlying assumptions cannot be met.
The SPSS package necessitates determining the nature of obser­
vations being made and then applying appropriate techniques.
Rules
and assumptions of "levels of measurement" (Klecka et al. 1975*13)
must be followed.
Among the Pinacate tools, assigned numbers only
serve as labels for variable states.
This is measurement on a nominal
scale because assumptions are not made about rank or scale, only the
presence or absence of an observation is recorded. SPSS provides sub­
programs for discrete (nominal scale) variables, such as CROSSTABS.
Other Pinacate tool attributes are measurable by interval
scales because zero is definable and intervals from that point can be
measured.
An example of this variable type is a dimensional measure­
ment recorded in millimeters. SPSS has separate Bets of sub-programs
designed for these "continuous" variables, such as FREQUENCIES.
The mechanics of recording tool attributes were simple.
They
involved assigning to each variable a set "field" (a series of coding
blocks on a FORTRAN sheet) in which a numeric code could be written.
When a zero was employed, the variable was not present.
If am attri­
bute was present, a numerical symbol was written corresponding to the
observation.
Observations on each case were recorded on coding sheets
and then transferred to computer cards.
A tool's attributes were
ready for comparison with other tools after the coding was complete.
26
Instructions for the desired SPSS sub-program were then sent to the CDC
6^00 computer along with 1386 cards, each representing a tool (case).
The third stage of research involved the application of one of
four SPSS sub-programs to the Pinacate cases. SPSS contains two sub­
programs for computing statistics which summarize the distribution of
cases for each individual variable. These are CONDESCRIPTIVE and
FREQUENCIES.
CONDESCRIPTIVE computes summarized measures of central tendency
and variation for variables on interval scales. It was suitable for
analyzing Pinacate tool dimensions. Mean, variance, standard devia­
tion, skewness and kurtosis were computed using this sub-program to
determine deviation from normal distributions, and to compute central
scores.
FREQUENCIES, a second sub-program, reports the frequency of
each variable state's occurrence. A table is presented listing counts
of cases for each, the percentage of cases based on the total number
of observations made on the variable under study, and the cumulative
frequency of the states of a variable. FREQUENCIES was designed to
compute the presence or absence of a limited number of states for vari­
ables measured on a nominal scale, as is the situation among most tool
attributes that cannot be ranked or measured.
After examining distributions of variable relationships, two
or more variables could be investigated. The sub-program CROSSTABS
provides a contingency table tabulating joint frequencies of cases for
two or more nominal variables.
Because of the size of CROSSTAB tables,
and the number of variable states only two variable relations were
27
studied; however, future research can look at multiple relations. The
table permits the direct description of attribute relationships. Addi­
tionally, this program calculates a chi-square statistic indicating the
likelihood that the compared variables are not associated one with
another and whether the occurrence results from expectable chance vari­
ation.
CROSSTABS with chi-square tests are particularly congenial to
investigating nominal variables with a limited number of states. Fur­
ther, CROSSTABS provides a means to test hypotheses about relations
among variables. For example, the statement if cores are predominately
basalt, then local stone is being employed, can be tested by looking at
the association of cores and stone types and calculating cross tabula­
tions.
Then the variables' association can be investigated by checking
a chi-square's significance.
CROSSTABS and a second sub-program ONEWAY
provides the basis for my fourth research stage, the stating of corol­
lary hypotheses about individual characteristics which comprise the
tool tradition.
To test if the continuous variables' means have the same or
equal variance at sampled sites, the sub-program ONEWAY was applied.
Hypothesizing that site samples would not produce different effects on
the variability of mean scores, results were then tabulated.
A "F-
test" provided by the sub-program was employed to ascertain if dis­
crepancies in results were sufficient to reject the hypothesis.
ONEWAY
provides a summary table listing sum of squares, mean of squares, "Ftest," level of significance, minimum and maximum scores and the 99#
confidence interval for the mean (Nie et al. 1975s398).
28
The fifth stage of research is the last step prior to inter­
pretations:
relations.
ordering artifacts into groups based upon their perceived
The groupings are at two levels, the population (collec­
tion) and the samples (site).
The first summary of variables was com­
puted for 1386 tools (cases) of the Pinacate collection.
Subsequently-
noting that several categories were poorly represented, some obser­
vations were combined to define new variable states.
Thirteen sites had sufficient cases for grouping at the site
level, and these were then each analyzed by creating sub-files of
cases.
A minimum case number of 30 was selected before grouping was
undertaken. For samples less than 30, the approximation of a normal
distribution (typical representation) is not good, while with increased
numbers of tools a more normal distribution of variable states might
be expected (Spiegel 1961:188).
After individual site sub-files were
organized, the sets of tools with common attributes were grouped.
The Variables
What variables were identified as the subjects of analysis to
support or nullify the research hypothesis?
Tool features are often
referred to as morphological attributes of two types:
functional and
stylistic. These contrast the tool's role as an operator (working at
a task) with its role as an indicator of cultural tradition (Sackett
1973I320).
To investigate tools as parts of technological traditions,
stylistic attributes become the research variables.
Morphological
attributes of style may include dimensions and shape, the parent mate­
rial and its features, and attributes indicative of manufacture
29
methods.
In discussing the computer code, I shall consider each of the
variables in turn, elaborating the reasons for its inclusion in this
study.
Table 1 lists the variables studied, while Appendix A presents
the complete lithic code.
A glossary is included of technological and
typological terms employed during tool analysis.
The initial variable sequences (l-*f) listed in the lithic code
identified the tool and its provenience.
Hayden,
This included surveyors
May, Ezell or Woodin (1), site number (2), catalog number (3)
and a sequential coding number (*)•)•
Such variables allow sub-filing
for special grouping and permit the location of coding or key punching
mistakes.
The first variables recorded were continuous variables of
dimension and edge inclination.
A sliding caliper and contact gonio­
meter were employed to measure dimensions of greatest length (21),
width (22), thickness (23), work angle (19) and use angle inclination
(20) to the nearest millimeter or degree.
Dimensions were recorded to
indicate differential, standardized or preferred size.
All dimensional
readings were taken from either a perpendicular or parallel orientation
to the platform. Partial dimensions (on broken tools) were not re­
corded.
Angle readings were taken at two points, the work angle within
one millimeter of the edge, and the edge angle (inclination at five
millimeters on the tool's face at the least damaged portion of the
edge.
The former defined the angle of use at a tool's abandonment,
the latter the initial angle of the tool's face.
Again, standards of
desired edge and averages of angles at abandonment were the objective
of the measurements.
30
Table 1. The variables.
1. Survey Definition
1^. Core and Worked Nodule Direc­
tionality and Platform
Preparation
2. Site Number
3.
Catalog Number
15. Descriptive Tool Type
b.
Coding Sequence Number
16. Location of Use Damage
5.
Material
17. Edge Profile
6.
Category-
18. Platform Preparation on Flakes
7. Percussion Reduction
19. Edge Working Angle
Flaking
20.
Edge Use Angle (Inclination)
8. Pressure of Controlled
Percussion Retouch
9. Surface Alteration
10.
Amount of Alteration
11. Technological Attributes
12. Possible Affiliation
13. Flake Stage of Manufacture
21. Greatest Length
22. Greatest Width
23. Greatest Thickness
31
The first nominal variable recorded was the parent stone used
for artifact manufacture.
Several attributes existed:
textlire, color,
transparency and origin (rock type). These were grouped into a single
variable, material, which included color and texture variation.
A
crypto-crystalline silicate was termed chert or chalcedony to describe
origin and color, while igneous rocks were described with porphyritic
modifiers like porphyritic rhyolite.
Twenty-seven stones were identi­
fied for the purpose of investigating material preferences, local use,
trade and workability.
A second variable (9) identified was surface alteration result­
ing from a stone's exposure to weather.
Eleven forms of alteration
can be recognized in the Pinacate tool population:
varnish, ground-
staining by clays, caliche, surface erosion, oxidation (weathering as
color change), hydration (weathering on obsidian), green lichen ad­
herent, hematite staining (red), and black oxide.
Co-occurrence of
these was common; therefore, 30 states were recorded.
The degree of alteration (10) also appeared to vary with dif­
ferential exposure.
This was ranked as light if visible; medium if
present in considerable amounts; and heavy if substantially changing
the tool's appearance.
Combinations of degrees were noted and recorded
creating ten states. The kind and degree of surface alteration was
observed to test Hayden's (1967» 1976) hypothesis that the degree of
surface alteration could indicate chronological affiliation, given
uniform exposure.
A majority of the variables was concerned with tool manufacture.
First, the tool's descriptive category (variable 6) was noted as flake,
core, worked nodule, or utilized stone.
A special attempt was made in
analysis to separate cores and nodules.
Cores were stone from which
usable flakes had been removed and later the artifact was used as a
tool.
Worked nodules were basalt "ejecta" or stone cobbles reduced in
mass, shaped and sharpened, and unlike cores their primary purpose was
not to produce usable flakes (see Glossary).
Utilized stones were
stones battered by use but revealing no evidence of "deliberate"
shaping.
The mechanics of tool manufacture were studied by developing a
series of variables including "reduction," "retouch," "flaking direc­
tion on cores," platform preparation, and technological attributes of
damage or breakage.
fabricated.
Each variable was a clue to how the tool was
When these were combined with another variable, stage of
manufacture, I hoped to identify the patterns of tool-making.
If percussion techniques were employed to make a tool, obser­
vations were made about percussion (variable 7) reduction, the re­
moval of flakes from single edges to thin and shape the tool.
Along
with unifacial, bifacial and multi-facial reduction, two other codes
were added to accommodate unique patterns of percussion.
Alternate
percussion reduction refers to flake removal alternating along a single
edge bifacially but with alternate blows, so that flake scars are not in
opposition.
The result is a sinuous serrated edge. Reverse percus­
sion reduction describes the practice of removing flakes from opposing
edges on opposite faces; a trapezoidal-shaped tool results (see Glos­
sary).
33
A second variable used to describe manufacture method is re­
touch (variable 8).
Retouching defines the deliberate serial removal
of small, parallel percussion, or more often pressure flakes.
Notching
was also considered a type of retouch.
The distinction between percussion reduction and percussion or
pressure retouch was made to provide data on tool thinning and shaping
(reduction) and final tool modification or resharpening (retouching).
Thus, retouching as employed in this analysis has a more specific
application than in most studies.
Further, in emphasizing deliberate
removal of small flakes, an attempt was made to separate use damage
(microflaking) from deliberate manufacture, by considering sporadic
microflaking as indicative of the latter.
Tools used as core tools were analyzed to determine the pattern
of flake removal.
Direction of flaking and platform preparation (vari­
able 1*0 were combined in 10 categories, three of direction and seven
of platform preparation and direction.
preparation was also recorded.
Similarly, flake platform
The absence of a platform was a zero,
while cortical meant the platform was not prepared. "Preparation" in­
cluded the use of planar breaks in the material and use of a single,
double (dihedral) or multiple scars as platforms. Six values were
noted, though no distinction was made between deliberate or accidental
preparation. Grinding, or abrading of platforms was initially coded
but was never observed.
Therefore, it was eliminated from the code.
Technological traits (variable 11) relating to manufacture or
use were recorded because they indicate material workability or imper­
fection.
In retrospect, manufacture breaks could have been separately
3^
listed, but since it is difficult to determine how a break occurred,
fractures were grouped so an arbitrary decision was unnecessary. Plat­
form crushing, step and hinge fractures (Crabtree 1972:68, 93), "per­
verse" fractures (angular breaks due to material imperfections), and
combinations of these traits were noted.
A final variable was designed to suggest a tool's stage of
manufacture (variable 13)•
This variable provides data on tool size
and shape preferences as well as core reduction steps. Flake types
included cortical and cortex removal flakes.
The former are the first
flakes removed and are predominantly cortical, retaining more than 70&
to 1C$ retained.
A primary flake, the third type, is an initial non-
cortical flake generally removed from a large block and like the other
two flake types, is over 3 cms in length.
Specialized primary flakes
refers to diverging flakes wider than they are long representing the
thinning of cores, cobbles, or bifaces (side-struck or bifacial thin­
ning flakes).
Secondary flakes were of two types, cortical and non-cortical.
Both are less than 3 cms in length and were established to identify
trimming flakes removed from cores or other flakes during tool finish­
ing or resharpening.
Other coded states included blades:
flakes,
twice as long as they are wide, having parallel edges and at least two
longitudinal exterior flake scars. Flake
cores are large flakes with
negative scarring suggestive of removal of additional flakes for tool
production.
A final manufacture stage was shatter. During manufacture
portions of cores or flakes break, but they can still be used. Hence,
broken chunks with either striking platforms or bulbs of force were
35
termed shatter.
Shatter was used only when the reduction stage could
not be inferred from the object, yet the artifact was obviously the
product of manufacture.
The tool type was also described (variable 15). This wa6 done
in an attempt to provide comparative data for other researchers and
additionally to record multiple reductions, retouches or use.
Thirty-
one values were identified and listed descriptively by their manufac­
ture or modification rather than by their presumed function.
Thus, a
unifacially retouched flake tool iB the traditional "scraper," and the
unifacially reduced ejectum is a "chopper."
Descriptive exceptions
were made for categories like projectile points or reworked groundstone.
A single functional variable (number 16) was created; the
location of edge damage.
This was recorded because it was possible
that tool use might vary in relation to other variables.
A 10-powered
hand lens and a 30-powered binocular microscope were both employed to
check tools for micro-flaking, edge rounding and/or polishing (Tringham
et al. 197*0.
The location of damage was recorded by orienting the
tool's striking platform or length axis and noting if interior (bulbar),
exterior, lateral or transverse platform damage or combinations of
these were present.
Another nominal variable of form was the working "edge profile."
This simply suggested the shape of the tool's utilized portion as con­
vex, concave, straight or a combination of these. Form is a variable
created during both tool manufacture and use, and thus may be purely
arbitrary. However, it was recorded to identify any edge form occur­
rence pattern.
36
A final variable (number 12) was included to provide informa­
tion for chronological comparisons.
If the tool's provenience sug­
gested association with objects or structures of identified cultural
affiliations, then these data were noted.
When an Amargosan sleeping
clearing was recognized, then Amargosa was recorded as a possible
indicator of tool affiliation.
Twenty-three variable observations were made about each tool.
Frequency of attribute occurrence and co-occurrence provided the means
for examining questions about the traditions of tool-making.
Obser­
vation of discrete variables was emphasized instead of tool measure­
ments, because the study's purpose was to analyze methods of manufac­
ture.
This approach permitted me to explore the potential of
systematically recording attributes directly without the intervening
step of translating all attributes into a measurable entity.
CHAPTER 5
THE CHIPPED STONE TOOL COLLECTION
The Pinacate tool collection represents selected chipped stone
artifacts from 5^ sites.
Distributions of variables within this tool
population were analyzed to discover what attributes characterize the
collection.
Each variable and variable state was individually tabu­
lated to find the commonest attributes.
Then some variable states
were combined (those with minimal numbers) and the CROSSTAB sub-program
was applied.
The results summarize the tool attributes which are fre­
quently associated.
This chapter presents data on the prevalent attri­
butes of the total tool collection.
The chipped stone tool collection consists of 1,386 artifacts:
9^5 flakes, 173 cores, 265 worked nodules or ejecta, 2 natural stone
slabs, and one unclassifiable tool.
The tools are fabricated primarily
from olivine basalt (42.790, obsidian (15.290, and rhyolite (11.390,
but quartz (10.6?0, chert (8^), andesite (390 and igneous porphyries
(1.790 are often worked.
Minor amounts of quartzite, granite and
scoriacious basalt, labradorite, jasper, siltstone, arkose sandstone,
limestone, ignimbrite and mudstone were also collected (Fig. 3)«
Basalt, obsidian, rhyolite and chert, the principal tool mate­
rials are mostly thinned by hard-hammer percussion prior to their use
(72.8%), and many aire further shaped or resharpened by either
37
38
FINE
BASALT
58B
41
ANDESITE
210
OBSIDIAN
III
CHERT
RHY0LITE
is3
QUARTZ
147
CHALCEDONY
28
QUARTZITE
18
PORPHYRITIC
RHY0LITE
17
SCORIACEOUS
BASALT
15
GRANITE
IS
Figure 3*
Absolute frequency of major materials
39
percussion or pressure retouch (60.5^). Both percussion reduction and
retouch tend to be unifacial (53»8?6 and 65£j, respectively), or bifacial
(53«7& and 28^), but there are minor amounts of alternate reverse and
multifacial flaking.
Data on flake manufacture stage indicates preference of pri­
mary flakes for tool fabrication (26.b%).
However, one quarter of all
flakes are cortex or cortex removal flakes.
Few "specialized primary"
(sidestruck bifacial thinning) flakes are present but small secondary
flakes are relatively common. Blades and shatter are insignificantly
present in the tool inventory (Fig. *f).
A majority of the flake tools
having striking platforms (35«3$) display single flake scars (kj>.l°/o)
but cortical platforms (30»3?0 and natural breaks in the stone (12.39^)
are also present.
Grinding and abrading preparation techniques were
not observed on striking platforms.
Cores and some nodular pieces and ejecta have dual functions:
flake production and use as tools. From these artifacts techniques of
flake production can be partially inferred.
Better than two thirds
have "prepared" platforms in the simplest sense with the cortex having
been removed.
Most "preparation" is merely the expedient use of a
single flake scar but occasionally full shaping of the platform occurs.
Flakes are removed in one (57*7%) or two (27«6&) directions (Fig. 5)*
For comparative purposes 3^ descriptive tool categories were
identified in the collection, but few are found in amounts exceeding 5
Unifacially (serially) retouched flakes (32?6) are most common
followed by large unifacially (9%) and bifacially (119a) percussion
reduced nodules, ejecta or natural flakes.
Utilized unmodified (not
ko
146
CORTICAL
CORTEX
REMOVAL
188
366
PRIMARY
SPECIAL
PRIMARY
73
139
SECONDARY
29
SHATTER
FLAKE
CORE
i 10
Figure k. Flake reduction stage frequencies.
kl
UNIDIRECTIONAL
(UNPREPARED)
BIDIRECTIONAL
(UNPREPARED)
MULTIDIRECTIONAL
(UNPREPARED)
UNIDIRECTIONAL
PREPARED
BIDIRECTIONAL
PREPARED
MULTIDIRECTIONAL
PREPARED
Figure 5. Nodule and core platform preparation and directionality.
retouched) flakes (IO.89S) and utilized cores (8&) are also found in
considerable numbers.
Less frequent but not uncommon are bifaces
(8.4$) and projectile points (6.590. Minor tool constituents include
notched, notched-retouched, retouched-utilized and denticulated tools
(Fig. 6).
Evidence of tool use can be observed on outer faces of flakes
as well as on cores and nodular tools as crushed and step-fractured
edges or slight wear-polishing.
Edge damage from use most often can
be identified on outer lateral-transverse (13.2&) edges.
of platform use also is present (8.3^).
Minor amounts
Tools showing damage on both
faces and all edges constitute 15•?£ of the collection (Fig. 7).
Despite abundant tool modification by retouching, tool shapes
are not distinctly patterned.
Edges are primarily either convex
(29.29d) or undulating (concave-convex, 30.8%), while relatively fewer
concave edges (14.1%) documenting notching and denticulation are
found.
Maximum length, width and thickness of the tools vary widely.
Mean length is 5»8 cms, width 4.4 cms and thickness 2.1 cms.
measurements do not appear to cluster.
Angle
Work angles range between 61
and 89 degrees with a mean of 75.5, while the rake angles range is
smaller, 49 to 78 degrees with a mean of 63-5 degrees. These angles
are on the average wider than the series described by Wilmsen (1968),
but as his methods for measuring edge angles differ from those employed
in this study, the results are not comparable.
Seventy-five percent of the tools' surfaces are altered by
oxidation (30.45&), defined as a definitive change in the surface
^3
UNIFACIAL
REDUCED
124
BIFACIAL
REDUCED
152
UNIFACIAL
RETOUCH
443
UTILIZED
FLAKE
149
UTILIZED
CORES
III
90
POINT
117
BIFACE
DENTICULATE
35
NOTCHED
29
UNIFACIAL
RETOUCH-NOTCH
36
Figure 6. Major tool type frequencies.
PLANAR
CORE
ON
INNER
55
19
247
OUTER
INNER
TRANSVERSE
19
OUTER LATERALTRANSVERSE
181
INNER
LATERAL
OUTER
LATERAL
OUTER
TRANSVERSE
194
57
BIFACIAL
95
LATERAL
ALL FACES
ALL E06ES
INNER
ALL EDGES
216
33
BIFACIAL EDGES-
100
TRANSVERSE
PLATFORM
59
Figure 7. Use pattern frequencies.
^5
color; for example basalt changes from dark to light gray.
Oxidation
combined with groundstaining (12.8?o), hydration of obsidian (11.190,
varnishing-groundstaining and oxidation combined (10.5?j), oxidation
and caliche (8.T&) or green lichen adherent (8.79o) are also common
forms of surface alteration. Surface alteration may be ranked as
mostly light (57*3^) or moderate (22.9&).
There is a small quantity
of heavily altered tools (6$).
The relative frequency of tool attributes appears to indicate
that the Pinacate collection is a basalt flake industry, flakes and
basalt being commonest.
Closer inspection reveals a more complex
manufacture pattern which is masked by frequencies of single attri­
butes.
CROSSTABS contrasting category with material variables demon­
strates that actually basalt tools are equally on flakes and cores or
nodule-ejecta, and other materials are contributing relatively more
flakes to the tool population.
To avoid other facile conclusions
about the Pinacate collection, a series of questions about the co­
occurrence of tool attributes were formulated.
as:
They were posed simply
is attribute one (like the variable material) occurring inde­
pendently, or is its occurrence related to the presence of attribute
two (such as stage of manufacture)?
CROSSTABS was employed to generate
the contingency table used to answer the question while the chi-square
test was applied to judge the significance of the relationship. The
attributes which appeared to be associated were then used to determine
material selection, tool-making methods, tool types and surface alter­
nation, all of which combine to delineate the tool manufacturing
traditions.
k6
Question 1: How is Raw Material Selected
and Employed in the Pinacate?
Most Pinacate tools are made from local basalt flakes, cores,
nodule-ejecta or naturally occurring flakes, with basalt cortical and
cortex removal flakes (the initial steps in stone reduction) occurring
in greater quantities than similar flakes of other materials.
Few
other materials are worked on sites as indicated not only by the flake
data but also by the paucity of chert and obsidian cores.
Because
large quantities of basalt are locally available and flakes or small
nodules (like Apache tears) of other stones appear to be imported, dif­
ferences are evident.
Most non-basaltic tools are fabricated on large
non-cortical flakes or, in the case of obsidian, on smaller secondary
flakes.
Initial percussion reduction is primarily documented on
basalts and to a more limited extent on rhyolite and quartz.
Final
tool finishing by pressure retouch and the use of smaller flakes may
be seen in cherts, chalcedonies and obsidians.
Because fine-textured and vitreous materials are not locally
available, greater care in the manufacture and use of easily worked
stone might be expected.
Such attention is observable in the higher
relative frequency of simple platform preparation characterizing
cherts, rhyolites and andesites and the predominance of cortical plat­
forms on basalts.
Further, though flake removal is predominantly uni­
directional on basalt cores (6C$5), it is more often bidirectional on
finer-textured stones.
Raw material also appears to determine per­
cussion flaking methods.
as do rhyolite and chert.
Basalt tools tend to be unifacially percussed
Chert and rhyolite tools need little
k7
post-detachment percussion reduction prior to final shaping as their
initial removal is partially controlled by core shaping. Therefore,
there is a lower frequency of percussion flaking of tools made from
these stones. The predominant unifacial percussion on basalt tools
typifies the difficulty of removing even the cortex from many nodules,
ejecta and flakes. Bifacial percussion reduction of obsidian, on the
other hand, can be more easily accomplished because the material frac­
tures easily when moderate force is applied; however, small nodule
size and flake thinness makes percussion thinning uncommon on obsidian.
Specialized styles of reduction are also material specific.
Alternate
patterns typify basalt, while reverse patterns are found mostly on
quartz.
Retouch modification to further shape tools is characteristic
of finer-textured stones: rhyolites and cherts as well as some basalt.
Retouching basalt is difficult and in particular pressure retouch is
not often performed, thus only
of basalt tools are retouched.
Within the population obsidian tool retouching (during biface or point
fabrication) is
of all tool retouching. Cherts and rhyolites are
also often unifacially retouched.
Materials appear to be selected with certain tool forms in
mind. Many denticulated and notched tools are made on cherts, rhyo­
lites, and chalcedonies which have, as mentioned above, higher retouch
frequencies. Similarly, many basalt tools are simply unifacially or
bifacially percussion flaked tools, or unmodified flakes, therefore the
lower relative frequency of retouch is understandable.
kS
Stone workability also is reflected in manufacture mistakes and
breakage.
Non-conchoidally fracturing stone tends to be structurally
weaker and subject to unpredictable fracturing (Crabtree 1972:5)•
Step
and Jiinge fractures are common on basalt tools and less frequently
appear on cherts, chalcedonies and rhyolite.
The two major patterns of stone selection are first the use of
small, fine textured (chert, obsidian) flakes; second, the use of
cortex-retaining basalt flakes or large nodules and ejecta; and third,
the use of primary flakes of basalt, rhyolite, chert and quartz.
Question 2: What are the Principal
Tool-making Methods?
Although the Pinacate collection is primarily a flake tool
industry this does not mean that tools represent expedient use of
flakes.
Most tools are percussion reduced and a high incidence of
flake thinning and shaping accounts for much of this reduction.
The
use of percussion to shape tools is understandable given the quantity
of large cortical and non-cortical flakes selected by the tool-maker.
These specimens need thinning prior to their final sharpening.
Percussion reduction shapes most flakes unifacially (3890 or
bifacially (20#). Percussion flaked nodule-ejectum tools are 20# of
all percussed tools while flake-producing cores contribute the remain­
ing 22#.
Five hundred eleven tools are not only percussion reduced but
may have been further shaped by retouching.
Reduction and retouching
occur together in unifacial (51#) and bifacial (20.7#) patterns.
Occasionally after unifacial thinning, edges will be bifacially
49
retouched (12.9&).
Retouching is primarily unifacial on nodules and
flakes, however, when bifacial retouch occurs it is always on flakes.
Retouching techniques vary according to a flake's stage of
reduction.
Unifacial retouching predominates on cortical, cortex re­
moval and primary flakes.
In contrast, secondary flakes are mostly
bifacially retouched (66.9&).
Pinacate retouching appears to be a method of shaping and
regularizing edge profiles.
Bifacial retouching is most observed on
convex edges (29.4?o) and concave-convex forms (24.7?o).
Alternate re­
touch lends itself to the serration and the production of concaveconvex edges.
Question 3? What are the Attributes
of the Pinacate Tool Types?
Epstein (1964) suggested that when standardized types are not
present in a tool collection, final modification may be used to create
tool categories. The Pinacate tool types were defined by modification
(Chapter 4) however, additional variables can help refine the tool
classification.
In particular, location of edge damage aids in dif­
ferentiating flakes, blades or core tool types, all of which may be
used without being modified by reduction or retouching. The most
common Pinacate tool form ("type") is the unifacially retouched flake,
which has two prominent use patterns:
transverse (22.6^).
outer lateral
and outer
Although many tools fall into a traditional formal
category of "side-scraper" (a laterally, serially retouched flake) data
indicate the variety contained in this class.
A similar pattern of
use damage occurs on unmodified flakes (29.7S& outer lateral and 23.6^
50
outer lateral-transverse).
It appears that despite technological dif­
ferences the two tool classes may have similar uses, and retouching
may be a tool resharpening technique.
Small notched, denticulated or unifacially retouched-notched
tools display outer or outer lateral-transverse wear.
Many of these
smaller tools have limited edge damage.
Multi-edge damage can be observed on both projectile points
and bifaces.
Initial platform preparation may, during analysis, have
been inadvertently identified as use.
However, the tools also may have
been used initially for something other than projectile points.
Use evidence can be seen on the outer faces of unifacially and
bifacially reduced nodule-ejectum (chopping and chopping tools-) tools
and on cores.
This damage which includes rounding and crushing of
edges is apparent even on cortical surfaces and supports the concept
that these are principally tools and not reused cores.
A second variable aiding in delineating tool classes is edge
shape.
When tool and edge associations are analyzed they suggest that
the least definitive aspect of tool style among the Pinacate is the
final edge shape.
The only exceptions are the standardized outlines
of points and bifaces.
There is a noticeable scarcity of converging-
retouched flakes (gravers, drills or perforators), and straight-edged
blades.
Flakes, before modification or use, often have convex edge
profiles and this shape is retained by most of the collection's re­
duced, retouched, or utilized tools.
51
Question h: What Patterns of Surface
Alteration are Present?
One thousand forty tools display surface alteration, a high
frequency of occurrence, but not unexpected given their possible long
exposure to weathering.
In this surface context, 57.3^ are lightly
weathered, 2^.9% oxidized, and 10.6?o are "hydrated" obsidian.
Most
moderately altered tools have been affected by groundstaining and
oxidation.
Heavily altered tools combine either heavy varnish with
groundstaining and oxidation, or moderate varnish with heavy ground­
staining and oxidation.
When such alterations are found they can provide a relative
chronology for the occurrence of differing tool-making methods.
Six
hundred sixty-four of 9^5 flakes are surface altered, either hydrated
(17.j$), caliche encrustation (13.^) and varnishing (10.3&) all appear.
On the inner faces of cortex and cortex removal flakes, oxidation alone
can be seen in moderate frequencies (l5«7/& and 26.7% respectively), but
combined groundstaining and oxidation is more prevalent (31.8^ and
27.3&) and both with varnish added are 31^*$ and 10.*$.
Therefore
many flakes from the initial reduction of cores are moderately to
heavily oxidized, groundstained and/or varnished.
In contrast, light
oxidation or hydration is characteristic of large and small noncortical flakes.
This brief discussion only highlights major tool attribute co­
occurrences in the Pinacate collection.
Several preliminary con­
clusions about distributions of material, manufactiire style, tool
shaping and stir face alteration can be drawn before the individual site
differences are detailed.
A variety of flake-tools — most often a simple unifacially
retouched flake displaying outer lateral or lateral-transverse use —
are predominant in the Pinacate collection.
This "scraper" appears on
numerous types of stone, particularly basalt, rhyolite, chert or
quartz, and is made by selecting generally a large, primary flake,uni­
facially percussing it (on occasion bifacially) and then unifacially
retouching its edge.
A second common tool group consists of uni­
facially or bifacially percussion reduced basalt nodules or ejecta
that are unretouched.
High relative frequencies of moderate and heavy
surface alternation — oxidation, groundstaining and in some instances
varnishing — are associated with this second tool group.
Data indi­
cate that the only standardized tool foras frequently occurring are
small projectile points and bifaces.
These are made by bifacially
pressure retouching small obsidian secondary flakes. Specialized
tools with denticulation and notching are uncommon.
Along with the
aforementioned types only utilized, unmodified flakes and reused cores
occur in notable quantities.
To better understand where and why this tool attribute vari­
ation exists, the 13 sites which contributed the major samples to the
Pinacate tool population will be discussed next.
CHAPTER 6
THE PINACATE SITE SAMPLES
Thirteen Pinacate sites provided adequate samples for quantita­
tive analysis of variable occurrence.
This section details distribu­
tions at each site based upon variable tabulations performed by the
SPSS programs.
Attribute co-occurrence information is derived from
CROSSTABS tables tabulating absolute and relative frequencies.
Vari­
able pairs were recorded to see if they were associated or occurred
independently.
As site descriptions have yet to be completed, Hayden
provided brief oral summaries of site situation, environment and arti­
fact locales.
His comments have been incorporated with my own brief
introductions to each site's tool collection to provide background for
detailing distinctions among the samples.
Tina.ja Maria:
HPS 1
Tinaja Maria is the northern most site encountered during
Hayden's survey of the Pinacate's western lava flows.
At Maria lava
flows front a large swale draining away from a series of rock tanks
which were probably only temporary water holes because they are shallow,
barely half a meter deep.
The lava flow vegetation is saguaro, low palo verde and ocotillo cover and in the deflating swale creosote grows while the tribu­
tary emptying the tanks supports palo verde and mesquite.
53
5^
Maria's artifacts were located vdthin the swale and to the west
where a varnished desert pavement has formed.
Bedrock mortars surround
the tinaja and broken metates and manos in the swale are evidence of
local plant processing activities.
Most chipped stone tools were
collected in the swale and were associated with Patayan ceramics of
all time periods.
Several tools, however, were gathered from the
varnished pavement area.
Tinaja Maria's chipped stone tool sample consists of just 39
pieces:
28 flakes, 7 cores and
worked nodules.
Basalt, crypto-
crystalline silicates (chert and jasper), and quartz each represent
25& of the raw material at the site while rhyolite and andesite con­
tribute the remaining percentage. No obsidian tools were collected.
Thirty-seven tools are percussion reduced (82.1$), primarily
on one face.
Some of these tools (61.550 are also modified and shaped
or resharpened by retouching, mostly unifacially.
Flake tools are predominately made on primary (3^-5^) and
bifacial thinning (20.?^) flakes. Some cortical (6.9&) and cortex
removal (17-290 flakes are employed as tools, but few secondary (6.99a)
or secondary trimming flakes are found in the sample (3.^)•
Twenty-
two flake tools retain platforms and nearly half of these have single
flake scars (40«99o) but several multiply scarred platforms (9*1$) are
present.
forms.
Seven of the 11 core or nodular tools have prepared plat­
These tools are equally divided among unidirectional (3) and
bidirectional ('f).
Despite the small sample size the lack of variation is note­
worthy when compared to the whole Pinacate collection.
Unifacially
55
retouched flakes constitute kB.7% of all specimens and utilized flakes
contribute an additional 30.8%.
These are the only tool types on
flakes except for a single biface. Five utilized cores were collected
(12.8%) and two bifacially reduced nodular pieces were employed as
tools.
Tool wear and damage is primarily observable on outer lateral
(21.650, outer transverse (10.396) and both outer lateral and trans­
verse edges (15.^9o).
Core platforms also exhibit damage.
Tool sizes are relatively small averaging k.J> cms in length,
3.^ cms in width, and 1.6 cms in thickness. The mean rake single is
60.7° and the mean working angle 67°•
Tinaja Maria has only 18 surface altered tools in its collec­
tion.
Three are varnished, 17 are oxidized and three are both var­
nished and oxidized.
On 11 tools the surface alteration is moderate
to heavy.
Some flakes appear to be specifically selected for particular
tool types.
Cortex and cortex removal flakes are primarily unifacially
retouched as are bifacial thinning flakes.
In contrast most unmodi­
fied, used flakes are flakes, and the lone biface is made from a very
small secondary flake.
Flakes show approximately a 2:1 ratio of moderate to light
alteration.
Cores and worked nodules are predominately moderate to
heavily oxidized.
This variation pattern suggests that the sample
represents two periods: (1) a lightly oxidized to unaltered late phase
probably contemporary with the site's ceramics; and (2) an earlier
moderately to heavily oxidized affiliation.
The varnished tools from
56
the pavement suggest that an earlier group of basalt tools may not have
been extensively sampled by Hayden.
Maria1s interest lies in its primary and bifacial thinning
flake tools whose moderate alteration suggests a placement in the preceramic period.
Certainly they are distinct from the varnished and
heavily oxidized specimens.
Tina.ja Badilla:
HPS 2 and HPS 3
Tinaja Badilla is situated close to Tinaja Maria on the north­
western curve of the Pinacate lava flows.
Badilla Crater lies directly
southeast as do rock tanks which are so eroded that they hold little
water and so may be considered extinct.
and HPS 3»
Badilla has two sites, HPS 2
The former is found at the upper end of a broad bajio
densely thicketed by palo verde and ironwood along its water course.
The bajio site has at least two components.
Patayan ceramics
were abundant and additionally a quartz chipped and groundstone tool
inventory was recorded which Hayden believes is of Amargosa I or early
Amargosa II affiliation (Hayden personal communication, 1977)*
Manos,
metates and gyratory crushers suggest food collecting and processing
occurred in the vicinity.
Sheep and deer could also have been hunted
while entering and feeding in the crater.
Badilla*s pavement site was briefly described by Hayden in 1967
as its figure and large sleeping circle with cached metates are note­
worthy.
HPS 3 designates this pavement and deflating loess-cinder
locale southwest of the tanks.
The two chipped stone samples were separately collected and the
original site designations were maintained as sufficient material was
57
gathered from both the bajio and the pavement locales to permit indi­
vidual study.
Chipped Stone Tools:
HPS 2
HPS 2's collection is composed of 8*f tools:
60 flakes, 12
cores, 11 worked nodules or ejecta, and a single utilized piece. The
most commonly employed material is quartz (5696).
Rhyolite (25^) artifacts also occur.
Basalt (2590 and
Cherts (4.8?a), chalcedonies
and obsidian (2.^) are uncommon.
Most tools are substantially
percussion reduced (78.690 and a smaller percentage are additionally
retouched (71.*$).
Slightly less than two-thirds of this modification
is unifacial.
The 60 flake tools represent all stages of reduction.
Over
half are large non-cortical flakes (51«75°) while cortical and cortex
removal flakes constitute only 16,7% of the total, and few secondary
flakes were used (13.1/6).
The few core tools collected indicate standardized tool prepa­
ration techniques.
In contrast to the flakes, all cores have prepared
platforms and many of these showed multidirectional flake removal
(*f6.2£).
Most tools are simple unifacially retouched flakes (58.3#) or
utilized unmodified flakes, but moderate numbers of denticulates and
bifacially retouched tools were also recovered (3-6$), and a single
point and a biface were also collected.
Tool use damage can be ob­
served predominantly on exterior surfaces (over 50^) but planarplatform use is also distinctly present (13.1$)•
58
Seventy percent of the tools have some form of surface altera­
tion. Particularly notable is the presence of a green lichen (38.1/0,
which has developed on the underside of these quartzes.
Most tools range between 2 and 6 cms in length and 1.5 to ^.5
cms in width.
Thickness averages 1.6 cms.
and does not diverge widely.
The rake angle averages 69°
The work angle ranges between 57° and 85°
and has a mean of 72.
Badilla1s bajio site appears to have two tool-making patterns
with quartz flakes and small tools constituting one, and cores and
nodular tools reflecting a second set of tool-making ideas.
Addi­
tionally, some intrusive tools, perhaps from Badilla1s pavement basal­
tic tool-kit appear in the site sample.
Chipped Stone Tools:
HPS 3
The HPS 3 sample of 100 tools is from a compact locus and re­
flects local tool production and multiple subsistence activities.
Fifty-nine flakes, 27 cores, 13 nodular-ejecta, and one utilized piece
were recovered.
This sample produced the highest utilized core fre­
quency of any Pinacate site and although error may be involved, never­
theless, core tools are more numerous here than elsewhere.
A high percentage of basalt occurrence at Badilla's pavement
site supports a concept of localized expedient core and flake produc­
tion (51^)•
Quartz (l8.C$) and obsidian (12&) are the only other
frequently found materials.
Finer stone like chert, chalcedony and
rhyolite compose barely 16# of the collection.
59
Because cores are common, percussion reduction is prevalent
(78^).
This percussion work is primarily unifacial (53-8%) on flakes
and unidirectional on cores.
Retouching is less common, occurring on
about half of the tools and being mostly unifacial (79»29j)«
Many tools
are made on large non-cortical flakes (**7»5&), however cortical and
cortex removal flakes constitute 3C$> of all flake tools.
A major dif­
ference between core and flake tools is that core platforms are often
prepared (809S).
Most tools at Badilla (HPS 3) pavement loci are either utilized
flakes or cores, however, retouched flakes and bifacially percussion
reduced ejecta are frequently found.
Notched
and notched and
retouched tools (3?°)» points (3&) and bifaces (6$) are also present.
Tool edge damage patterns reflect the prevalence of utilized cores as
use and retouch damage on platform areas was identified on 19$ of all
tools.
Tools from HPS 3 are generally thicker than at other sites.
Mean lengths are 5*6 cms, widths k.2 cms and thickness 2.,b cms.
In
particular the massiveness of the core and cobble tools is reflected
in the widths and thickness although a standard deviation of 2.2 and
1.6 cms, respectively indicates the presence of a number of thin,
narrow tools. Tool thickness is also documented in the wide use and
rake angles. The mean rake angle is 73° and the mean work angle is
77°.
Seventy-five percent of the tools have been surface altered.
Oxidation alone or in combination with caliche or varnish is most
common. Only seven tools actually are varnished.
60
Point associations indicate that at least some material is late
Amargosa II (19&) or Amargosa III (12&) (Sand Papago). Several core
forms are similar to those found in the Cochise complex (Sayles and
Antevs 19^1) and the milling stone assemblages of California (Kowta's
1969 "scraping plane").
Though predominately a basalt collection, HPS 3 lacks homo­
geneity, and several distinct patterns may be present.
First, there
is a collection of unidirectional planar cores that have extensive
platform use damage. Second, there is a series of lightly hydrated
obsidian points and bifaces.
Third, a few small unifacially retouched
and laterally used chert and chalcedony flake tools are present.
Finally, a number of lightly varnished and moderately oxidized uni­
facially reduced nodular-ejecta tools were collected.
This information
indicates a temporally and culturally diverse sample.
Papago Tanks;
HPS k
Papago Tanks is a famous Pinacate campsite that was a favored
area of the last Pinacateno (Sand Papago), Juan Carvajales who lived
there until 1912. Lumholtz visited and camped at Papago Tanks while
exploring the Pinacate and Ezell (1955) collected artifacts there on
his initial survey of the north-central Papagueria.
Papago Tanks is southwest of Sykes crater in the Papago arroyo
which drains the Sierra's northwest face.
A lava flow lies above large
semi-permanent plunge pools drained by the arroyo.
The wash draining the pools is forested with mesquite, ironwood, and palo verde.
On either bank terraces with pavements are
found.
The wash widens and spills across a thickly vegetated plain,
eventually draining all the way to the coastal dunes.
Hayden has located San Dieguito sleeping circles and Amargosa
windbreaks and sleeping clearings at Papago Tanks.
A single, aceramic
locus produced sleeping clearings, "shrines," a burial, several points
and ball-shaped quartz manos which Hayden dates to late Amargosa I or
early Amargosa II.
At Papago Tanks numerous loci were collected but
no single locus has a sufficiently large sample to provide an inde­
pendent analytical unit.
Therefore, the sample forms a single, very
limited, picture of the site's great temporal range and spatial dis­
tribution.
The Hayden sample consists of 164 chipped stone tools:
93
flakes (56.7$)t 36 cores (22#) and 35 worked nodules or ejecta (21.3?«).
Basalt is the primary manufacturing material (59.8^)•
Numerous other
stone types are also present, but only obsidian and quartz are found
in amounts greater than 1C$.
Tools are percussion reduced (81.75&), unifacial (5^»296). Less
than half of the tools are retouched, and 64.1# of these are unifacial.
Although primary flakes are the commonest reduction state
(42.7%), considerable numbers of cortex (16.7#) and cortex removal
(22.7?£) flakes were collected.
Secondary flakes and shatter were sel­
dom used for tool manufacture (12.590 •
The high frequency of flakes
retaining cortex is reflected in the predominance of unprepared plat­
forms (40.1^).
Core tools are also often unprepared (32.7^)•
Simple prepared
platforms are found in similar amounts (25^). Unidirectionality
predominates (57»7#) but bidirectionality and multidirectionality are
minor constitutes of the core forms (23*1# and 19«2#, respectively).
Many tools at Papago Tanks are tools percussion reduced unifacially (11.7#) or bifacially (19#), but the unifacially retouched
flake is the commonest tool type.
Utilized cores are frequently found
(lk.7%), but utilized flakes are only modestly represented (9»2#).
Points (6.7#) and bifaces (5»5#) together comprise 12# of tools, while
notching either unifacially or as denticulation constitutes 6.6# of
the sample.
Many core tools show planar use (8.6$) or platform re­
touching (8$), but other tool types display lateral damage particu­
larly on outer faces (11.7#).
Bifacial damage tends to appear on all
edges (1^.2#).
Tool lengths diverge widely, averaging 6.6 cms in length but
having a standard deviation of 3»1 cms.
ness 2.7 cms.
Widths average 2.6 and thick­
Rake single mean is 75° and working angles are an even
wider 79°•
Eighty percent of all artifacts have some surface alteration.
Light oxidation is the prominent visible alteration; caliche is also
common (22#).
Severed variable co-occurrence patterns can be discerned.
Understandably given the basalt and flake frequencies, many tools are
made on basalt flakes.
But there is
alBO a
fairly moderate repre­
sentation of obsidian and quartz flakes (11# and 6.7# each). Only
basalt or other igneous stones are used for cores, while basalt,
rhyolite and quartz appear as nodule or ejecta tools.
63
The Papago Tanks tool inventory is homogeneous only in the use
of local basalt material.
Tool types and manufacture styles vary.
Although many tools are substantially modified by reduction, retouching
appears to associate with a series of reused cores, denticulates, bifaces, points and some flakes.
These slightly weathered tools seem to
represent the site's late component, while a variety of other tools
showing minimal retouch and having moderate oxidation may be consider­
ably earlier.
Tina.ja del Tule:
HPS 20
Near the foot of the Pinacate Peak's southwest flank 1*+ miles
from the Gulf coast is Tinaja del Tule. Here a deep barranca cuts the
lava flow with pools holding water nearly year round.
Above the tanks
are desert pavements interspaced with sandy drainages.
Vegetation is
presently a sparse creosote-bursage cover except for a few saguaros,
palo verde and small ironwood.
Tule is a complex site, located not only on a major trail to
the high tanks at the Peaks but also on the major north-south trail on
the mountain's west side.
Archaeological features include shrines,
trails, "figures," cleared pavement strips ("paradas") and windbreaks.
Although Hayden has noted tools from Malpais on, later Amargosa mate­
rial from wash areas has been the primary sampling focus.
Tule produced an extensive percussion-flaked shell tool indus­
try which I have discussed elsewhere (Rosenthal et al. 1978). Because the
Pinacate shell assemblage is unique, artifact collections from Tule
reflect intensive shell sampling rather than extensive collecting of
64
the stone tools. Therefore, the chipped stone sample from Tule is,
according to Hayden, less representative of tool variety than it should
be.
Forty-seven chipped stone tools were gathered from two loci at
Tinaja del Tule. This inventory contains 38 flake, 5 core and 4
nodular tools principally from obsidian (29.8%) and basalt (25.5$)•
Chert (12.8%), chalcedony (6.490, rhyolite (10.6$) and quartz (10.6$)
are also present adding to the material variety at Tule.
A considerable amount of percussion reduction (83$) equally
divided between unifacial and bifacial work (46.2$, respectively) may
be observed.
Specimens were further modified by pressure or percussion re­
touching (66$).
common.
Unifacial (51*6$) and bifacial (38.7$) retouching are
Large non-cortical flakes are the prominent reduction state
in the sample (50$), but secondary flakes occur also relatively fre­
quently (21$). Few specimens are cortical (5»3$) or cortex removal
(5.3$) flakes, but shatter is common (10.^0). Platform preparation
is limited to the use of a flake scar as a platform.
Five of nine core and nodular tools are unidirectional and the
remainder bidirectional.
Platform preparation is observable on six
of the nine tools.
Tool types represented in the sample include unifacially
(10.6$) and bifacially (6.4$) reduced (chopper-chopping tool) pieces,
utilized flakes (12.8$), utilized cores (4.3$), bifaces (12.8$), points
(17$), denticulates (4.3$)« bifacially retouched flakes (4.3$) and
unifacially retouched-notched tools (2.190.
The prevalent tool, how­
ever, is the simple unifacially retouched flake (25»9&).
Tool lengths averaged 4.8 cms, widths 3*2 cms and thickness
1.5 cms, however, there is considerable value range around these means.
Maximum tool size is 25.6 x 12.2 x 7.2 cms and minimum is 1.5 x .5
x .2 cms.
Angle means are wide, working angles average 72° and rake
angles 64.2°.
Surface alteration is present on less than two-thirds of tools.
Seventy-five percent of all alteration is ranked as light. A rela­
tively high frequency of alteration can be accounted for by green
lichen (2C$), hydration (23.3?0 and oxidation (23»3&)•
Tule's chipped stone collection is one of the smallest studied.
It should primarily represent the late occupation of the site as Hayden
has noted but not collected the extensive earlier material (Hayden
personal communication 1977). The tool's generally light weathering
and the paucity of heavy oxidation and varnish confirm this impression.
The late lightly hydrated obsidian association is the site's most
distinctive tool grouping. Here are found bifaces and points made
from small secondary obsidian flakes and also several small unifacially
retouched pieces of shatter.
It appears that at least point finishing
and perhaps complete manufacture from flakes occurred on site.
The
small number of cores, none of which related to point manufacture, how­
ever, suggests that little or no tool fabrication was locally done.
A rhyolite-chert-quartz complex having limited bifacial per­
cussion reduction and retouching also appears distinctive, and may not
be contemporaneous with the point-making activity.
Although made on
66
flakes the tools are formed from larger primary types rather than
secondary flakes or shatter.
choice.
Selectivity appears to affect flake
Two varnished nodular tools are hard hammer percussed bi-
facially and additionally percussion retouched in alternate and reverse
styles, differing from all other tools.
The use of basalt to make some
tools which are more moderately oxidized than others, is also noted.
Although the tools show great variation in material, manufacture tech­
niques and tool types, the sample is small enough to make interpreta­
tion difficult.
However, multiple visits by many different groups
over time is indicated.
Chivos Tanks;
HPS 23
Slightly southwest of Tule Tanks on a tributary arroyo lies
Chivos Tanks, one of the Pinacate's deepest tinajas. Hayden estimates
that when full Chivos can hold 30,000 gallons of water and, therefore,
it should be considered a permanent water source.
Above the tank is
a broad loess filled lava bench and immediately above lies a paved
area.
Although the tank is an abrupt cut in the lava, it flows into
the broad Chivos drainage a mile away where smoketrees, palo verde,
mesquite and ironwood abound.
This bajio continues for a considerable
distance and leads directly to the historic dune campsites at Sunset
Camp.
Chivos tanks were the permanent water holes supporting these
late refuge camps.
Most archaeological material at Chivos may be found in the
loess bench and on the pavements.
The bench produced ceramics:
sleeping circles and windbreaks exist at both loci. Metates, gyratory
crushers and much laevicardium shell were observed on site, while to
the north several shrines were recorded.
Although Gifford sampled
Chivos in 19^6, he apparently ignored its stone inventory, therefore,
Hayden's material represents the primary collection from a previously
known site.
Forty-one flakes, 5 cores, 2 worked nodular pieces and one
unidentified tool were collected at Chivos Tajiks.
Most of the bS tools
are manufactured from obsidian (Mt-,9#) and chert, rhyolite occur moder­
ately (l6.j$5 and 12.295, respectively).
Seventy-three percent of the tools are percussion reduced,
while "Tl^o show additional retouch modification. In both cases bifacial
work predominates (58.396 reduction, 62.9?*> retouch) making Chivos tool
manufacture styles quite distinctive.
Flake tools are commonly manufactured from small secondary
flakes (**0.590, primary flakes (33*390» and few cortex (4.890 or cor­
tex removal (9-3&) flakes appear in the tool collection. Seven core
and nodular tools were identified.
Most are bidirectional and have
prepared platforms (71«*$)»
Chivos may be a point manufacturing locus as the 14 bifaces
(28.656) and seven points (1^.39°) collected represent a major portion
of the tool kit. Unifacially retouched flakes are likewise common
(28.695).
Accompanying the tools' generally convex edge shape is a rela­
tively small mean tool size:
2.0 cm in thickness.
5-6 cm in length, 2.7 cm in width and
A single large nodular tool 41 x 13 x 8 cms
slightly skews the mean.
Average work and rake angles are relatively
68
acute compared to the total collection.
Mean work single is 65° and
mean rake single is 58°.
Fewer than half the tools are surface altered (46.9&)j mostly
the obsidian tool's hydration is quantified in this variable (52.2&).
Varnish is absent; oxidation occurs on one quarter of all altered
tools.
All but three artifacts must be considered lightly altered.
Almost all tools with ceramics and by point associations
should be either late prehistoric or Sand Papago. The points made at
the site are small triangular projectiles with slightly concave bases.
The Chivos collections' pattern is apparent from variable fre­
quencies alone, however, variable contingency tables do indicate spe­
cific attribute relations.
Heavily oxidized nodular basalt tools
account for the potential early material at the site, while cores and
flakes of obsidian, chert and rhyolite, lightly to moderately unaltered
seem to be the late component.
Notable is the absence of obsidian core
tools despite the presence of incomplete preforms, unfinished bifaces
and points in the sample.
exceptions, to obsidian.
These tool types are restricted, with few
In contrast, cherts, rhyolites and quartz
are found mainly as retouched flakes, while large unifacially percus­
sion reduced nodules and large flakes are of basalt.
Although most tools are both reduced and retouched, percussion
at Chivos is a shaping method prior to pressure retouching, where
elsewhere it may be the sole method of manufacture.
The Chivos point-making style is seen in the variable co­
occurrences. The selection of obsidian, its bifacial percussion
thinning (100$ of sill points), and then its final bifacial pressure
69
flaking (10095) is clearly documented.
Further, although some bifaces
are lightly weathered this alteration is absent from points suggesting
that some bifaces may be earlier than the points.
Chivos' collection is a fairly uniform, homogeneous group of
tools.
It is dominated by small obsidian bifaces, points and flake
tools complemented by small rhyolite and chert pieces of similar tech­
nological style.
Although Hayden1s site data suggest that occupation occurred
from San Dieguito period on, the phases prior to Amargosa III are poorly
represented in the collection. Few tools can be attributed to early
complexes by their surface alteration. The sample instead appears to
represent fairly recent hunting and perhaps butchering activities.
Tina.ja Emilia;
HPS 51
High on the Sierra's eastern slopes below the peaks in a deep
barranca is Tinaja Emilia. In the floor of the barranca a series of
shallow pools have been carved in the igneous stone.
These tanks are
essentially permanent drying up only after prolonged draught.
Several
sediment filled lava benches are located on the arroyos' and tanks*
north sides.
Most artifacts are found near wind breaks constructed on the
benches.
Some tools come from the varnished pavements on the tank's
south side.
Hayden collected
5 worked nodules.
tools at HPS 31s
36 flakes, k cores and
Obsidian is the prevalent stone employed in tool-
making (38.6$) followed by basalt (27.390•
70
Percussion reduction is present on 77*3# of tools and retouch­
ing is even more frequent.
Retouching is present on one or more faces
in equal numbers, however, bifacial percussion reduction is more common
than unifacial work.
Although many tools are on large, primary flakes and bifacial
thinning flakes, a considerable number of secondary flakes are employed
in tool-making (18.2#), and cortical and cortex removal flakes repre­
sent 22,7% of the flake tools.
Despite the moderate frequency of pri­
mary and secondary flakes, platform preparation is not characteristic
of these stages of reduction.
In contrast, the cores all have prepared
platforms.
Most HPS 31 tools can be placed into three distinct categories:
unifacially retouched flakes (31.85a), bifaces (2596), and points (15.99=)•
Another tool type, bifacially retouched flakes, occurs less frequently
(9.1%)•
Flake and core utilization are minimally present (4.3# each).
Dimensions are uniformly small; lengths have means of only 4.3 cms,
widths of 2.8 cms and tools are very thin, just 1.2 cms. Notable also
are the narrow rake angles (mean of 50°) but working angles are fairly
wide, around 70°.
Many tools have altered surfaces (68.290, 8C$ of which are only
lightly weathered. Oxidization (50#) and hydration (30#) account for
most alteration.
Since most material is associated with ceramics and
has only light alteration, it appears attributable to either ceramic
or historic times.
Distinct variable co-occurrence patterns are present in the
tool sample.
Although most flakes are of obsidian (44#), cores are
71
of various stones including basalt, rhyolite, chert and also
obsidian.
The predominant feature of the collection are the obsidian
flake tools which occur as points or bifaces.
Because here, unlike
elsewhere obsidian core tools were collected, some manufacture prob­
ably was conducted on site.
Along with these tools, obsidian work
includes making unifacially retouched flakes, and every obsidian tool
shows some form of retouching.
Rhyolite and cherts, on the other hand,
are often found with use damage only, whether cores or flakes.
Al­
though material differs, tools are similarly manufactured and modified.
These data imply that the tools are of fairly recent origin.
The presence of ceramics and the late point styles confirm the general
impression of late occupancy.
Additionally, other traits like small
size, use of secondary flakes, variety among materials selected and
light alteration may be further features of the Pinacate's late Ceramic
horizon.
Tina.ja del O.jo:
HPS 32
Southeast and downslope from Tinaja Emilia on a broad loesscovered ridge lies the Tinaja del 0jo site.
On the ridge's south side
a deep-cutting arroyo contains the large deep tank from which the site
name is taken.
This arroyo has a dense cover of desert trees:
verde, mesquite and ironwood.
palo
Creosote and bursage survive on the
loess ridge.
Site features include numerous "windbreaks" at the north and
on a ridge above the tanks. Figures and large, late shrines are
72
observable as are elongated cleared-paved alignments that Hayden dates
to Malpais times.
Gyratory crushers were incorporated into the con­
struction of several windbreaks which stand two to three courses high.
Most chipped stone artifacts come from the loess-covered ridge and the
northern area adjacent to the ridge.
Additionally, Hayden has identi­
fied an early 20th century sheep-hunting camp near the tanks.
A large collection of 105 chipped stone tools including 82
flakes, 7 cores and 16 worked nodules was gathered at Tinaja del Ojo.
Although obsidian is the commonly employed stone (33*39o)» rhyolite is
also frequently used (21.990 and basalt (17.15&) somewhat less so.
Percussion reduction is employed to shape 76.2^ of tools and
this work is predominantly bifacial (61.290.
Retouching occurs even
more often (78.1$) and is generally unifacial (59»8£).
Flakes represent 31»3^» cortical and cortex removal flakes
constitute an additional 15.7& and 19*3secondary flakes account for
l8.1# of artifacts.
Bifacially and unifacially percussion reduced tools account
for barely 10# of the sample. Most tools are simply unifacially flakes
(A-2.99»)• however, points (13«390 and bifaces (12.*$), are well repre­
sented in the collection.
Bifacially retouched flakes are also rela­
tively common at Tinaja del Ojo (12.*$).
Tools are small averaging ^.5 cms (mean) in length, 3»3 cms
in width and l.*f cms in thickness. Several tools have skewed this
mean; most tools are smaller but a large nodular tool 13-2 cms in
length increased the mean.
The tools* working angles have a mean of
75°» and the rake angle mean is 62°•
73
Sixty-six percent of tools have surface alteration, either
light (71%) or moderate (21.7#).
Commonest is hydration (3^.8^) or
oxidation alone (29$)•
Obsidian, the most abundant material is mostly found as flake
tools (97.1$)» and rhyolites (73«9&) and cherts (10090 are also mainly
found as flakes. Still, other materials particularly basalt, are
equally divided between flakes and cores or worked nodules.
Small obsidian tool predominance, points, bifaces and unifacially or bifacially retouched flakes typify the
del Ojo sample.
The collection appears fairly homogeneous in style and material selec­
tion, because rhyolite and chert tools manufacture and use are similar
to the obsidian.
The few basalt flake, core and nodular tools varnished and
generally lacking retouch seem to represent a second separate com­
ponent.
The majority of the tools appears to belong to a late prehis­
toric to historic complex with points, bifaces and small tools docu­
menting hunting and perhaps butchering activities.
Tina.ja del Cuervo;
HPS 37
Tinaja del Cuervo lies at the extreme southeast periphery of
the Pinacate lava flows, where a deep arroyo contains a nearly per­
manent water tank.
Below the tank a winding bajio exists, lushly
vegetated with mesquite, ironwood, and saguaro, while northwest of the
main arroyo and closer to the flow a desert pavement maintains a sparse
creosote cover.
74
At Tinaja del Cuervo the major artifact concentrations are on
the pavement.
top.
Sleeping circles and windbreaks are located on the flow
Trails lead away from the tinaja to a shrine where a single,
probably early, figure was constructed.
The tool sample from HPS 37 consists primarily of flakes
(88.8#), with several nodular tools (9^) and two cores (2.290.
Twenty-two and one half percent of the 89 tools are made from rhyolite
and a nearly equal number from basalt (21.390•
There are minor amounts
of three other stones, quartz (13«5$), chert (12.W and obsidian
(12.W.
Most tools are percussion reduced (83.130 and additionally re­
touched (74.2/0.
Unifacial reduction and retouch predominate (58.1&
and 62.1$, respectively), but bifacial work is not uncommon (37«8# and
30.3#).
The predominately flake sample has a high frequency of large,
non-cortical flakes (48.790, a moderate occurrence of cortical flakes
(22.590 and minor constituents of secondary (13.7/0 and cortex removal
(11.2#) flakes.
Tools that are unifacially retouched flakes are the single
most prevalent type (38.2^); however, unmodified utilized flakes have
a high relative frequency (11.290.
Also, bifaces (11.230 and points
(7*950 are common as well as several small bifacially retouched flakes
(5.656) and denticulates (4.590•
Tools are fairly small averaging 4.3 cms in length (mean), 3*1
cms in width and 1.3 cms in thickness. One or two large tools have
75
skewed the mean.
The working angle's mean is 76° and the rake angle's
mean is 59°•
Surface alteration is present on 55# of tools and is often
simple oxidation (53-WO or oxidation combined with groundstaining or
caliche.
Obsidian weathering is found on one half of the tools. Var­
nish was not observed on any tool in the sample.
The typical pattern of tool-making at Tinaja del Cuervo is the
thinning and finishing of tools. Though, specifically, large noncortical flake selection for tool-making is common these specimens are
substantially reduced and finally shaped by retouch.
Final tool-
shaping, whether retouching or notching, correlates strongly with un­
altered or lightly altered (oxidized or hydrated) pieces.
Denticula-
tion of rhyolite, however, is typical of later Amargosan (II) and San
Pedro Cochise point styles (Rosenthal et al. 1978)-
As fewer than
half of the tools are surface altered, data suggest at least two and
possibly three components at Cuervo, based upon alteration and manu­
facture style.
Tool analysis indicates the presence of an obsidian pointmaking (or finishing) activity documented in projectiles and bifaces.
These may be contemporaneous or slightly later than the slightly
oxidized rhyolite flake complex.
Hard hammer serration or denticula-
tion is known for late preceramic times.
The artifacts themselves,
with wide variety of material, yet with a standardized reduction and
retouching technique are quite similar to Sells phase material redovered in the Quijotoa valley northeast of the Pinacate (Rosenthal
et al. 1978; L. Voegler personal communication 1977)*
76
Tina,ja de las Figuras:
HPS 40
Tinaja de las Figuras is, as its name implies, the tank of the
figures.
In its vicinity a number of zoomorphic and naturalistic stone
rock alignments were discovered.
Figuras is located on the eastern slopes of the Pinacate flows
north of Tinaja del Ojo.
paved flow.
Hayden found artifacts on a small, heavily
Its western flank is cut by a deep arroyo containing an
almost completely eroded tank which presently holds little water.
A
small drainage with sandy loess soils crosses the site.
Because of the extinct tanks and distinctive spiral figures,
Hayden assigns portions of the site sample to preceramic Amargosan
times when the tinaja was probably a major ceremonial area for Pinacate
peoples.
More than one period should be represented, because artifacts
lie upon the pavement which incorporates the spiral figures.
At HPS *+0, 106 tools were collected:
21 worked nodules or ejecta.
78 flakes, 7 cores and
Basalt represents 40.6% of the sample
while rhyolite (21.7%) and chert (11.3$) are also common.
Only
of the tools are reduced by percussion flaking, how­
ever, 85% are either pressure or percussion retouched. This shaping
and modification is primarily unifacial (58.3% for reduction and 78.9%
for retouch).
There are greater numbers of bifacial percussed pieces
(31»3%)» small percentages of bifacial retouching (13»3%)•
Flakes recovered from HPS kO (73*6%) suggest the reduction of
local igneous rock occurred at the site. High incidences of cortical
(27.5%) and cortex removal (23-8%) flakes were found (over 50% of all
flakes).
Additionally substantial numbers of primary flakes were
77
collected (32.9&).
Despite the number of flakes representing the ini­
tial stages of core reduction, most tools have prepared platforms with
single flake scars (52.5&)•
A continuum between cores and large cobble tools occurs at
Figuras as many nodular pieces have large flake scars and could have
produced usable flakes.
heavy use.
Excessive damage to their edges indicates
Most of the cores and nodular tools are unidirectionally
flaked (7%).
Reflecting the high frequency of flake occurrence in the
sample, tools can be primarily categorized as unifacially retouched
flake tools (47.2&). Utilized flakes are few in number but utilized
cores are not uncommon (6.6S6).
There is the high relative frequency
of denticulated (2.89s), notched (8.5#), and retouched-notched (7.5$)
tools.
These smaller, distinctively retouched tools total 25.4# of
the sample.
Denticulation and notching is also documented in the
relatively high frequency of concave (29.29o) and concave-convex (24.9&)
edges.
Although tool sizes are not small overall (mean length 4.8 cms
suid mean width 4.1 cms), they are thin with mean thickness of 1.7 cms.
Working angles average a wide 77° and rake angle mean is 61°.
Surface alteration is common suggesting considerable length
of exposure for tools.
Varnish is present on about yjjb of tools often
combined with groundstaining and oxidation, while 9C$ of all tools are
surface altered.
There is a fairly even distribution of light and
moderate alteration.
Retouching is the major method for distinguishing tool types
at Figuras because denticulation and notching reflect the application
of specific technique of retouching.
A pattern of alternate or reverse
retouching occurs on flakes, which is distinguishable from a natural
curved edge, and represents one of the few examples of deliberate
notching among Pinacate material.
HPS 40 maintains enough variation in its sample that establish­
ing tool complexes is quite difficult.
This is primarily because the
local basalt material appears as flake, core and nodular tools.
A
variety of other materials are also employed in differing ways further
confusing patterns of tool-making. However, heavy and moderate oxi­
dation co-occur with varnish and groundstaining, while hydration and
caliche are always light. Basalt is commonly the most heavily altered
material.
Several aspects of Tinaja de las Figuras are interesting.
Although basalt tools predominate over other stones, rhyolite and
chert in particular are common.
Obsidian is, however, less frequent
here than elsewhere. Several heavily altered, varnished, unifacially
and bifacially reduced basalt nodules are found, representing most of
the percussion reduction at the site.
Tools of rhyolite and chert
(and sometimes basalt) are mostly unreduced and are simply modified
by either unifacial or bifacial retouching.
There is a minor component
of denticulated and notched tools present.
Tinaja de las Figuras, probably represents numerous aboriginal
visits of preceramic times.
Most tools are surface altered, and a
paucity of either small secondary flakes of obsidian or distinctive
late point types seems to indicate no late occupation.
The tool sample,
though, culturally and temporally mixed does reveal local manufacture
and selection of initial flakes for tool-making. Much more intensive
study of the site must be performed before the different traditions
are separated.
Tina.ja Doble:
HPS 48
Tinaja Doble is situated north of Papago arroyo and below a
massif on the Pinacate Peak's north face.
As its name implies, Doble
is a double tank with two badly eroded pools.
The site is a paved
plain on a flow top, the front of which is stepped by a series of large
boulders.
The lava flow and adjacent paved sectors have mixed cacti
(saguaro, cholla) and brittlebush cover.
Hayden recorded two aceramic loci here. Oxidized tools came
from both the eastern pavement and rocky flow front.
In among the
flow's boulders, cleared areas with southern exposures have occupa­
tional evidence. The sandy area feeding the tanks produced limited
amounts of broken metates and chipped stone.
From the eastern pavement, Hayden collected a unified tool-kit,
the only identified contemporaneous work area located during the sur­
vey, excepting Sunset Camp's glass artifact industry.
The Doble tool­
kit consisted of several oxidized basalt flakes and bifacially reduced
tools.
Hayden's Doble survey collection contains 86 chipped stone
tools; 63 flakes, 11 cores and 12 nodules.
basaltic (fine 82.6%, scoriacous
is present (7?0•
The sample is primarily
and porphyritic 1.2&).
Rhyolite
A majority of tools are percussion reduced (53-9?0
equally unifacially (47.856) and bifacially (^3.59°)•
Often further
80
modification occurs in the form of retouching (60.5$), but this is
predominately unifacial (71«2$) and only occasionally, bifacial (21.2$).
Flake tools are primarily selected from initial reduction
stages either cortical (33-3$) or partially cortical (30.39s)• Fewer
are primary flakes (28.8$) and secondary flakes (3$) and shatter (1.5%)
are uncommon.
This may indicate local core production, flake removal,
and the subsequent incorporation of initial flakes into the tool-kit.
About two-thirds of the flakes have platforms; these are often simply
prepared (65$), but cortical platforms are present (32.5&)•
Sixteen core and nodular tools show distinct flake-scar direc­
tionality and platform preparation.
tional and 31.3$ bidirectional.
Sixty-two percent are unidirec­
All but 12.5$ have prepared platforms.
Unifacially retouched (32.6$) and utilized (15.1$) flake tools
are common.
sample.
Utilized cores constitute an additional 8.1$ of the
Denticulated (?$) tools occur in small quantities.
were collected but a number of bifaces (10.5$) were.
No points
Large uni­
facially (7$) and bifacially percussion reduced (12.8$) tools comprise
the remaining sample.
Use damage evidence is not distinctly patterned except in its
general occurrence on outer faces.
5.6 cms and thicknesses 2.5 cms.
Mean lengths are 7-^ cms, widths
Work angles have a wide 79° mean
while rank angles are a much narrower 57° mean.
Almost all Doble's tools have some form of surface alteration.
Primarily oxidation combines with groundstaining (48.2$); however, 14
tools (16.7$) are additionally varnished.
The alteration can be ranked
81
as light (3^.9£) or moderate (30.1%) with small numbers of heavily
altered tools (8.*$).
Hayden observed few potsherds at Doble, so he concluded that
the site is primarily preceramic.
The predominance of weathered basalt
tools, the paucity of obsidian and chert, the absence of late prehis­
toric and historic points appear to confirm this impression.
Doble is noteworthy for its high basalt frequency, yet unlike
Sitio Celaya (to be discussed next) it has low quantities of nodular
basalt tools.
Instead, the tool-kit is comprised of basalt flakes
and cores, especially cortex retaining flakes.
Two patterns of tool production cire indicated by attribute co­
occurrences. First, initial reduction (cortical and cortex removal)
are employed for many tools; second, some nodular tools are found.
There is only a moderate amount of shaping during manufacture, and
when present retouching is by percussion and mostly unifacial.
Addi­
tionally extensive use of obsidian and chert is not documented in the
collection; the materials employed seem to be primarily local.
On
site tool manufacture is a definite possibility.
HPS J+8 differs from other Pinacate sites in the above aspects.
Although Doble's nodular tools are similar to others that Hayden
affiliates with San Dieguito material, the moderately weathered basalt
flake associations are atypical and have not been previously described
by Hayden (196?, 1976) for the complex. The selection of cortical
flakes seen here differs from the small chert and obsidian flake tools
found at loci with Patayan ceramics, emphasizing the difference between
the manufacture methods occurring here and at sites having pottery.
82
Sitio Celaya:
HPS 66
Sitio Celaya's name derives from Celaya crater, the large lowrimmed caldera lying at the north edge of the lava flows. Celaya has
a broad eastern rim covered with heavily varnished basalt ejecta.
Within the crater is a large playa fed by a series of arroyos cutting
the rim, maintaining a thick mesquite and palo verde bosque, a habitat
for bighorn sheep and whitetail deer.
Below the east rim outside the crater is a loess-filled,
shallow plain covered with creosote.
An arroyo trenches the crater's
rim and small rock tanks have formed. These retain water for short
periods after storms.
Cultural material at Celaya was found primarily on and in the
pavements beside the rim and dispersed throughout the broad plain and
in sandy minor drainages.
The rim's pavements contain tools and
debitage quarried and fabricated from the ejecta spewed out by the
volcano.
Adjacent to the rim in the pavement are a barely visible
series of sleeping clearings.
Hayden (1976:282) attributes these
tools and features to the earliest occupation of the Sierra Pinacate,
the Malpais phase. The broad plain has produced Amargosan period
shell and stone artifacts and some Patayan ceramics.
Hayden extensively collected the pavement quarry area because
of its distinctive tools, and few now remain in situ on-site.
There­
fore, a near total collection of the "Malpais" locus has been made.
Fewer tools were gathered in the playa area.
Hayden (personal com­
munication 1978) considers this site to be the most important dis­
covery of his Pinacate survey.
83
One hundred fifty three chipped stone tools were collected
from primarily two loci at Sitio Celaya:
ejecta.
71 flakes, 15 cores and 6?
These are predominately of basalt (69.9SO with minor amounts
of obsidian (6.5$), quartz (7.2%) and rhyolite (5»9&).
Only small
numbers of chert, chalcedony and porphyry tools were recovered.
Percussion reduction, occurring on 78.*$ of the tools, is
mostly unifacial (6^.250 with some bifacial (2b.29o) work. Celaya*s
quantity of retouch is remarkably low (26.1%), and when apparent is
most often unifacial (62.5^).
The 71 flake tools are made on large,
non-cortical flakes (^3»850, cortex (17»8%), or cortex removal flakes
(17.8&). Platforms are mostly cortical (^8.7^) or planar (7.750«
unprepared types.
On *+9 core and nodular tools, platforms are also
predominately unprepared (67.^), and the tools are almost always
unidirectional (83.790•
Tool types are dominated by unifacially and (36.690 bifacially
percussion reduced ejecta (11.890.
Unifacially retouched flakes
(11.190 are present in some quantities sis are utilized cores (9»2/0»
Points and bifaces are uncommon (2.6# and 6.5$, respectively) and few
denticulates, notched tools and other small flake objects were
collected.
Tools are on the average larger here them elsewhere.
length is 8.5 cms, width 6.8 cms and thicknesB 3 cms.
Mean
These figures
indicate that tools are wide and thick sis well as long, reflecting the
large tools made on ejecta.
average 63°•
Working angle mean is 78.5°, rake angles
8if
Almost all of the 65 nodular-ejecta of the 153 tools have
varnish;
9k.1% have some form of surface alteration.
present on another 5^»
Oxidation is
Groundstaining is also common.
The Sitio Celaya sample has several distinct manufacture and
stone use patterns that were documented by variable CROSSTABS.
basalt nodular and ejecta tools predominate (40.59o).
tools are also common (22.9&).
First,
Basalt flake
Cores and nodular tools are made only
from basalt, andesite and rhyolite, while flake tools represent the
small percentages of chert, quartz and obsidian present at the site.
Materials are selectively worked.
Careful attention to plat­
form preparation is found on cherts and obsidians while basalt and
rhyolite flakes have cortical platforms.
The site's importance lies in the extensive surface alteration
found on most tools.
About one in three tools is lightly altered but
these are mostly obsidian flakes or quartz with green lichen. When
other forms of alteration are present they are moderate to heavy
groundstaining, oxidation and varnishing. In particular, combinations
of groundstain, oxidation and varnish in moderate or heavy accumula­
tions are found on one-quarter of the tools.
The hydrated obsidian and caliche encrustation are both found
on tools collected from the plain and sandy drainages.
small bifaces and late prehistoric points were found.
to document the recent component at Celaya.
Here several
These appear
Most other tools, however,
with heavy and moderate surface alteration, come from the rise beside
the crater either from within the varnished pavement or adjacent to
the sleeping clearings.
85
Sitio Celaya is the site which contradicts the usual toolmaking styles of other Pinacate sites. The sample is dominated by
core and nodular tools where elsewhere flakes predominate. Percussion
reduction is the prominent technique for final shaping in contrast to
the retouching at other sites.
Unifacially and bifacially reduced
large, natural flakes and ejecta tools characterize the sample, while
at other sites a unifacially retouched flake tool is commonest. Spe­
cialized small tools are minimally present despite the large sample,
and these may be partially accounted for by the presence of late points
and bifaces.
Sitio Celaya appears dissimilar to other sites and has three
probable components.
The heavily oxidized, varnished basalt uni­
facially and bifacially percussion reduced tool is distinct as are
the few utilized basalt cortical and cortex removal flakes.
A moder­
ately to lightly oxidized rhyolite and basalt, retouched and used flake
complex also is indicated. Finally, the aforementioned small obsidian
and chert collected from a separate locus and having little or no
alteration, may also differ.
Sitio La Playa;
HPS 68
Northwest of Cerro Colorado is the most recently discovered
major Pinacate site, Sitio La Playa. Here a wide playa which contained
water during the late Pleistocene is located.
The area appears to
again have been inundated during the medithermal and presently holds
water after storms.
Low lava flows border the Playa on the south.
86
Archaeological material was collected from sandy areas of water
sorted cinders beside the lowest flow.
Cached cardium shell and gyra­
tory crushers were noted and projectile points and ceramics were
collected.
Due to changes in the lake level, Hayden suggests that material
remains at La Playa represent a lakeshore camp of late Preceramic and
Ceramic times.
Site activities may have included camping to kill
migratory water fowl using the lake, or hunting game watering in the
vicinity.
A small artifact sample (^5 tools) was collected at Sitio La
Playa, including 36 flakes, 3 cores and 6 nodular tools.
Obsidian
tools are commonest (20.2%) followed by quartz (17.8#), rhyolite
(15.65o), and chert (15.6%). Basalt is infrequently found (11.190 as
are porphyritics.
Slightly more than one-third of the tools are reduced by per­
cussion, primarily unifacially (68.8%), or bifacially (2590•
Most
tools are, in contrast, frequently retouched (73*3%)» mostly uni­
facially (63.5%)» and sometimes bifacially (2*+.2%).
Although cortical flakes are few (15«3%), cortex removal types
are often seen (23«7%)«
Large non-cortical (28.99s) and bifacial
thinning flakes (21.1%) are the major component and there is but a
minor percentage of secondary flakes.
their platforms.
Twenty-three flakes retained
Most have single flake scars (3*t.8%) but some are
multifaceted platforms (21.7%).
Forty percent of all tools at Sitio La Playa are unifacially
retouched flakes and an additional 20% are simple, unmodified,
87
utilized flakes.
Two points
and two bifaces
lected along with five bifacially retouched flakes.
were col­
Conspicuous by
their low representation are bifacially reduced tools and no unifacially percussion reduced types were collected.
Tools regularly have convex working edges; however, some simple
concave edges reflect the sample's minor amount of notching.
Tools
are also small with means of 3.8 cms in length, 3 cms in width and 1.3
cms in thickness.
Working angles are a wide mean of 70.3° while the
rake angle mean is an acute 53°•
Although two-thirds of the material is surface altered, most
weathering is light (9056).
Varnish is absent; groundstaining is mini­
mally observable (12.2$), but light oxidation is prevalent and light
hydration also fairly common.
Late prehistoric and his toric point types and Patayan ceramics
are present at La Playa, therefore, at least a portion of the tools
are contemporaneous with these artifacts.
The lake's covering of the
site during wetter Preceramic (medithermal) times means that little
material is attributable to Amargosa I or earlier times.
The La Playa collection is a fairly uniform, simple tool sample
with several co-occurring variables delineating manufacture style.
Limited percussion reduction characterizes the flake tool sample, with
this technique appearing primarily during biface and point preparation,
being uncharacteristic of general tool fabrication.
Most tool modi­
fication is retouching which is applied even to reused cores.
The
commonest tool is a unifacially retouched, large non-cortical or bi­
facial thinning flake while another prevalent form, the unmodified,
88
utilized flake is simply a cortex flake selected for use.
For the most
part, points and bifaces are retouched secondary flakes.
This sample appears to be the most homogeneous recovered.
Al­
though there is variety, the consistent use of unreduced flakes for
tool manufacture, and the absence of nodules for unifacially and bifacially percussion reduced tools indicates uniformity of technique.
The low basalt use incidence differentiates this flake complex from
other sites, while its associations, point style, and physical loca­
tion permit its assignment to the late prehistoric to historic period.
This brief summary of the 13 site samples details general
features only. Specific attribute frequencies are presented in the
site comparison tables which are discussed in the succeeding chapter.
CHAPTER 7
site sample COMPARISONS
Changing variable frequencies among the 13 samples indicate
that attributes of raw material selection, tool fabrication techniques
and surface alteration differ from site to site. CROSSTABS and ONEWAY
were employed to tabulate frequencies while "Chi-square" and "P-tests"
were applied to test for significant deviation.
This chapter dis­
cusses where actual frequencies deviate from expected frequencies and
what attributes are contributing to the deviances.
Except for plat­
form preparation, all paired variables appear to be strongly associ­
ated.
Several cautionary notes should be made.
It must be remembered
that these samples are not randomly selected; therefore, inferring the
total site population attributes based on these frequencies is not
possible.
All the results suggest are the characteristics of these
13 site samples.
Many attributes are not present at certain sites and
although this is useful negative data, it presents a statistical interpretational problem. Too many small frequencies or absent categories
(empty cells) can yield significant chi-squares when there is no
association between two variables.
This problem may be partially alleviated by combining variable
states, and this has been done.
However, it does not make sense
89
90
conceptually to combine, arbitrarily, attributes like utilized stone
with other categories just because they infrequently occur.
The suggested solution to this problem is to determine just how
small values are contributing to the table. When the total number of
cases is large, cells with low frequencies generally contribute little
to the chi-square value; however, when numerous empty cells are present
the chi-square statistic probably is not an acceptable statement of
attribute relations.
By comparing the expected frequency (theoretical values) with
the obtained frequency of each variable's states, the deviation from
random occurrence can be observed. Then using the chi-square statistic
the discrepancy can be computed and judged as significant. Unfor­
tunately the SPSS sub-program CROSSTABS does not present the expected
frequency. However, this may be calculated by multiplying the column
frequency by the row frequency for each variable state and then
dividing the figure by the table's grand total.
Because the contingency tables present both the absolute and
the relative frequency of occurrence, results cannot be summarized as
simply an increased frequency. The particular tool attributes which
have increased or decreased proportionally provide the means for
interpreting why two variables are associated. The terms — overrepresented and under-represented — are used to imply not only greater
or fewer numbers, but also an increase or decrease in proportional
representation.
Table 2 summarizes material distributions at the 13 sites.
Quantities and percentages of raw material vary widely.
Basalt
Table 2. Crosstabulation of material by site. — The first number in
each site = count; the second number = row
the third
number = column %\ smd the fourth number = total %.
Material
Scor.
Basalt
?
Site
No.
Fine
Basalt
1
1
10
25.6
7.1
.9
1
2.6
7.7
.1
1
2.6
2.9
.1
6
15.4
4.8
.5
0
0
0
0
0
0
0
0
21
25.0
4.4
1.9
0
0
0
0
0
0
0
0
5
6.0
4.0
.5
0
0
0
0
2
2.4
1.3
.2
4
4.8
4.7
.4
2.4
9.5
53
53,0
11.1
4.9
0
0
0
0
3
3.0
8.6
.3
8
8.0
6.4
.7
0
0
0
0
12
12.0
7.6
1.1
6
6.0
7.0
.5
1
1.0
4.8
.1
96
58.5
20.0
8.6
3
1.8
23.1
.3
6
3.7
17.1
.5
3
1.8
2.4
.3
0
0
0
0
19
11.6
12.0
1.7
8
4.9
9.3
.7
2
1.2
9.5
.2
20
13
27.7
2.7
1.2
0
0
0
0
1
2.1
2.9
.1
5
10.6
4.0
.5
0
0
0
0
13
27.7
8.2
1.2
6
12.8
7.0
.5
3
6.4
14.3
.3
23
6
12.2
1.3
.5
0
0
0
0
1
2.0
2.9
.1
8
0
0
0
0
22
44.9
13.9
2.0
8
16.3
6.4
.7
16.3
9.3
.7
0
0
0
0
12
27.3
2.5
1.1
0
0
0
0
0
0
0
0
4
9.1
3.2
.4
0
0
0
0
17
38.6
10.8
1.5
4
9.1
4.7
.4
4.8
.1
32
20
19.0
4.2
1.8
0
0
0
0
6
5.7
17.1
.5
23
21.9
18.4
2.1
1
1.0
11.1
.1
35
33.3
22.2
3.2
11
10.5
12.8
1.0
0
0
0
0
37
20
22.5
4.2
1.8
2
2.2
15.^
.2
3
3.4
8.6
.3
20
22.5
16.0
1.8
4
4.5
44.4
.4
11
12.4
7.0
9
10.1
10.5
1.0
.8
1
1.1
4.8
.1
2
3
if
31
Andesite
Rhy^lite
Gragite
Obsidian
7
Chgrt
Chalc«
9
5
5
12.8
12.8
23.8
.5
5.8
.5
2
.2
1
2.3
91
2 by site. -- The first number in
md number = row %; the third
fourth number = total %.
Material
ite
6
Gragite
ObsidiEin
Chgrt
Chalced.
9
Quartz
10
Por.
Basalt
PO
Othgr
0
Row
Total
0
0
0
0
5
12.8
5.8
.5
5
12.8
23.8
.5
10
25.6
7.2
.9
0
0
0
0
1
2.6
4.0
.1
OM
0
0
0
39
3.5
8
5
0
0
0
0
5
0
0
5
0
0
0
0
2
2.it
1.3
.2
it
it.8
4.7
.it
2
2.4
9.5
.2
47
56.0
33.8
4.2
0
0
0
0
3
3.6
12.8
.3
OM
0
0
0
84
7.6
8
0
4
7
0
0
0
0
12
12.0
7.6
1.1
6
6.0
7.0
.5
1
1.0
4.8
.1
16
16.0
11.5
1.4
0
0
0
0
1
1.0
4.0
.1
OM
0
0
0
100
9.0
3
8
4
3
0
0
0
0
19
11.6
12.0
1.7
8
4.9
9.3
.7
2
1.2
9.5
.2
19
11.6
13.7
1.7
2
1.2
10.0
.2
6
3.7
24.0
.5
OM
0
0
0
164
14.8
5
6
0
5
0
0
0
0
13
27.7
8.2
1.2
6
12.8
7.0
.5
3
6.4
1^.3
.3
5
10.6
3.6
.5
0
0
0
0
1
2.1
4.0
.1
OM
0
0
0
47
4.2
8
3
if
7
0
0
0
0
22
itit.9
13.9
2.0
8
16.3
9.3
.7
0
0
0
0
2
4.1
1.4
.2
2
4.1
10.0
.2
0
0
0
0
OM
0
0
0
49
4.4
if
1
2
it
0
0
0
0
17
38.6
10.8
1.5
it
9.1
4.7
.it
1
2.3
4.8
.1
2
4.5
1.4
.2
1
2.3
5.0
.1
3
6.8
12.0
.3
OM
0
0
0
44
4.0
3
9
it
1
1
1.0
11.1
.1
35
33.3
22.2
3.2
11
10.5
12.8
1.0
0
0
0
0
6
5.7
4.3
.5
3
2.9
15.0
.3
0
0
0
0
OM
0
0
0
105
9.5
0
5
0
8
it
4.5
itit.it
.it
11
12.4
7.0
1.0
9
10.1
10.5
.8
1
1.1
4.8
.1
12
13.5
8.6
1.1
3
3.4
15.0
.3
4
4.5
16.0
.4
OM
0
0
0
89
8.0
Table 2—-Continued.
Material
Site
No.
Fine
Basalt
Scor.
Basalt
Andesite
Rhy^lite
Gragite
Obsidian
Chgrt
Chanced.
4o
45
42.9
9.4
4.1
1
1.0
7.7
.1
8
7.6
22.9
.7
22
21.0
17.6
2.0
2
1.9
22.2
.2
7
6.7
4.4
.6
12
11.4
14.0
1.1
0
0
0
0
48
71
82.6
14.8
6.4
4
4.7
30.8
.4
1
1.2
2.9
.1
6
7.0
4.8
.5
0
0
0
0
1
1.2
.6
.1
1
1.2
1.2
.1
0
0
0
0
66
107
69.9
22.3
9.6
1
.7
7.7
.1
5
3.3
14.3
.5
9
. 5.9
7.2
.8
0
0
0
0
10
6.5
6.3
.9
5
3.3
5.8
.5
2
1.3
9.5
.2
68
5
11.1
1.0
, .5
1
2.2
7.7
.1
0
0
0
0
6
13.3
4.8
.5
2
4.4
22.2
.2
9
20.0
5.7
.8
7
15.6
8.1
.6
4
8.9
19.0
.4
Column 479
Total 43.2
13
1.2
35
3.2
125
11.3
9
.8
158
14.2
86
7.7
21
1.9
Number of missing observations, 1.
92
Material
ite Qragite Obsidian Chgrt Chanced.
Quartz Basalt Oth^r
in
?0
2
0
6
0
2
1.9
22.2
.2
7
6.7
4.4
.6
12
11.4
14.0
1.1
0
0
0
0
0
0
0
0
2
1.9
10.0
.2
6
5.7
24.0
.5
6
0
8
0
0
1
1.2
.6
.1
1
1.2
1.2
.1
0
0
0
0
1
1.2
.7
.1
1
1.2
5.0
.1
0
0
0
0
5
3.3
5.8
.5
2
1.3
9.5
.2
11
7.2
7.9
1.0
3
2.0
15.0
.3
0
0
0
0
5
0
0
0
1M
0
0
0
105
9.5
OM
86
7.7
0
0
0
9
9
2
8
0
0
0
10
6.5
6.3
.9
6
3
8
5
2
4.4
22.2
.2
9
20.0
5.7
.8
7
15.6
8.1
.6
4
8.9
19.0
.4
8
17.8
5.8
.7
3
6.7
15.0
.3
0
0
0
OM
0
0
0
5
3
9
.8
158
14.2
86
7.7
21
1.9
139
12.5
20
1.8
2.3
0
25
Row
Total
OM
0
0
153
13.8
0
45
4.1
0
1110
100.0
93
representation is as high as 82.6% at HPS 48 and as low as 11# at HPS
68, while quartz distribution is as high as 56# at HPS 2 and as low as
1.2# at HPS 48. Cherts and chalcedonies are minor portions of the
samples at all sites; however, they are relatively more frequent at
HPS 68
than elsewhere. Rhyolite is moderately but consistently
represented among sites; in contrast obsidian is scarce or absent at
two sites, HPS 1 and 48, and overabundant at two sites, HPS 23 and 31•
From the data in Table 2, the following groups of material
occurrences are discernible.
1.
Basalt is over-represented at HPS 48 and 66.
2.
Basalt and rhyolite are over-represented at HPS 3i *+» and 40.
3. Basalt is under-represented at HPS 68.
4.
Obsidian and chert are over-represented at HPS 23 and 31.
5.
Quartz is over-represented at HPS 2.
Flake tools dominate the Pinacate collection (68.2#), varying
from above 70# to below 50# among samples (Table 3)•
Sites HPS 23» 37
and 68 are contributing relatively greater amounts of flakes to the
total collection. In contrast, sites HPS 3 and 4 have higher relative
frequencies of core tools while HPS 40 and particularly HPS 66 have
numerous nodular stone pieces or ejecta. Sites grouped by categories
indicate that:
1. Flakes are over-represented at HPS 20, 23» 31» 32, 371 and 68„
2.
Cores are over-represented at HPS 3 and
3.
Nodular pieces or ejecta are over-represented at HPS 66.
Tool fabrication patterns may be suggested by observing vari­
ables indicating where tool percussion reduction and retouch occur.
94
Table 3«
Crosetabulation of category by site. — See Table 2.
Flake
1
Category
Worked
Nodule
Core
2
3
1
28
71.8
3.7
2.5
7
17.9
4.9
.6
4
10.3
1.9
.4
0
0
0
0
39
3.5
2
60
71.4
7.9
5.4
12
14.3
8.3
1.1
11
13.1
5.3
1.0
1
1.2
50.0
.1
84
7.6
3
59
59.0
7.8
5.3
29
29.0
20.1
2.6
12
12.0
5.8
1.1
0
0
0
0
100
9.0
4
91
55.5
12.0
8.2
36
22.0
25.0
3.2
36
22.0
17.4
3.2
1
.6
50.0
1
164
14.8
20
37
78.7
4.9
3.3
6
12.8
4.2
.5
4
8.5
1.9
.4
0
0
0
0
47
4.2
23
41
83.7
5.4
3.7
5
10.2
3.5
.5
3
6.1
1.4
.3
0
0
0
0
49
4.4
31
35
79.5
4.6
3.2
4
9.1
2.8
.4
5
11.4
2.4
.5
0
0
0
0
44
4.0
32
82
78.1
10.8
7.4
7
6.7
4.9
.6
16
15-2
7.7
1.4
0
0
0
0
105
9.5
37
78
87.6
10.3
7.0
2
2.2
1.4
.2
9
10.1
4.3
.8
0
0
0
0
89
8.0
4o
76
71.7
10.0
6.8
7
6.6
4.9
.6
23
21.7
11.1
2.1
0
0
0
0
106
9.5
Site
No.
Utilized
Stone
4
Row
Total
95
Table 3—Continued.
Site
No.
Flake
1
Category
Worked
Nodule
Core
2
Utilized
Stone
3
4
Row
Total
48
63
73.3
8.3
5.7
11
12.8
7.6
1.0
12
14.0
5.8
1.1
0
0
0
0
86
7.7
66
71
46.4
9.4
6,4
15
9.8
10.4
1.4
67
43.8
32.4
6.0
0
0
0
0
153
13.8
68
37
82.2
4.9
3.3
3
6.7
2.1
.3
5
11.1
2.4
.5
0
0
0
0
45
4.1
Column
Total
758
68.2
144
13.0
207
18.6
2
.2
1111
100.0
96
Percussion reduction is employed to remove flakes from cores, to thin
and shape flakes, and to establish working edges.
cussion is applied to either one or two faces.
Hard hammer per­
Table b summarizes the
varied distributional data on percussion reduction at the 13 sites.
Unifacial flaking is most frequent at HPS 1 (92.2^) and although never
very common, bifacial flaking is relatively more frequent at HPS 20
(**7.5&), HPS 23 (56.8$), HPS 31 (50&), and HPS 32 (61.7550.
Alternate
and reverse techniques appear in increased amounts at HPS *+ and 66 and
at the quartz dominated HPS 2.
Site grouping by percussion reduction
include:
1. Percussion being over-represented at HPS 1.
2. Percussion being under-represented at HPS 40, ^8 and 68.
3.
Unifacial flaking being over-represented at HPS 1, and 2.
k.
Unifacial flaking being under-represented at HPS 23 and 32.
5.
Bifacial flaking being over-represented at HPS 20, 23, 311 32
and 37.
6.
Bifacial flaking being under-represented at HPS 1, 2, kO and 68.
Retouching is present on 60.5$ of all tools. As Table 5 shows,
this variable has an increased frequency at eight of the sites, HPS 2,
20, 23, 31, 32, 37t *+0» and 68, and is poorly represented at one site,
HPS 66.
Unifacial and bifacial retouching differ distinctly.
Sites
HPS 1, 2, and bo have greater quantities of unifacially retouched
tools than expected, while site HPS 23 alone has more bifacial than
unifacial retouching.
HPS 68 is noteworthy because it has a series of
97
Table 4.
Crosstabulation of percussion reduction by site
Table 2.
— See
Percussion Reduction
Site
No.
Unifacial
1
32
86.5
7.2
3.9
2
5.4
.7
.2
0
0
0
0
0
0
0
0
3
8.1
4.8
.4
2m
0
0
0
37
4.6
2
52
78.8
11.7
6.4
9
13.6
3.1
1.1
0
0
0
0
0
0
0
0
5
7.6
8.1
.6
18m
0
0
0
66
8.1
3
42
53.8
9.5
5-2
22
28.2
7.7
2.7
0
0
0
0
2
2.6
18.2
.2
12
15.4
19.4
1.5
22m
0
0
0
78
9.6
if
57
42.5
12.9
7.0
49
36.6
17.1
6.0
1
.7
10.0
.1
0
0
0
0
30m
0
0
0
134
16.5
18
45.0
4.1
2.2
19
47.5
6.6
2.3
0
0
0
0
2.5
9.1
.1
27
20.1
43.5
3.3
2
5.0
3.2
.2
7m
0
0
0
40
4.9
15
40.5
3.4
1.8
21
56.8
7.3
2.6
0
0
0
0
0
0
0
0
1
2.7
1.6
.1
12m
0
0
0
37
4.6
15
44.1
3.4
1.8
17
50.0
5.9
2.1
0
0
0
0
1
1
2.9
9.1
.1
2.9
1.6
.1
10m
0
0
0
34
4.2
32
31
38.3
7.0
3.8
50
61.7
17.5
6.2
0
0
0
0
0
0
0
0
0
0
0
0
24m
0
0
0
81
10.0
37
42
56.8
9.5
5.2
29
39.2
10.1
3.6
1
1.4
10.0
.1
0
0
0
0
2
2.7
3.2
.2
15m
0
0
0
74
9.1
30
60.0
6.8
3.7
15
30.0
5.2
1.8
0
0
0
0
0
0
0
0
1
2.2
1.6
.1
56m
0
0
0
50
6.2
20
23
31
40
Bifacial
Alternate
Reverse
1
Mul­
tiple
0
Row
Total
98
Table 4—Continued.
Percussion Reduction
Site
No.
Unifacial
1
Bifacial
Alternate
2
3
22
47.8
5.0
2.7
20
43.5
7.0
2.5
0
0
66
77
64.2
17.4
9.5
68
Column
Total
48
Reverse
Mul­
tiple
5
Row
Total
0
1
2.2
1.6
.1
40M
0
0
0
46
5.7
0
0
3
6.5
27.3
.4
29
24.2
10.1
3.6
8
6.7
80.0
1.0
4
3.3
36.4
.5
2
1.7
3.2
.2
33M
0
0
0
120
14.8
10
66.7
2.3
1.2
4
26.7
1.4
.5
0
0
0
0
0
0
0
0
1
6.7
1.6
.1
30M
0
0
0
15
1.5
443
54.6
286
32.2
10
1.2
11
1.4
62
7.6
299M
0
812
100.0
Number of missing observations, 299
99
Table 5-
an;e
No.
Crosstabulation of retouch by site. ~ See Table 2.
Retouch
Unifacial
1
Bifacial
2
Alternate
Reverse
3
4
KQW
Total
0
1
21
87.5
4.7
3.1
0
0
0
0
3
12.5
12.5
.4
0
0
0
0
15m
0
0
0
24
3.6
2
51
85.0
11.4
7.6
2
3.3
12.5
.3
0
0
0
0
60
8.9
41
78.8
9.2
6.1
2
3.3
8.3
.3
0
0
0
0
24m
0
0
0
3
5
8.3
2.7
.7
11
21.2
6.0
1.6
48m
0
0
0
52
7.7
4
51
64.6
11.4
7.6
24
30.4
13.0
3.6
3
3.8
12.5
.4
1
1.3
6.3
.1
85m
0
0
0
79
11.8
20
16
53.3
3.6
2.4
11
36.7
6.0
1.6
1
3.3
4.2
.1
17m
0
0
0
30
4.5
23
13
37.1
2.9
1.9
22
62.9
12.0
3.3
0
0
0
0
2
6.7
12.5
.3
0
0
0
0
14m
0
0
0
35
5.2
31
18
50.0
4.0
2.7
17
47.2
9.2
2.5
0
0
0
.0
1
2.8
6.3
.1
8m
0
0
0
36
5.4
32
47
58.7
10.5
7.0
33
41.3
17.9
4.9
0
0
0
0
0
0
0
0
25m
0
0
0
80
11.9
37
39
60.9
8.7
5.8
20
31.3
10.9
3.0
4
6.3
16.7
.6
1
1.6
6.3
.1
25m
0
0
0
64
9.5
69
78.4
15.4
10.3
12
13.6
6.5
1.8
4
4.5
16.7
.6
3
3.4
18.8
.4
18m
0
0
0
88
13.1
40
100
Table 5—Continued,
one
No.
Retouch
Row
Total
Unifacial
1
Bifacial
2
Alternate
3
Reverse
k
0
kS
37
71.2
8.3
5.5
11
21.2
6.0
1.6
2
3.8
8.3
.3
2
3.8
12.5
.3
3^M
0
0
0
66
25
62.5
5.6
3.7
10
25.0
5.^
1.5
5
12.5
20.8
.7
0
0
0
0
113M
0
0
0
68
20
62.5
^.5
3.0
8
25.0
^.3
1.2
0
0
0
0
k
12.5
25.0
.6
13M
0
0
0
32
k.S
Column
Total
H8
66.7
l8if
2?A
2k
3.6
16
2.^
^39M
0
672
100eO
Number of missing observations, ^39.
52
7.7
kO
6.0
101
tools with reverse retouching. Inspecting the results, four groupings
appear to be present.
1.
Retouching is over-represented at HPS 2, 21, 32, 37, and 40.
2.
Unifacial retouching is over-represented at HPS 1, 2, and 40.
3.
Retouching is under-represented at HPS 3, 4, and 66.
4.
Bifacial retouching is over-represented at HPS 23, 31, and 32.
Tool size, thickness, steepness and curvature can be partly
controlled by the flake or core the flintknapper selects to further
modify or use (White and Thomas 1972:278-279)*
One can identify flake
selection among the 13 sites by reviewing the different occurrences of
the variable stage of reduction.
Twenty-six percent of all the collection's tools are made on
large non-cortical (primary) flakes, but both cortical and cortex re­
moval flakes are found in moderate amounts.
Similarly, HPS 2, 20, and
37, over one-third of the flake tools, are made from large non-cortical
(primary) flakes. Cortical flakes are frequently used at HPS 37, 40,
and 48, while cortex removal flakes are commoner at HPS 48 and 68.
Small, non-cortical (secondary) flakes are more often selected for
tool-making by individuals visiting HPS 31, if the collected sample
is indicative of the site as a whole.
Combining selective factors for
flakes, the following groupings can be suggested (Table 6).
1. Initial flakes (cortex and cortex removal) are over-represented
at HPS 40 and 48.
2. Primary and special primary flakes are over-represented at HPS
1, 2, 20, 37, and 68.
Table 6. Crosstabulation of flake reduction stage by site number.
See Table 2.
Flake Reduction Stage
Cortical
1
Cortex
Removal
2
Primary
3
Special
Primary
4
Secondary
5
Blade
6
Sec. Trim,
7
1
2
6.9
1.6
.3
5
17.2
3.5
.6
10
34.5
3.4
1.3
6
20.7
10.2
.8
2
6.9
1.8
.3
1
3.4
25.0
.1
1
3.4
25.0
.1
2
3
5.0
2.4
.4
7
11.7
4.9
.9
31
51.7
10.4
4.0
5
8.3
8.5
.6
11
18.3
9.9
1.4
2
3.3
50.0
.3
0
0
0
0
3
5
8.6
4.1
.6
12
20.7
8.5
1.6
21
36.2
7.0
2.7
6
10.3
10.2
.8
10
17.2
9.0
1.3
0
0
0
0
0
0
0
0
4
16
16.8
13.0
2.1
21
22.1
14.8
2.7
4o
42.1
13.4
5.2
4
4.2
6.8
.5
12
12.6
10.8
1.6
0
0
0
0
0
0
0
0
20
2
5.4
1.5
.3
2
5.4
1.4
.3
19
51.4
6.4
2.5
2
5.4
3.4
.3
8
21.6
7.2
1.0
1
2.7
25.0
.1
0
0
0
0
23
2
4.9
1.6
.3
4
9.8
2.8
.5
13
31.7
4.4
1.7
4
9.8
6.8
.5
17
41.5
15.3
2.2
0
0
0
0
0
0
0
0
31
3
8.8
2.4
.4
7
20.6
4.9
.9
10
29.4
3.4
1.3
4
11.8
6.8
.9
8
23.5
7.2
1.0
0
0
0
0
0
0
0
0
32
14
16.9
11.4
1.8
16
19.3
11.3
2.1
26 31.3
8.7
3.4
7
8.4
11.9
.9
15
18.1
13.5
1.9
0
0
0
0
0
0
0
0
37
18
22.8
14.6
2.3
9
11.4
6.3
1.2
40
50.6
13.4
5.2
0
0
0
0
11
13.9
9.9
1.4
0
0
0
0
0
0
0
0
40
21
26.9
17.1
2.7
18
23.1
12.7
2.3
26
33.3
8.7
3.4
3
3.8
5.1
.4
6
7.7
5.4
.8
0
0
0
0
2
2.6
50.0
.3
Site
No.
102
>y Bite number. ~
iduction Stage
r
Crtx.Rmvi.
Blade Sec. Trim.
Sp. Pri.
6
7
.
8
Shatter
9
Crtx.Rmvi.
Blade
10
Flake
Core
11
Row
Total
0
0
0
0
0
0
0
0
0
0
0
0
0
10m
0
0
0
29
3.8
0
0
0
0
7
6.9
100.0
.3
0
0
0
0
1
1.7
4.5
.1
0
0
0
0
0
0
o ..
0
24m
0
0
0
6o
7.8
0
0
0
0
0
0
0
0
2
3.4
9.1
.3
1
1.7
lOO'.O
.1
1
1.7
16.7
.1
km
0
58
7.5
0
0
0
0
0
0
0
0
0
0
0
0
1
1.1
4.5
.1
0
0
0
0
1
1.1
16.7
.1
69m
0
0
0
95
12.3
1
2.7
25.0
.1
0
0
0
0
0
0
0
0
3
8.1
13.6
.if
0
0
0
0
0
0
0
0
10m
0
0
0
37
4.8
0
0
0
0
0
0
0
0
0
0
0
0
1
2.4
4.5
.1
0
0
0
0
0
0
0
0
8m
0
0
0
41
5.3
0
0
0
0
0
0
0
0
0
0
0
0
2
5.9
9.1
.3
0
0
0
0
0
0
0
0
10m
0
0
0
34
4.4
0
0
0
0
0
0
0
0
0
0
0
0
5
6.0
22.7
.6
0
0
0
0
0
0
0
0
22m
0
0
0
83
10.8
0
0
0
0
0
0
0
0
0
0
0
0
1
1.3
4.5
.1
0
0
0
0
0
0
0
0
10m
0
0
0
79
10.2
0
0
0
0
2
2.6
50.0
.3
0
0
0
0
2
2.6
9.1
.3
0
0
0
0
0
0
0
0
28M
0
78
10.1
1
3.4
25.0
.1
1
3.4
25.0
.1
2
3.3
50.0
.3
0
0
0
0
0
0
0
0
Table 6--Continued
Flake Reduction Stage
Site
No.
Cortical
1
Cortex
Removal
2
Primary
3
Special
Primary
k
Secondary
5
Blade
6
Sec. Trim
7
^8
22
33.3
17.9
2.8
20
30.3
1^.1
2.6
19
28.8
6.if
2.5
2
3.0
3.^
.3
2
3.0
1.8
.3
0
0
0
0
0
0
0
0
66
13
17.8
10.6
1.7
11
15.1
7.7
l.k
32
^3.8
10.7
k.l
8
11.0
13.6
1.0
5
6.8
*t.5
.6
0
0
0
0
0
0
0
0
68
2
5.1
1.6
.3
10
25.6
7.0
1.3
11
28.2
3.7
l.b
8
20.5
13.6
1.0
k
10.3
3.6
.5
0
0
0
0
1
2.6
25.0
.1
Column. 123
Total 15.9
lk2
l8.it
298
38.6
59
7.6
111
lif.it
k
.5
Number of missing observations, 339.
k ,
.5
teduction. Stage
r
Blade Sec. Trim.
6
7
Crtx.Rmvl.
Sp. Pri.
8
Shatter
9
Crtx.Rmvl.
Blade
10
Flake
Core
11
Row
Total
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1.5
^.5
.1
0
0
0
0
0
0
0
0
20m
0
0
0
66
8.5
0
0
0
0
0
0
0
0
0
0
0
0
2
2.7
9.1
.3
0
0
0
0
2
2.7
33.3
.3
8om
0
0
0
73
9.9
0
0
0
0
if
.5
1
2.6
25.0
.1
if
.5
0
0
0
0
1
2.6
^•5
.1
0
0
0
0
2
5.1
33.3
.3
6m
0
0
0
39
5.1
2
.3
22
2.8
1
.1
6
.8
339m
0
772
100.0
104
3. Small non-cortical (secondary) flakes are over-represented at
HPS 20, 23, and 31.
Core preparation is observable in the platform preparation and
direction of flake removal variables, which are also indicative of
attempts to standardize the end product, the flake.
Cores are scarce at most Pinacate sites, but at three sites,
HPS 3, 4 and 66, core tools amount to about 30^ of the samples.
Only
at HPS 4 and 66 are cores with unprepared platforms more frequent than
prepared forms.
Elsewhere prepared platforms, simple cleared cortical
surfaces, predominate. It should be emphasized that platform prepara­
tion here refers simply to the removal of cortex and the clearing of
the impact area prior to flake removal.
No indications of elaborate
grinding techniques or platform faceting are present in the collection.
As Table 7 shows, directionality among prepared platform cores
is evenly distributed between uni- and bidirectional specimens. HPS
40 and 48 produced most of the unidirectional cores, while sites HPS 1
and 20 have more bidirectional work.
Site samples with few core
tools tend to have more prepared than unprepared forms.
Edge shape may reflect use or stylistic variation as a result
of manufacture.
The distribution of this variable shows two edge forms
being common overall:
(30.79$).
convex (29.1$) and undulating, concave-convex
At individual sites, however, one or the other predominates,
such as the increased frequency of concave edges at HPS 40 and of
straight edges at HPS 1 and 2. Greater quantities of convex edges are
found at HPS 23 and 32, while undulating concave-convex edges are
commoner at HPS 66.
The incidence of non-standardized edges
105
Table 7-
Site
No.
Crosstabulation of core or nodule by site. — See Table 2.
Unidirec.
1
Core or Nodule
Uni-prep.
Bi-prep.
6
5
Multi-prep.
0
7
0
32m
0
0
0
0
0
0
Row
Tota]
1
0
0
0
0
3
42.9
4.5
1.4
4
57.1
7.0
1.9
2
0
0
0
0
4
30.8
6.1
1.9
3
23.1
5.3
1.4
6
46.2
18.8
2.8
71m
0
0
0
13
6.0
3
7
21.9
11.5
3.2
10
31.3
15.2
4.6
9
28.1
15.8
4.2
6
18.8
18.8
2.8
68m
0
0
0
32
14.8
4
17
32.7
27.9
7.9
13
25.0
19.7
6.0
12
23.1
21.1
5.6
10
19.2
31.3
4.6
112m
0
0
0
52
24.1
20
3
30.0
4.9
1.4
2
20.0
3.0
.9
5
50.0
8.8
2.3
0
0
0
0
37m
0
0
0
10
4.6
23
1
14.3
1.6
.5
0
0
0
0
1
14.3
1.5
.5
2
40.0
3.0
.9
5
71.4
8.8
2.3
0
0
0
0
42m
0
0
0
7
3.2
2
40.0
3.5
.9
1
20.0
3.1
.5
39m
0
0
0
5
2.3
32
0
0
0
0
3
37.5
4.5
1.4
2
25.0
3.5
.9
3
37.5
9.4
1.4
97m
0
0
0
8
3.7
37
1
14.3
1.6
.5
3
42.9
4.5
1.4
1
14.3
1.8
.5
82m
0
0
0
7
3.2
40
1
8.3
1.6
.9
8
66.7
12.1
3.7
3
25.0
5.3
1.4
2
28.6
6.3
.9
0
0
0
0
94m
0
0
0
12
5.6
31
7
3.2
106
Table 7—Continued,
Site
No.
Unidirec.
1
Core or Nodule
Bi-prep.
Uni-prep.
6
5
Row
Total
Multi-prep.
7
0
48
2
12.5
3.3
.9
8
50.0
12.1
3.7
5
31.3
8.8
2.3
1
6.3
3.1
.5
70M
0
0
0
16
7.4
66
29
67.4
47.5
13.4
7
16.3
1-0.6
3.2
5
11.6
8.8
2.3
2
4.7
6.3
.9
110M
0
0
0
43
19.9
68
0
0
0
0
2
50.0
3.0
.9
1
25.0
1.8
.5
1
25.0
3.1
.5
4lM
0
0
0
4
1.9
61
28.2
66
30.6
57
26.4
32
14.8
895M
0
216
100.0
Column
Total
Number of missing observations, 895
107
represented by the category "other" has frequencies as high as 3*$ at
HPS 3, and 32# at HPS 31.
Dimensions provide information about desirable or standardized
tool sizes. The easiest comparison is simply large versus small tools.
More informative than the simple mean dimension is the 95% confidence
interval within which most individual tool dimensions fall. Figures
8, 9t and 10 present data on tool size and graphically illustrate site
differences.
Three sites have relatively larger tools; HPS
*t8, and 66.
Many other sites fall below the collection's mean dimensions.
In par­
ticular, HPS 68 Sitio La Playa and HPS 23 Chivos Tank produced samples
of relatively thin, smaller flakes.
HPS 20 has a wider range of dimen­
sions than other sites while 32 and 37 have a narrower range of scores.
These dimensional differences may reflect the concentrations of larger
nodular or ejecta found at some sites and the predominance of smaller,
secondary, thinning flakes at others.
Although the major objective during research was to differen­
tiate variable patterns related to manufacture, a descriptive tool
typology was developed and employed during site comparison to further
qualify site differences.
For this purpose the initial 32 types
(because many were insignificantly present, less than 190 had to be
reduced and 13 types were finally included in the site comparison
analysis.
The prevalent tool type among the 13 sites (as well as in
the total population) is the unifacially retouched flake.
This tool
which is unifacially retouched on lateral, transverse or both edges
is a general category.
No attempt was made to define it further as
108
SITE
68
_____
66
46
______
40
37
32
______
31
23
_______
20
4
_____
3
2
I
______
302 4 6 84 0 2 4 6 6 50 2 4 6 6 60 2 4 6 6 70 2 4 6 8 80 2 4 6 6 90 2 4
MILLIMETERS
Figure 8. 9% confidence interval for means of
lengths.
109
SITE
66
66
48
40
37
32
31
23
20
4
3
2
10 2 4 6 6 20 2 4 6 6 30 2 4 6 8 402 4 6 6 SO 2 4 6 6 602 4 6 870 2 4 6
MILLIMETERS
Figure 9. 9% confidence interval for means of
widths.
110
66
66
46
40
37
32
31
23
20
4
3
2
I 2 3 4 5 6 7 6 9 10 I 2 3 4 5 6 7 8 92 0 I 2 3 4 5 6 7 B 9 30 I 2 3
MILLIMETERS
Figure 10. 9% confidence interval for means of
thickness.
Ill
better than two-thirds of flake tools lacked platforms for orienting
the tool for locating retouch modification (Table 8),
Along with the
inter-site differences for unifacially retouched flakes, bifaces,
points, small denticulated and notched tools, unifacially and bi­
facially reduced nodules and ejecta all vary (Table 9)•
Tool quanti­
ties suggest that large unifacially reduced tools are common only at
Sitio Celaya, HPS 66, while unifacial and bifacial reduced pieces
appear in relatively higher frequencies at HPS k and 66. Severely
limited amounts of these types are present at sites HPS 1, 2, 23* 31*
37»
and 68.
Although utilized, unmodified flakes are 11.6& of the 13 site
samples, at eight of the sites they are more common than in general;
HPS 1, 2, 3» ^8, and 68 have high frequencies of utilized flakes.
Likewise utilized cores, though never abundant (8.?&) are found in
higher relative frequencies at HPS 1, 3» and 4, while they are nearly
absent at HPS 37.
points:
Several sites have few or even lack diagnostic
HPS 23» 31« and 32. Seven of the 13 sites have fewer points
and bifaces than might be expected.
Notched, denticulated and bifacially retouched tools have
varied frequencies among sites. HPS 1 and 23 have none of these tools.
Denticulates and notched tools are absent at five of the site samples,
and bifacially retouched flakes are absent at three sites. Several
sites, however, have many more of these specialized, modified tools.
In particular, HPS 32 and
have numerous denticulated and notches
specimens; HPS kO has an increased frequency of bifacially retouched
tools, and HPS 48 has an unexpectedly high frequency of denticulates.
Table 8.
Site
No.
Crosstabulation of flake platform preparation by site. —
See Table 2.
Cortical
1
Platform Preparation
MultiUnifaceted
Dihedral
faceted
2
3
4
5
0
2
9.1
5.3
.5
0
0
0
0
2
9.1
4.0
.5
17m
0
0
0
22
5.5
9
19.6
18.0
2.3
38m
0
0
0
46
11.5
1
2.9
2.6
.2
6
17.6
12.0
1.5
66m
0
0
0
34
8.5
9
16.1
23.7
2.3
6
10.7
12.0
1.5
108m
o
o
o
56
14.0
0
0
0
0
2
20.0
4.0
.5
4
33.3
8.0
1.0
37m
0
0
0
10
2.5
37m
0
0
0
12
3.0
31m
0
0
0
13
3.3
73m
0
0
0
32
8.0
54m
0
0
0
35
8.8
1
9
40.9
7.0
2.3
9
40.9
5.5
2.3
0
0
0
0
2
14
30.4
10.9
3.5
14
41.2
10.9
3.5
22
39.3
17.2
5.5
20
43.5
12.2
5.0
3
6.5
15.0
.7
0
0
0
0
20
2
20.0
1.6
.5
6
6o.o
3.7
1.5
3
5.4
15.0
.7
0
0
0
0
23
2
16.7
1.6
.5
6
50.0
3.7
1.5
0
0
0
0
0
0
0
0
31
5
38.5
3.9
1.2
5
38.5
3.0
1.2
0
0
0
0
1
7.7
2.6
.2
32
7
21.9
5.5
1.7
14
43.8
8.5
3.5
4
12.5
20.0
1.0
3
9.4
7.9
.7
2
15.4
4.0
.5
4
12.5
8.0
1.0
37
9
25.7
7.0
2.3
15
42.9
9.1
3.7
2
5-7
10.0
.5
5
14.3
13.2
1.2
4
11.4
8.0
1.0
3
4
13
38.2
7.9
3.3
16
28.6
9.8
4.0
Row
Tota]
Planar
113
Table 8--Continued.
Site
No.
Cortical
1
Platform Preparation
TJniMultiDihedral
faceted
faceted
4
2
3
Row
Total
Planar
5
0
40
6
15.8
4.7
1.5
19
50.0
11.6
4.7
2
5.3
10.0
.5
6
15.8
15.8
1.5
5
13.2
10.0
1.2
68m
0
0
0
38
9.5
48
13
32.5
10.2
3.3
20
50.0
12.2
5.0
2
5.0
10.0
.5
4
10.0
10.5
1.0
1
2.5
2.0
.2
46m
0
0
0
4o
10.0
66
19
48.7
14.8
4.7
13
33.3
7.9
3.3
2
5.1
10.0
.5
2
5.1
5.3
.5
3
7.7
6.0
.7
114M
0
0
0
39
9.7
68
6
26.1
4.7
1.5
8
34.8
4.9
2.0
2
8.7
10.0
.5
5
21.7
13.2
1.2
2
8.7
4.0
.5
22M
0
0
0
23
5.7
Column, 128
Total 32.0
164
4l.o
20
5.0
38
9.5
50
12.5
711M
0
400
100.0
Number of missing observations, 711.
Table 9. Crosstabulation of tool types by site. — See Table 2
Tool Type
Site
No.
UniReduced
BiReduced
UniRetouch
Used
Flake
1
2
3
7
8
Core
Point
Biface
Denticu'
late
9
10
14
1
0
0
0
0
2
5.1
1.7
.2
19
48.7
5.1
1.7
12
30.8
9.3
1.1
5
12.8
5.2
.5
0
0
0
0
1
2.6
1.0
.1
2
2
2.4
1.8
.2
5
6.0
4.3
.5
49
58.3
13.1
4.4
15
17.9
11.6
1.4
5
6.0
5.2
.5
1
1.2
1.5
.1
1
1.2
1.0
.1
3
3.6
11.5
.3
3
8
8.0
7.3
.7
13
13.0
11.1
1.2
28
28.0
7.5
2.5
17
17.0
13.2
1.5
18
18.0
18.6
1.6
3
3.0
4.6
.3
6
6.0
6.3
.5
0
0
0
0
4
19
11.7
17.4
1.7
31
19.0
26.5
2.8
47
28.8
12.6
4.2
14
8.6
10.9
1.3
24
14.7
24.7
2.2
11
6.7
16.9
1.0
9
5.5
9.4
.8
2
1.2
7.7
.2
20
5
10.6
4.6
.5
4
8.5
3.4
.4
11
23.4
2.9
1.0
6
12.8
4.7
.5
2
4.3
2.1
.2
8
17.0
12.3
.7
6
12.8
6.3
.5
2
4.3
7.7
.2
23
0
0
0
0
3
6.1
2.6
.3
14
28.6
3.8
1.3
6
12.2
4.7
.5
4
8.2
4.1
.4
7
14.3
10.8
.6
14
28.6
14.6
1.3
1
2.0
3.8
.1
31
0
0
0
0
3
6.8
2.6
.3
14
31.8
3.8
1.3
2
4.5
1.6
.2
2
4.5
2.1
.2
7
15.9
10.8
.6
11
25.0
11.5
1.0
0
0
0
0
32
3
2.9
2.8
.3
10
9.5
8.5
.9
43
41.0
11.5
3.9
3
2.9
2.3
.3
5
4.8
5.2
.9
14
13.3
21.5
1.3
13
12.4
13.5
1.2
0
0
0
0
37
5
5.6
4.6
.5
4
6.7
5.1
.5
32
36.0
8.6
2.9
10
11.2
7.8
.9
1
1.1
1.0
.1
7
7.9
10.8
.6
10
11.2
10.4
.9
4
4.5
15.4
.4
4o
5
4.7
4.6
.5
9
8.5
7.7
.8
48
45.3
12.9
4.3
4
3.8
3.1
.4
7
4.6
7.2
.6
1
.9
1.5
.1
4
3.8
4.2
.4
3
2.8
11.5
.3
0
0
0
0 .
114
e Table 2.
ol Type
Biface
10
1
2.6
1.0
.1
Denticulate
14
0
0
0
0 .
Bifacial
Retouch
16
Notched
17
Uni Retouch
Notched
18
Eccentric
27
Row
Total
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0M
0
0
0
39
3.5
1
1.2
1.0
.1
3
3.6
11.5
.3
3
3.6
7.0
.3
0
0
0
0
0
0
0
0
0
0
0
0
OM
0
0
0
84
7.6
6
6.0
6.3
•5
0
0
0
0
1
1.0
2.3
.1
3
3.0
10.3
.3
3
3.0
12.5
.3
0
0
0
0
OM
0
0
0
100
9.0
9
5.5
9.4
.8
2
1.2
7.7
.2
0
0
0
0
4
2.5
13.8
.4
2
1.2
8.3
.2
0
0
0
0
1M
0
0
0
163
14.7
6
12.8
6.3
-5
2
4.3
7.7
.2
2
4.3
4.7
.2
0
0
0
0
1
2.1
4.2
.1
0
0
0
0
OM
0
0
0
47
4.2
14
28.6
14.6
1.3
1
2.0
3.8
.1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
OM
0
0
0
49
4.4
11
25-0
11.5
1.0
0
0
0
0
4
9.1
9.3
.4
1
2.3
3.^
.1
0
0
0
0
0
0
0
0
OM
0
0
0
44
4.0
13
12.4
13.5
1.2
0
0
0
0
13
12.4
30.2
1.2
0
0
0
0
1
1.0
4.2
.1
0
0
0
0
OM
0
0
0
105
9.5
10
11.2
10.4
.9
4
4.5
15.4
.4
5
5.6
11.6
.5
3
3.4
.3
2
2.2
100.0
.2
CM
0
0
0
89
8.0
10.3
4
4.5
16.7
.4
4
3.8
4.2
.4
3
2.8
11.5
.3
7
6.6
16.3
.6
10
9.4
34.5
.9
8
7.5
13.3
.7
0
0
0
0
OM
0
0
0
106
9.5
Table 9—Continued.
Site
No.
W
66
68
Column
Total
UniReduced
BiReduced
UniRetouch
Used
Flake
1
2
3
7
Core
8
Point
Biface
Denticu­
late
9
10
Ik
1.0
28
32.6
7.5
2.5
13
15.1
10.1
1.2
7
8.1
7.2
.6
0
0
0
0
9
10.5
9.^
.8
6
7.0
23.1
.5
56
36.6
51.^
5.0
18
11.8
15A
1.6
23
15.0
6.2
2.1
17
11.1
13.2
1.5
Ik
9.2
Ik.k
1.3
if
2.6
6.2
A
10
6.5
lO.k
.9
k
2.6
15»k
A
0
0
0
0
2
10
22.2
7.8
.9
3
6.7
3.1
.3
2
k.k
3.1
.2
2
1.7
.2
17
37.8
k.6
1.5
1
2.2
3.8
117
10.5
373
33.6
129
11.6
97
8.7
65
5.9
6
7.0
5.5
.5
11
12.8
9A
109
9.8
Number of missing observations» 1.
if.if
2.1
.2
96
8.6
.1
26
2.3
Bi:
Re-
"
t
*
j
I!
I!
115
oint
9
Biface
10
Denticu­
late
14
Bifacial
Retouch
16
Notched
17
6
7.0
23.1
.5
1
1.2
2.3
.1
2
2.3
6.9
.2
3
3.5
12.5
.3
0
0
0
0
OM
0
0
0
86
7.7
2.6
15.4
.4
2
1.3
4.7
.2
4
2.6
13.8
.4
1
.7
4.2
.1
0
0
0
0
OM
0
0
0
153
13.8
5
11.1
11.6
.5
2
4.4
6.9
.2
1
2.2
4.2
.1
0
0
0
0
OM
0
0
0
45
4.1
43
3.9
29
2.6
24
2.2
2
.2
1M
0
1110
100.0
0
0
0
0
9
10.5
9.4
.8
4
2.6
6.2
.4
10
6.5
10.4
2
if.it
3.1
.2
2
2.1
.2
1
2.2
3-8
.1
65
5.9
96
8.6
26
2.3
.9
Uni Retouch
Notched
18
Eccentric
27
Row
Total
0
116
Generally, sites HPS 4 and 66 and, to a lesser extent, 40 and
48 contributed relatively more unifacially and bifacially percussionreduced tools.
These reduced and otherwise unmodified tools are un­
common at other sites.
Specialized small tools also add variety to the
samples from HPS 37, 40, and 48.
At HPS 1, 2, and 3 utilized flakes
combine with higher frequencies of unifacially retouched flakes to
distinguish these sites.
Point making, as illustrated by the presence
of both bifaces and finished projectiles, differentiates HPS 23, 31,
and 32 from the other sites.
Three final variables show inter-site variation:
(1) location
of use damage; (2) the angle of the immediate edge (edge angle); and
(3) the angle 5 mm along the face (rake angle).
Use evidence, which
may be primarily observed with minor exceptions, on outer, outer
lateral, and outer transverse edges, reproduced at individual sites
the general collection pattern.
Table 10 summarizes site comparisons.
Evidence of inner (bulbar) use, whether lateral or transverse, never
rises above 9»3& (being highest at HPS 48); however, use of planar
platform surfaces is well defined at four sites:
HPS 1, 2, 3, and 4.
Additionally, use of all edges is commonest at four sites:
HPS 20,
23, 31, and 32.
Figures 11 and 12 record means and standard deviations for
rake and work angles at the 13 sites. The rake angle measurements are
narrower at seven sites (1, 23, 31, 32, 37, 40, and 48), wider at four
sites (2, 3, 4, and 20) and are very narrow at HPS 68. The standard
deviation at HPS 1, 3, and 68 suggest that there are greater ranges of
mean scores at some sites, while other samples have less deviation from
Table 10. Crosstabulation of use location by site
Site
No.
Planar
Core
1
Inner
3
Outer
4
Outer
Inner
Trans.
Both
8
7
Inner
Lateral
9
See Table 2
Use Location
Outer
Outer
Lateral Trans.
10
11
In-Out
Lat.
12
1
5
13.5
10.0
.5
0
0
0
. 0
6
16.2
3.2
.5
1
2.7
7.7
.1
6
16.2
4.0
. .5
2
5.4
4.1
.2
8
21.6
5.1
.7
4
10.8
8.0
.4
2
5.4
2.5
.2
2
7
8.6
14.0
.6
1
1.2
5.9
.1
7
8.6
3.7
.6
0
0
0
0
10
12.3
6.7
.9
4
4.9
8.2
.4
21
25.9
13.3
1.9
5
6.2
10.0
.5
6
7.4
7.4
.5
3
14
14.3
28.0
1.3
1
1.0
5.9
.1
8
8.2
4.2
.7
0
0
0
0
15
15.3
10.0
1.4
1
1.0
2.0
.1
19
19.4
12.0
1.7
5
5.1
10.0
.5
3
3.1
3.7
.3
if
14
8.7
28.0
1.3
3
1.9
17.6
.3
15
9.3
7.9
1.4
3
1.9
23.1
.3
19
11.8
12.7
1.7
1
.6
2.0
.1
19
11.8
12.0
1.7
6
3.7
12.0
.5
11
6.8
13.6
1.0
20
1
2.1
2.0
.1
1
2.1
5.9
.1
3
6.4
1.6
.3
0
0
0
0
8
17.0
5.3
.7
1
2.1
2.0
.1
7
14.9
4.4
.6
0
0
0
0
5
10.6
6.2
.5
23
1
2.0
2.0
.1
0
0
0
0
0
0
0
0
0
0
0
0
5
10.2
3.3
.5
1
2.0
2.0
.1
6
12.2
3.8
.5
3
6.1
6.0
.3
9
18.4
11.1
.8
31
0
0
0
0
0
0
0
0
0
0
0
0
1
2.7
7.7
.1
4
9.8
2.7
.4
3
7.3
6.1
.3
3
7.3
1.9
.3
1
2.4
' 2.0
.1
3
7.3
3.7
.3
32
1
1.0
2.0
.1
1
1.0
5.9
.1
14
13.3
7.4
1.3
0
0
0
0
23
21.9
15.3
2.1
2
1.9
4.1
.2
10
9.5
6.3
.9
6
5.7
12.0
.5
7
6.7
8.6
.6
37
1
1.1
2.0
.1
1
1.1
5.9
.1
13
14.6
6.9
1.2
3
3.4
23.1
.3
9
9.0
5.3
.7
6
6.7
12.2
.5
20
22.5
12.7
1.8
2
2.2
4.0
.2
11
12.4
13.6
1.0
40
1
.9
2.0
.1
4
3.8
23.5
.4
27
25.5
14.3
2.5
2
1.9
15.4
.2
21
19.8
14.0
1.9
6
5.7
12.2
.5
17
16.0
10.8
1.5
2
1.9
4.0
.2
5
4.7
6.2
.5
117
3ee Table 2.
Jse Location
ter
Outer
sral Trans.
3
11
In-Out
Lat.
12
Every­
thing
13
Inner
All
14
Inner
Outer
15
Plat.
Deliberate
16
Edge
23
Row
Total
0
8
.6
.1
.7
4
10.8
8.0
.4
2
5.4
2.5
.2
1
2.7
.6
.1
1
2.7
5.0
.1
0
0
0
0
0
0
0
0
1
2.7
9.1
.1
2M
0
0
0
37
3.4
21
.9
.3
.9
5
6.2
10.0
.5
6
7.4
7.4
.5
4
4.9
2.4
.4
1
1.2
5.0
.1
6
7.4
6.3
.5
4
4.9
8.2
.4
5
6.2
45.5
.5
3M
0
0
0
81
7.4
L9
.0
.7
5
5.1
10.0
.5
3
3.1
3.7
.3
11
11.2
6.6
1.0
0
0
0
0
13
13.3
13.7
1.2
6
6.1
12.2
.5
2
2.0
18.2
.2
2M
0
0
0
L9
.8
.0
.7
6
3.7
12.0
.5
11
6.8
13.6
1.0
23
14.3
13.9
2.1
1
.6
5.0
.1
32
19.9
33.7
2.9
13
8.1
26.5
1.2
1
.6
9.1
.1
3M
0
0
0
161
14.7
7
.9
5
10.6
6.2
.5
16
34.0
9.6
1.5
0
0
0
0
5
10.6
5.3
.5
0
0
0
0
0
0
0
0
CM
0
0
0
47
4.3
.6
0
0
0
0
6
.2
.8
.5
3
6.1
6.0
.3
9
18.4
11.1
.8
18
36.7
10.8
1.6
1
3
6.1
6.1
.3
0
0
0
0
OM
0
0
0
49
4.5
5.0
.1
2
4.1
2.1
.2
3
.3
.9
.3
1
2.4
2.0
.1
3
7.3
3.7
.3
20
48.8
12.0
1.8
0
0
0
0
4
9.8
4.2
.4
2
4.9
4.1
.2
0
0
0
0
3M
0
0
0
41
3.7
LO
.5
.3
.9
6
5.7
12.0
.5
7
6.7
8.6
.6
26
1
1.0
11
10.5
11.6
1.0
3
2.9
6.1
.3
0
0
0
0
CM
0
0
0
105
24.8
15.7
2.4
20
.5
.7
.8
2
2.2
4.0
.2
11
12.4
13.6
3.0
17
19.1
10.2
1.5
1
1.1
1
1.1
2.0
.1
0
0
0
0
OM
0
0
0
89
8.1
.1
5
5.6
5.3
.5
L7
.0
2
1.9
4.0
.2
5
4.7
6.2
5
4.7
3.0
.5
5
4.7
25.0
.5
5
4.7
5.3
.5
6
0
0
0
0
OM
0
0
0
106
9.7
.8
.5
.5
2.0
5.0
.1
5.0
5.7
12.2
.5
98
8.9
4
9.6
Table 10—Continued
Site
No.
Planar
Core
1
Inner
3
Outer
4
Outer
Inner
Both
Trans.
8
7
Inner
Lateral
9
UBe Location
Outer
Outer
Lateral Trans.
10
11
In-Out
Lat.
12
48
1
1.2
2.0
.1
0
0
0
0
Ik
16.3
7.4
1.3
2
2.3
15.4
.2
15
17.4
10.0
1.4
8
9.3
16.3
.7
10
11.6
6.3
.9
6
7.0
12.0
.5
7
8.1
8.6
.6
66
4
2.6
8.0
.4
If
2.6
23.5
.4
75
49.0
39.7
6.8
1
.7
7.7
.1
10
6.5
6.7
.9
13
8.5
26.5
1.2
10
6.5
6.3
.9
6
3.9
12.0
.5
5
3.3
6.2
.5
68
0
0
0
0
1
2.2
5.9
.1
7
15.6
3.7
.6
0
0
0
0
6
13.3
4.0
.5
1
2.2
2.0
.1
8
17.8
5.1
.7
4
8.9
8.0
.4
7
15.6
8.6
.6
Column 50
Total 4.6
17
1.5
189
17.2
13
1.2
150
13.7
49
k.5
158
14.4
50
4.6
81
7.4
Number of missing observations, 13•
118
Be
Location
Outer
In-Out
Trans.
Lat.
11
12
Every­
thing
13
Inner
All
Ik
Inner
Outer
15
Plat.
Deliberate
16
Edge
23
Row
Total
0
6
7.0
12.0
.5
7
8.1
8.6
.6
10
11.6
6.0
.9
2
2.3
10.0
.2
if
if.7
if.2
.if
6
7.0
12.2
•5
1
1.2
9.1
.1
OM
0
0
0
86
7.8
6
3.9
12.0
.5
5
3.3
6.2
.5
17
7.8
7.2
1.1
5
3.3
15.0
.5
5
3.3
5.3
.5
2
1.3
if.l
.2
1
.7
9.1
.1
OM
0
0
0
153
13.9
k
8.9
8.0
A
7
15.6
8.6
.6
3
6.7
1.8
.3
2
if.lt
10.0
.2
3
6.7
3.2
.3
3
6.7
6.1
.3
0
0
0
0
OM
0
0
0
^5
50
Jf.6
81
7.*+
166
15.1
20
1.8
95
8.7
^9
^.5
11
1.0
13M
0
1098
100.0
119
SITE
68
66
46
40
37
32
31
S3
20
4 56 7 6 9S O ! 2 3 4 5 6 7 8 9 6 0 1 2 3 4 5 6 7 8 97 0 I 2 3 4 5 6 7 8
DEGREES
Figure 11. 9% confidence interval for reans of
rake angle.
120
SITE
68
66
_______
4 8
40
_______
37
32
______
31
23
20
______
A
3
2
I
_______
—-_____
____________
6 0 I 2 3 4 5 6 7 8 97 0 1 2 3 4 5 6 7 8 9 8 0 1 2 3
DEGREES
Figure 12. 9556 confidence interval for means of
work angle.
121
the mean than in the whole tool population (standard deviation 14.9).
The individually higher or lower means at each site appear attributable
to greater ranges around the central score.
Work angles are often wide at individual sites, but, neverthe­
less, seven of 13 have mean scores lower than the total collections.
Relatively smaller angles are observable at HPS 1, 2, 20, 23, 31, 32,
and 68 with the smallest mean work angle being found at HPS 23«
angles are observable at HPS 3i
^0, ^8, and 66.
Wider
Wider angle measure­
ment at HPS 3 and 4 appear to indicate greater use of planar core
tools.
Standard deviations suggest broad ranges around the means.
Similarly rake angles produce the same pattern, being generally between
5 and 7° narrower than work angles; however, it is notable that there
is not a standard increase in degrees among the two.
Instead, other
factors like retouching or excessive use seem to influence the differ­
ing angle widths, as does, possibly, the flake size and width.
Still,
where narrow work angles are found, narrow rake angles also occur
(Fig. 12).
The frequency of different forms of surface alteration vary
from site to site (Table 11). Seventy percent of all tools are surface
altered; sample frequencies range from as high as 96.3$ (HPS 66) to
less than 50# at HPS 1 and 23.
Although too many forms of surface
alteration variation are present to specifically cite each one, the
tables choose comparative data on the major ones.
Notable is the
paucity or absence of "hydrated" obsidian at HPS 1, 2, 3,
and ^8,
compared with its high frequency at HPS 23, 31i and 32. Green lichen
is strongly present only at HPS 2, 3» and 20.
Groundstaining combined
Table 11.
Site
No.
Crosstabulation of surface alteration by site. ~ See
Table 2.
Groundstaining
2
Caliche
3
Oxidized
5
Hydrated
6
Surface Alteration
Varnish
Grstain. Var.Grstain. Gr
Oxidized Oxidized
Oxidized
Li
8
10
7
1
0
0
0
0
1
5.0
3.0
.1
8
40.0
3.0
1.0
0
0
0
0
2
10.0
10.5
.2
5
25.0
4.4
.6
2
10.0
2.2
.2
2
0
0
0
0
2
3.4
6.1
.2
13
22.0
4.9
1.5
2
3.4
2.3
.2
3
5.1
15.8
.4
3
5.1
2.6
.4
3
5.1
3.2
.4
5
4
0
0
0
0
2
2.6
6.1
.2
34
44.7
12.8
4.1
3
3-9
3.4
.4
3
3.9
15.8
.4
4
5.3
3.5
.5
3
3.9
3.2
.2
1
1
1
.8
5.0
.1
7
5.4
21.2
.8
70
53.8
26.3
8.3
5
3.8
5.7
.6
2
1.5
.2
8
6.2
7.0
1.0
0
0
0
0
0
0
0
0
2
6.7
6.1
.2
7
23.3
2.6
.8
6
20.0
6.8
.7
1
3.3
5.3
.1
0
0
0
0
0
0
0
0
23
2
9.1
10.0
.2
0
0
0
0
5
22.7
1.9
.6
12
54.5
13.6
1.4
0
0
0
0
0
0
0
0
0
0
0
0
31
0
0
0
0
1
3.6
3.0
.1
15
53.6
5.6
1.8
9
32.1
10.2
1.1
0
0
0
0
'0
0
0
0
0
0
0
0
32
2
2.9
10.0
.2
6
8.6
18.2
.7
21
30.0
7.9
2.5
24
34.3
27.3
2.9
l
1.4
5.3
.1
4
5.7
3.5
.5
1
1.4
1.1
.1
37
2
4.0
10.0
.2
if
8.0
12.1
.5
26
52.0
9.8
3.1
5
10.0
5.7
.6
0
0
0
0
7
14.0
6.1
.8
1
2.0
1.1
.1
3
4
20
10.5
1
2
122
>ee
m.
:ed
Var.Grstain. Green
Oxidized
Lichen
10
11
Oxidized
Caliche
13
Hematite
14
Varn.Cal.
Oxidized
16
Grstain.
Cal. Oxi.
21
Row
Total
0
5
)
f
5
2
10.0
2.2
.2
0
0
0
0
0
0
0
0
2
10.0
16.7
.2
0
0
0
0
0
0
0
0
19M
0
0
0
20
2.4
J
3
5.1
3.2
.4
32
54.2
44.4
3.8
0
0
0
0
1
1.7
8.3
.1
0
0
0
0
0
0
0
0
25M
0
0
0
59
7.0
3
3.9
3.2
.2
10
13.2
13.0
1.2
16
21.1
18.2
1.9
1
1.3
8.3
.1
0
0
0
0
0
0
0
0
24M
0
0
0
76
9.1
0
0
0
0
8
6.2
11.1
1.0
28
21.5
31.8
3.3
1
.8
8.3
.1
0
0
0
0
0
0
0
0
34m
130
15.5
0
0
0
0
6
20.0
8.3
.7
6
20.0
6.8
.7
1
3.3
8.3
.1
1
3.3
5.9
.1
0
0
0
0
17M
0
0
0
30
3.6
0
0
0
0
0
0
0
0
2
9.1
2.3
.2
1
0
0
0
0
27M
0
0
0
22
2.6
8.3
.1
0
0
0
0
0
0
0
0
1
3.6
1.4
.1
2
7.1
2.3
.2
0
0
0
0
0
0
0
0
0
0
0
0
16M
0
0
0
28
3.3
1
1.4
1.1
.1
3
4.3
4.2
A
2
2.9
2.3
.2
4
5.7
33.3
.5
1
1.4
5.9
.1
1
1.4
5.9
.1
35M
0
0
0
70
8.3
1
2.0
1.1
.1
1
2.0
1.4
.1
3
6.0
3.4
A
1
2.0
8.3
.1
0
0
0
0
0
0
0
0
39M
0
0
0
50
6.0
l
5
ih
5
5
>
i
•
1
1
i
i
4.5
0
0
0
Table 11—Continued
Site
No.
Surface Alteration
Varnish
Grstain. Var.Grstain.
Oxidized
Oxidized Oxidized
8
10
7
Groundstaining
2
Caliche
3
Oxidized
5
Hydrated
6
4o
if
4.2
20.0
.5
1
1.0
3.0
.1
23
24.0
8.6
2.7
6
6.3
6.8
.7
4
4.2
21.1
.5
16
16.7
14.0
1.9
16
16.7
17.2
1.9
48
3
3.6
15.0
.4
1
1.2
3.0
.1
12
14.5
4.5
1.4
1
1.2
1.1
.1
1
1.2
5.3
.1
40
48.2
35.1
4.8
13
15.7
14.0
1.5
66
3
2.1
15.0
.4
1
.7
3.0
.1
21
14.6
7.9
2.5
8
5.6
9.1
1.0
2
1.4
10.5
.2
26
18.1
22.8
3.1
54
37.5
58.1
6.4
68
3
9.7
15.0
.4
5
16.1
15.2
.6
11
35.5
4.1
1.3
7
22.6
8.0
.8
0
0
0
0
1
3.2
.9
.1
.0
0
0
0
20
2.4
33
3.9
266
31.7
88
10.5
19
2.3
114
13.6
93
11.1
Column
Total
Number of missing observations, 272.
123
Alteration
Jrstain. Var.Grstain.
Jxidized
Oxidized
8
10
Green
Lichen
11
Oxidized
Caliche
Hematite
Varn.Cal. Grstain.
Oxidized Cal. Oxi.
Row
Total
13
14
16
21
0
3
3.1
17.6
.4
10m
0
0
0
96
11.4
6
7.2
35.3
.7
3m
0
0
0
83
9.9
16
16.7
14.0
1.9
16
16.7
17.2
1.9
0
0
0
0
17
17.7
19.3
2.0
0
0
0
0
40
48.2
35.1
4.8
13
15.7
14.0
1.5
0
0
0
0
6
7.2
6.8
.7
0
0
0
0
6
6.3
35.3
.7
0
0
0
0
26
18.1
22.8
3.1
54
37.5
58.1
6.4
8
5.6
ll.l
1.0
5
3.5
5.7
.6
0
0
0
0
9
6.3
52.9
1.1
7
4.9
41.2
.8
9m
0
0
0
144
17.2
1
3.2
.9
.1
0
0
0
0
3
9.7
4.2
.4
1
3.2
1.1
.1
0
0
0
0
0
0
0
0
0
0
0
0
14m
0
0
0
31
3.7
114
13.6
93
11.1
72
8.6
88
10.5
12
1.4
17
2.0
17
2.0
272m
839
0 100.0
124
with oxidation, although minimally present at most sites, is noticeably
present at HPS 40, 48, and 66. Finally, desert varnish is uncommon at
all sites except at HPS 40, 48, and 66.
Most alteration may be ranked
as light, but three sites have increased frequencies of moderate
alteration (HPS 40, 48, and 66), while HPS 48 and HPS 66 both have
higher relative frequencies of heavy alteration (Table 12).
It is difficult to compare variable associations among sites.
Neither time, funding, or SPSS sub-programs permitted additional tests
for co-occurrence of nominal variables beyond two variable contingency
tables.
Observations on what attributes appear jointly rest on in­
specting each site's tables and then comparing sites.
were used to compare sites.
Two approaches
Variable co-occurrence percentages were
compared among sites to establish which attributes often appeared to­
gether.
Second, sites were paired and their cumulative relative fre­
quencies compared. By examining the site contingency tables* attribute
associations some generalizations may be made.
The first major difference among sites is in the categories of
material selected for tool manufacture.
At four sites (1, 2, 48, and
68) low absolute and relative frequencies and absence of some materials
altogether sire documented.
At three sites basalt flakes were chosen
for tool-making in increased quantities:
48.
HPS 4, 40, and particularly
A greater relative frequency of basalt core and ejectum tools is
also found at these sites as well as at HPS 66 where core and ejectum
tools combined are more common than flakes alone.
Numerous obsidian
flakes are found at HPS 23» 31, and 32. HPS 37 has relatively higher
quantities of chert and rhyolite flakes, while HPS 2 is distinctive
125
Table 12. Crosstabulation of amount of alteration by site. — See
Table 2.
Amount of Alteration
Hvy. Ox- Lt. Ox- Hvy. GsHeavy Partial Lt.Varn. Mod.Cal. Lt. Ox.
Lipht
Moder­
ate
1
2
3
1
6
33.3
1.3
.7
9
50.0
4.6
1.1
1
5.6
2.1
.1
0
0
0
0
1
5.6
4.2
.1
2
17
29.8
3-6
2.0
6
10.5
3.1
.7
1
1.8
2.1
.1
32
56.1
65.3
3.8
3
51
67.1
10.8
6.1
19
25.0
9.8
2.3
2
2.6
4.3
.2
96
73.8
20.3
11.5
25
19.2
12.9
3.0
20
21
72.4
4.4
2.5
23
Site
No.
Row
Total
8
0
0
0
0
0
1
5.6
5.6
.1
21m
0
0
0
1
1.8
4.2
.1
0
0
0
0
0
0
0
0
27m
57
0 6.8
0
0
0
0
0
0
0
0
0
0
0
0
0
0
24m
76
0 9.1
0
0
1
.8
2.1
.1
4
5.3
8.2
.5
4
3.1
8.2
.5
0
0
0
0
4
3.1
13.8
.5
0
0
0
0
34m 130
0 15.6
0
0
4
13.8
2.1
.5
1
3.4
2.1
.1
1
3.4
2.0
.1
0
0
0
0
2
6.9
6.9
.2
0
0
0
0
18m
29
0 3.5
0
0
18
81.8
3.8
2.2
1
4.5
.5
.1
2
9.1
4.3
.2
0
0
0
0
0
0
0
0
1
4.5
3.4
.1
0
0
0
0
27m
0
0
0
22
2.6
31
23
82.1
4.9
2.8
3
10.7
1.5
.4
0
0
0
0
1
3.6
2.0
.1
0
0
0
0
1
3.6
3.4
.1
0
0
0
0
16m
0
0
0
28
3.4
32
50
71.4
10.6
6.0
15
21.4
7.7
1.8
3
4.3
6.4
.4
2
2.9
4.1
.2
0
0
0
0
0
0
0
0
0
0
0
0
35m
0
0
0
70
8.4
37
37
74.0
7.8
4.4
9
18.0
4.6
1.1
1
2.0
2.1
.1
2
4.0
4.1
.2
0
0
0
0
1
2.0
3.4
.1
0
0
0
0
50
39m
0 6.0
0
0
40
47
49.0
9.9
5.6
33
34.4
17.0
4.0
5
5.2
10.6
.6
2
2.1
4.1
.2
3
3.1
12.5
.4
6
6.3
20.7
.7
0
0
0
0
10m
96
0 11.5
0
0
4
5
6
18
2.2
126
Table 12—Continued.
Site
No.
Amount of Alteration
Hvy. Ox- Lt. Ox- Hvy. GsPartial Lt.Varn. Mod.Cal. Lt. Ox.
4
8
6
0
5
Light
1
Moderate
Heavy
2
3
48
29
3^.9
6.1
3.5
25
30.1
12.9
3.0
7
8.4
14.9
.8
0
0
0
0
11
13.3
45.8
1.3
5
6.0
17.2
.6
6
7.2
33.3
.7
3M
83
0 10.0
0
0
66
50
34.7
10.6
6.0
^3
29.9
22.2
5.2
23
16.0
48.9
2.8
l
.7
2.0
.1
8
5.6
33.3
1.0
9
6.3
31.0
1.1
10
6.9
55.6
1.2
9M 144
0 17.3
0
0
68
28
90.3
5.9
3.4
2
6.5
1.0
.2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
3.2
5.6
.1
Column 473
Total 56.7
194
23.3
47
5.6
49
5.9
24
2.9
29
3.5
18
2.2
Number of missing observations, 277•
14M
0
0
0
Row
Total
31
3.7
277M 834
0 100.0
127
because more quartz flakes are found in its sample. Basalt is the
prevalent material used in making cores (which become tools) and ejecta
tools; however, flake materials vary greatly.
When co-occurrence of
flake reduction stage and stone type is charted, some materials are
found only in final reduction stages while others seldom are. Flake
manufacture stages of particular materials are only found at one or two
sites.
Initial reduction flakes of basalt (cortex and cortex removal
flakes), for example, are present in increased frequencies at HPS 48,
while large non-cortical quartz (primary) flakes are most common at
HPS 2 and small obsidian non-cortical (secondary) flakes characterize
HPS 23 and 32.
Inter-site differences are recognizable in the application of
thinning and edge modification techniques indicated in the CROSSTAB
percussion reduction by retouch. Strong attribute associations are
present at five sites, while expected (random) variation is present at
three.
When distinct patterns are found three differences appear:
(1)
the absence of bifacial work (HPS 1); (2) the predominance of bifacial
percussion flaking and retouch (HPS 23); or (3) a co-occurrence of
unifacial percussion flaking, and retouch with bifacial reduction and
retouch (HPS 31* 32, and 66).
Absolute and relative frequencies suggest that the flakes
selected for tool use vary from site to site.
Initial, cortex-
retaining flakes are common at several sites; large, non-cortical
flakes predominate at others; and small non-cortical flakes predominate
at still others.
Some of this distinctness results from retouch modi­
fication. Flake reduction stage-retouch relationships deviate from
128
expected frequencies at several sites:
strongly at HPS 4, 31, 32, and
*f8; and less strongly at HPS 3» 23, and 66. Major contrasts are found
between unifacial retouching of cortex and cortex reraoveil flakes at
HPS k and ^8 (this is overwhelmingly hard-hammer percussion work,
possibly for resharpening) and the bifacial retouching of small noncortical secondary flakes at HPS 23, 31» and 32 (during manufacture).
Aside from these two patterns of material and category selec­
tion and of percussion reduction and retouch, only surface alterations
differ markedly among sites.
When the tool category is compared with
the form of surface alteration "empty cells" appear in the contingency
tables.
'
For example, varnished flakes, cores or nodule/ejectum are
absent from four sites, HPS 23, 31, 37, and 68, while hydrated obsidian
is not present at HPS 1. More subtle are the varying frequencies of
oxidized flakes (as high as 63& at HPS 32, but as low as 17$ at HPS 2).
Some sites have fairly uniform types of alteration, such as numerous
green lichen-covered quartz flakes (HPS 2) or varnished ejecta (HPS 66),
*
yet other sites show great variety both in the tool categories and in
the type and intensity of surface alteration.
Although only three
contingency tables, HPS 31i 32, and 66, indicate strong deviation from
expected frequencies, the complete absence of alteration forms on
certain categories at individual sites is important negative data.
To group sites having similar attribute frequencies, Robinson's
(1951:297) "coefficient of agreement" was employed.
Using attribute
frequency, percentages were compared between sites.
The lack of agree­
ment is simply the difference between percentages.
Taking attributes
129
characteristic of traditions,
material selected and methods of manu­
facture, a cumulative difference was calculated by subtracting the
coefficient of agreement from 200 (paired sites total percentage).
Table 13 records the data on nine paired sites and arrays them accord­
ing to greatest similarity, of material, category, reduction, retouch
and manufacture stage.
These attribute comparisons suggest that sites
3 and *+, 31 and 32, and 23 and 31 are similar with their cumulative
differences being small.
In contrast, the paired sites ^0 and ^8, 37
and 68, and ^8 and 66 show divergence, particularly in materials
selected and manufacture stage.
This contributes to a general lack
of agreement among the latter set of sites' collections.
Variable frequencies and co-occurrences describing attribute
associations differ among sites and reveal at least four distinct
patterns of tool-making with the possibility of several others. The
data suggest mixed components at four sites, single components at only
two, and at least two distinct traditions and time periods at several
other sites.
HPS 2's flake, core and nodule industry in quartz with its
green lichen adherent is markedly different from other sites.
This
tool association, though employing techniques similar to sites with
lightly to moderately oxidized tools, must be separated by the strong
preferential use of quartz "primary" flakes.
The obsidian "secondary" flake grouping with its distinctive
bifacial retouching and bifacial percussion flaking to shape tools
predominates at HPS 23, 31, and 32, all sites with datable ceramic
components and projectile points.
130
Table 13-
Paired sites coefficient of agreement for major variables.
Category
Reduction
Retouch
Manufacture
Stage
171.8
177.3
172.0
1?4.6
180.5
154.6
192.4
176.7
185.4
179.8
150.6
186.4
186.2
168.6
159.2
132.0
192.0
183.6
176.7
154.2
111.0
184.5
160.0
184.1
172.7
Material
3
4
31
32
23
31
1
2
40
48
37
143.0
189.0
172.2
174.9
112.1
164.4
140.3
153.9
178.8
138.6
68
48
66
131
The varnished basalt nodule-ejecta industry of HPS 66 is like­
wise readily separable from other samples.
These tools' surface alter­
ations suggest contemporaneity with the utilized cortical and cortex
removal flakes found at HPS 40 and 48.
Selection and use of large non-cortical (primary) flakes
characterizes a series of sites with both light and moderate oxidation
and limited amounts of caliche encrustation and groundstaining.
In
particular, components at HPS 3, 4, 20, 40, and 48 show selection of
basalt, rhyolite and occasionally chert primary flakes which are unifacially or bifacially flakes prior to unifacial retouching.
Primary
flakes and bifacial thinning flakes are also selected for tool-making
at HPS 1 and 68, but at these sites little tool percussion flaking
occurs (being mostly unifacial); instead unifacial retouch appears as
the single predominant tool-making technique.
These sites appear the
most homogeneous of all 13, having few intrusive tools indicated either
by surface alteration or technique.
They also have few tool types.
The site comparisons suggest less homogeneity than expected, indicating
that the collected samples are representative of several different
cultural groups and time periods.
Nevertheless, the attributes as
analyzed for individual sites and then compared by quantitative analy­
sis demonstrate major differences among sites.
CHAPTER 8
SUMMARY OF STUDY RESULTS
Results of the Sierra Pinacate chipped stone tool study show
the type of information obtainable when systematic, quantitative analy­
sis is applied to attributes of selected material, tool-making methods,
and surface alteration.
Although variable interrelations did not
always provide the clarity desired, four conclusions may still be
drawn; whole integration partially verifies the initial hypotheses.
These conclusions are:
1. There are at least four distinct patterns of tool fabrication
in the Pinacate collection.
2.
There are inter-site variations in the occurrence of both indi­
vidual attributes and these tool-making patterns.
3.
Most sites have two or more patterns of tool manufacture.
k. Form and degree of surface alteration distinguish some tool
fabrication techniques.
The hypothesis that if a single method of manufacture predomi­
nates, then the site represents primarily one cultural tradition was
supported in two sites only. If a tool-making tradition may be defined
as a group of co-occurring attributes, prevalent at a single site, then
samples often may have two or more methods of manufactures (traditions)
represented.
132
133
The second hypothesis that if a single tool-making pattern was
identified, then a consistent form of surface alteration should occur
was only partially verified.
Surface alteration often is material
specific; however, occurrences of surface alteration forms:
varnish,
oxidation, green lichen or hydration, appear chiefly at certain sites.
Because more sites than anticipated had multiple tool-making
methods and surface alteration forms present in their samples, support
for the assumptions underlying these hypotheses is not explicit.
One
explanation for the study's limited success is that only Tinaja
Badilla (HPS 2 and 3) had sufficient numbers of artifacts at loci to
be independently analyzed.
Equally important, however, in limiting
the usefulness of quantitative results was a lack of careful attribute
definition and/or the failure to include other attributes which might
have contributed information supporting the hypotheses.
A final prob­
lem was a product of selecting discrete variables and the resultant
contingency tables which contained many "empty cells."
In the following discussion I shall review each conclusion and
its pertinent supporting data, then elaborate upon problems encountered
during analysis.
The Traditions
Identifying tool traditions involves analyzing raw material
selection, the techniques applied to each material, and the final tool
function (Jelinek 1976:22-25)•
Though debitage is primarily employed
to indicate technique variations, the tools themselves can suggest much
about a culturally determined distinct technological tradition (Sheets
IJk
1975:367)•
In the three aspects of tool-making;
tradition, material,
technique and the final product, the Pinacate tool collection and the
site samples indicate four patterns of fabrication and hint at an addi­
tional two.
The initial attribute distinguishing a tradition is the stone
for tool-making. Five materials are primarily employed in the Pinacate
in decreasing frequency:
chert (Fig.
3).
basalt, obsidian, quartz, rhyolite, and
These stones may be found with and without surface
alteration; therefore, alteration and material selection can be com­
pared and a relative chronology established.
A review of their co­
occurrence indicates that basalt and rhyolite are most often heavily
weathered by varnish, oxidation and groundstaining, while chert and
obsidian show little or no exposure alteration. Heavily varnished,
groundstained and oxidized basalt and rhyolite tools appear to be the
oldest.
These specimens share several additional attributes indicative
of techniques applied to specific stones.
They are mostly made on
nodules or on ejecta and because the chosen material may be found in
either easily shaped natural nodules or lozenge-shaped ejecta, only
thinning and sharpening of the edges with a hammerstone is necessary
to convert them into large unifacially and bifacially percussion flaked
tools.
As this preferred stone category is readily available only on
the periphery of Celaya crater, it is not unexpected that tools exhibit­
ing these attributes occur primarily at Sitio Celaya.
A series of heavily oxidized and occasionally varnished basalt
flake tools suggest that a second aspect of the earliest tool tra­
dition was the expedient use of large cortical and cortex removal
135
flakes.
These are shaped only by percussion retouching and their
irregular outlines suggest that non-prepared flakes, by-products of
core or nodule initial percussion reduction, were chosen from among
the debitage and retouched to make flake tools or that utilized cor­
tical or cortex removal flakes were resharpened.
By combining these two aspects of tool manufacture with their
almost uniform surface alteration, the early tool-making tradition may
be defined as the preferential use of basalt nodules and ejecta shaped
by serial unifacial or bifacial percussion reduction with a hammerstone.
Accompanying these "chopper" and "chopping tools" are minor
numbers of cortical and partially cortical flakes with percussion re­
touch.
These flake tools have not been previously suggested as being
contemporaneous by Hayden.
Absence of soft hammer percussion, prepared
core and prepared flake use and pressure techniques is notable and
distinguishes this tool grouping from the three other traditions pres­
ently identifiable in the collection.
The first tradition is commonest
at HPS 66 and ^8 (Table lkf Figs. 13 and 1*0.
The second tool-making tradition is also based on the selec­
tive use of a single material, quartz.
Confined primarily to two
aceramic areas (at HPS 2 and 3), these tools are mostly large noncortical (primary) flakes which are seldom reduced by percussion, but
rather have edges retouched by both pressure and percussion (Fig. 15).
Often a green lichen adheres to tool surfaces.
Hayden (personal com­
munication 1978) suggests that the quartz sample is the predominant
tool material observable at HPS 2, so no attempt was made to selec­
tively collect this stone.
Therefore, the quartz flake and occasional
Table 14. Site comparisons.
Site
Basalt
Rhyolite
Nodules
Cortex
Flakes
Unifacial
Reduction
Unifacial
Retouch
Varnish
Oxidation
Relative frequency of Tradition 1 attributes (percentage)
( s i t e 40 being leaBt r e p r e s e n t a t i v e )
*40
48
66
Expected
Frequency
42.9
82.6
69.9
21.2
4.8
7.2
21.7
14.0
43.8
26.9
20.0
47.8
60.0
47.8
64.2
78.9
71.2
62.5
28.0
16.9
45.2
43.2
11.5
18.6
15.9
54.6
66.7
26.1
Relative frequency of Tradition 2 attributes (percentage)
(site 40 being least representative
Primary
Flakes
Multi-directional
Cores
Unifacial
Retouch
Green
Lichen
Site
Quartz
1
2
3
4
37
25.6
56.0
34.5
51.7
46.2
16.0
11.6
36.2
18.8
42.1
19.2
26.5
76.8
53.8
42.5
13.5
50.6
28.6
56.8
54.2
13.2
6.2
2.2
12.5
38.6
14.8
66.6
8.6
0.0
0.0
Table 14—Continued.
Site
Basalt
Rhyolite
Chert
Primary
Unifacial
Reduction
Unifacial
Retouch
Oxidation
Moderate
Alteration
Relative frequency of Tradition 3 attributes (percentage)
(sites 20 and 37 being least representative)
3
4
20
37
40
48
66
Expected
Frequency
53-0
58.5
27.7
22.5
42.9
82.6
69.9
8.0
1.8
10.6
27.5
21.0
7.0
5.9
6.0
4.9
12.8
10.4
11.4
1.2
3.3
36.2
42.1
51.4
50.6
33.3
28.8
43.8
53.8
42.5
45.0
56.8
6o.o
47.8
66.7
78.8
64.6
53.3
43.2
11.3
7.7
38.6
54.6
78.4
71.2
62.5
44.7
53.8
23.3
52.0
24.0
14.5
14.6
25.0
19.2
13.8
18.0
34.4
30.1
29.9
66.7
31.7
23.3
60.9
Relative frequency of Tradition 4 attributes (percentage)
(site 37 being least representative)
Site
23
31
32
37
Expected
Frequency
Secondary
Flakes
Bifacial
Retouch
Hydration
33.3
12.4
16.3
9.1
10.5
10.1
41.5
23.5
18.1
13.9
62.9
47.2
41.3
31.3
5^.5
32.1
34.3
10.0
14.2
7.7
14.4
27.4
10.5
Obsidian
44.9
38.6
Chert
Figure 13. Tools of Tradition 1 from Sitio Celaya
The following are catalogue numbers from the personal
files of Julian D. Hayden, Research Associate,
Arizona State Museum, Tucson. See also Figures
14-17.
a. 66.156
b. 66.182
c. 66.204
d. 66.148
e. 66.110
f. 66.146
Figure 13. Tools of Tradition 1 from Sitio Celaya,
Figure 14. Possible Tradition 1 tool kit from Tinaja Doble. —
a. 48.63
b. 48.82
c. 48.27
d. 48.74
g
Figure 14—Continued.
if8.79» f.
48.80,
g. 48.61
Figure 15. Quartz tools of Tradition 2» —
a. 17.31
c. 50J*
b.
9.1
d. 18.H
ikz
quartz core tools may be considered a separate stone working tech­
nology, although whether they represent a distinct cultural tradition
cannot be established.
The quartz tool group is technologically similar to the third
tradition which contains most Pinacate specimens. It has several
distinct attributes.
While material selectivity characterizes the
three other traditions, a variety of stones are similarly worked in
tradition 3»
Basalts and rhyolites are widely used, but cherts and
chalcedonies as well as andesite and prophyries are not uncommon.
Most materials show light to moderate oxidation and groundstain, while
some show caliche coatings.
This group may be distinguished by the selection of large noncortical primary and bifacial thinning or side-struck ("special pri­
mary") flakes. In contrast to the other traditions, these flakes are
substantially thinned by both hard hammer, and to a lesser degree,
soft hammer percussion on one or both faces.
Tool retouching, when
it occurs, is unifacial, a series of overlapping flake scars randomly
placed along lateral and transverse edges.
Unifacially retouched
flakes typify the third tradition, but utilized flakes do not, so re­
touching may be a resharpening procedure employed to extend the tool's
use (Fig. 16).
Several bifaces also exhibit the technique of extensive
percussion thinning and may represent preparation of San Pedro and
Amargosa points (Rosenthal et al. 1978) identified at HPS
20, and
37 as they show limited use of pressure work except in their basal
notching.
As most specimens come from sites with little or no ceramic
%
CR.W.
Figure 16. Tools of Tradition 3» —
a. 37-21
c. 40.30
b. 40.75
d. 40.31
e. 40.15
f. 40.34
H
-P-
3M
evidence and appear to have a technology similar to known pre-ceramic
points, this tool tradition may represent the Amargosa complex.
Hayden (1967) postulated a continuity in cultural tradition
between pre-ceramic and Ceramic times in the Pinacate, however, the
chipped stone tool collection does not confirm continuity in manufac­
ture technique.
Instead, sites that have Pat&yan ceramics in abundance
and historic and late prehistoric point forms (Rosenthal et al. 1978)
have distinct technological attributes.
Tradition 4 is characterized by the selection of small obsidian
and occasionally chert secondary flakes for tool-making (Fig. 17).
Specimens generally show bifacial pressure retouch alone, seldom being
thinned except by pressure flaking (as among bifaces).
The only
visible surface alteration is a change in the obsidian from clear
texture to an opaque matte surface(hydration).
At least two sites
were loci of point manufacture, having numerous small bifaces and tri­
angular concave-base arrowheads.
The specificity of this tool group
suggests limited activities related to hunting since the few tool
types are technologically very consistent.
It may, therefore, repre­
sent only a portion of the tool-making methods employed during late
prehistoric and historic times.
Two other tool attribute patterns are not distinctly repre­
sented in the collection due to their limited quantities.
At two
sites only very thin, fine-textured (chert, rhyolite) large, noncortical and bifacial thinning flakes were collected.
In contrast
to the primary flakes of tradition 3» which are large flakes of
initial manufacturing stages, these tools have little or no percussion
Small tools of Tradition k from Chivos Tanks. — a. obsidian 23-89; b* basalt
23.55; c. cherty rhyolite 23.81; d. obsidian 23.8lf; e. cherty phyolite 23.M;
f. chert 23«85»
146
thinning and may represent later tool-making stages. Their usual modi­
fication is unifacial pressure retouch, while a light oxidation is
observable on their faces.
A second less distinct technological pattern is the use of
certain core tool forms.
Some western desert cultures like the Cali­
fornia Milling stone horizon complexes and the Cochise have large core
tools called "pulping" or "scraping" planes (the Sayles complex of
Kowta 1969)•
Similar unidirectional cores (the aforementioned plane)
with a platform established on either a single flake scar or a natural
break in the material are found only occasionally in the Pinacate
collection.
p. 105).
They are relatively more common at some sites (Table 7»
Their alteration is mostly moderate oxidation and occasion­
ally caliche coating.
These are the four technologic traditions of the Pinacate.
Three appear to represent distinct time periods as established by
either surface alteration or the presence of tool making attributes
common to known point types. These distinct traditions suggest sepa­
rate stone working methods for the early period (San Dieguito complex),
for the pre-ceramic period (Amargosa I and II) and for the
eramic
period (Amargosa III and TV). The latter distinction is a new one
while the difference between the San Dieguito and Amargosa had been
previously hypothesized by Hayden (1967* 1976).
Site Variation
Inter-site variation, the occurrence or non-occurrence of
attributes as expressed in variable frequencies, is part of the
1^7
rationale for establishing the four tool-making traditions. Differ­
ences among sites are notable. Several sites produce mostly one pat­
tern of attributes; others have two or more, while still others have
such variety that if studied in isolation no pattern would be dis­
tinguishable. Although only 13 sites had large enough samples for
statistical analysis, cultural and temporal association of other sites
may now be clarified by comparing in situ artifacts with what is known
about the samples intensively studied, or by collecting additional con­
trolled samples.
Table 1^ groups sites by technological attributes into their
predominant traditions.
In each group a site which has only minor
representation of the patterned attributes has been added for compari­
son, as noted in the caption accompanying the tradition.
Surface
alteration is also recorded to provide the relative chronological re­
lation.
The table's four divisions detail the prominent attribute's
percentage representation and compare it with the expected frequency
at the 13 sites rather than with the collection's overall frequency
(already presented in Chapter 5)«
Summarizing Table Ik, the early basalt tradition of nodule and
ejectum hard hammer percussion is well established at hps 66, has a
minor component at hps *f8, and a small number of specimens at hps 40.
The quartz tool grouping, tradition 2, is the major component at hps 2,
and a lesser one at hps 3 and 37.
hps
A few quartz tools also appear at
which has been added for comparison.
basalt and rhyolite flakes is found at hps 3i
and is a minor constituent of the tool
The third tradition of
20, 37, **0, and ^8,
sample at most sites.
The
148
fourth tradition of small retouched obsidian tools exists at three
sites, HPS 23, 31, and 32, while minor numbers appear at HPS 37 and 20.
The two poorly represented traditions are each present at two sites.
Bifacial thinning and primary flakes with limited percussion flaking
characterizes HPS 1 and 68 while core tools of unidirectional form with
platforms established on planar, natural fractures are found at HPS 3
and 4.
As may be seen from an inspection of the data in Table 14, one
tradition may appear at several sites. Further, two or more groups of
tools displaying differing tool-making attributes may exist at a single
site.
Specifically, sites like HPS 4 or 66 have heavily varnished or
oxidized nodular-ejecta tools (though in differing amounts) and basalt
and rhyolite flake tools (tradition 3); HPS k additionally has tradi­
tion 2 quartz tools, while HPS 66 has a few late obsidian specimens,
tradition 4.
In contrast, only two sites have fairly uniform samples dis­
playing predominantly one tool-making pattern and one surface altera­
tion form.
They are HPS 1 and 68 where the large unmodified non-
cortical flakes predominate. Five other sites either have one major
component or primarily represent one major technological tradition;
the quartz dominated HPS 2, the small obsidian flakes at HPS 23i 31*
and 32, and the large basalt tools at HPS 66.
Variation among sites can be seen in the simple presence or
absence of an attribute (the absence of quartz at HPS 40, for example)
or in the presence or absence of a pattern of co-occurring attributes
(like the absence of varnished, large, bifacially percussion reduced
149
basalt ejecta at HPS 23)•
Even when distinct groups of variables are
not observable as the first hypothesis supposed, still, differences
can be established by just comparing single attributes.
Interpreting
both the single frequency and the co-occurrence frequencies is diffi­
cult because one must expect that at least minor fluctuations are a
result of Hayden's selection of some tools over others.
It appears that the major groups of tool-making attributes
cited represent existing tool differences, because these distinctions
exist not only in the total collection, but also among the site
samples. Further, the site groupings which resulted from an inspection
of the data are products of readily observable, quantitatively, greater
tabulations of the attributes. Only where percent representations
differed demonstrably for expected representation were traditions in­
ferred. Where variable frequencies appeared random or equally dis­
tributed, no decision about site's affiliation or the major component
present was made.
Surface Alteration
The previous discussion of traditions and site variation em­
ployed surface alteration when establishing site distinctions, because
the form and amount of alteration varied with other attributes. Al­
though analysis suggests that surface alteration is a relative indi­
cator of age (length or exposure), closer inspection of the series of
observed attributes is necessary. There is a clear relationship
between surface alteration and material, yet within each stone category
alteration may vary.
150
When analyzing a single material and alteration form, each must
be compared to the other variables representing attributes of manufac­
ture technique.
As such an analytical procedure is cumbersome, two
other approaches were taken.
First, the relative amount of alteration
was ranked and studied for each material (alteration X amount X mate­
rial); second, alterations that appear on different stones (like oxi­
dation or groundstaining) were analyzed. Groundstaining and oxidation
were found on most materials while green lichen and hydration were
material dependent so all patterns had to be studied independently.
Antiquity can be partly established by the noting of the cooccurring patterns among these alterations.
A tool that is heavily
groundstained and oxidized or coated with desert varnish should be
older than a tool having no varnish, light oxidation and groundstain­
ing.
Desert varnish co-occurs with both oxidation and groundstaining,
while the latter two may occur in isolation or together.
Varnish is
found only on basalt and rhyolite, predominantly on the former.
The
combined alteration is present on large nodule, or ejecta tools, or on
occasion on flakes retaining cortex which exhibit serial unifacial or
bifacial percussion retouch (mostly on cortical flakes).
Varnish is
almost completely absent from large and small non-cortical flakes and
from cores.
A second distinct alteration form is the appearance of a green
adherent identified as green lichen (C. Vance Haynes, personal communi­
cation 1977)•
It is found on quartz tools that were collected mostly
at HPS 2. Quartz from other samples have little of this alteration.
Although the reasons for the lichen occurrence is unknown, it may be
151
related to exposure, and Haynes has observed its formation on Cochise
artifacts from eastern Arizona.
A physical change in the surface of obsidian ("hydration") is
distinctively distributed among Pinacate sites and tools. At HPS 23,
44.9% of all tools are obsidian and 54.% of all these specimens are
altered. Similarly, 33*3% of all tools at HPS 32 are obsidian and
34.9# of obsidian tools have hydration. In contrast, obsidian tools
observed in small collections from HPS 17 and 18 (Sunset Camp, a his­
toric site in part) are primarily unaltered. A larger sample from the
latter loci may have permitted comparison of late prehistoric obsidian
techniques using relative surface alteration as a temporal index.
Oxidation of basalt or rhyolite, when appearing alone, also
suggests relatively longer periods of exposure for samples from some
sites.
The limited, light surface alteration observable at HPS 68
(9056 light; one of 4 tools being basalt or rhyolite), at HPS 37 (75%
of alteration is light; one of two are basalt or rhyolite tools),
contrasts with sites like HPS 40 where J>tyo of tools are moderately
altered (2 of 3 are basalt).
As Table 10 (p. 117) demonstrated, each
site is quite distinct in its alteration pattern and at sites having
one of the four traditions in predominance, the prevalent alteration
may be considered characteristic of the tool grouping.
Problems of Analysis
As can be seen from this summary, many ideas resulting from
the study are drawn from an inspection of the FREQUENCY, CROSSTAB, and
ONEWAY data. This means that the tool groupings proposed are not
152
mathematical associations (factors or clusters) but are interpretations
of attribute distributions.
The technique employed which used discrete
variables and permitted direct attribute tabulation, did not provide
grouping by calculating interrelations.
chosen approach.
This is the weakness of the
Although quantitative results cannot be disputed,
observations and interpretations can. Therefore, where possible I have
detailed calculations or scores to provide background for the interpre­
tive statements. Still, results are debatable.
A problem developed during contingency table generation.
When
this approach was chosen it was not anticipated that many major attri­
butes would be minimally represented or completely absent from indi­
vidual sites. The contingency table chi-square approach has two con­
straints limiting its usefulness (Clark 1976:260-261); the need to have
expected cell values of greater than five, and the increased likelihood
of significant results when large numbers are employed and more than
two-way associations are studied.
Both constraints exist in regard to
the Pinacate collection.
Even when similar attributes were combined, the problem of
minimal representation persisted. Further, when studying variation,
combing attributes simply to meet chi-square statistical requirements
was an unacceptable alternative.
The low representation of major values among attributes of
stone and flake selection, percussion reduction, retouching, and sur­
face alteration at several sites made the acceptance of the chi-square
results questionable. The chi-square was used, finally, to guide
selection of data for attribute comparisons, and final interpretations
relied upon comparing relative frequencies of attributes at paired
sites.
It was disappointing that so many attributes had limited dis­
tributions making statistical analysis difficult. However, the absence
or limited presence of certain attributes serves equally to suggest
site distinctions and aids in delineating site specific manufacture
patterns.
A second problem is not a result of the technique employed,
but is rather a product of laboratory procedures. During cataloging,
notes were taken to provide information for designing the computer
code. Additionally, a literature search for attributes was conducted;
in retrospect, some variables were omitted and others included that
need not have been. For example, the variable "technological mistakes
and breaks" was poorly defined and thus produced unclear results. The
seeming randomness of retouch, in contrast, might have been better
understood if location or number of serial scars variables had been
included in the code. In the future, the variable code and its
accompanying manual should be more carefully designed.
A final problem lies in the inability to interrelate nominal
and continuous variables without masking variation.
After study com­
pletion, it was recognized that measurement groupings might have been
added and compared. Such a procedure may be attempted in future
studies.
15^
Tool Typology
This study has tried to resolve the methodological problem of
analyzing surface collections.
It has inadvertently produced some
cautionary notes, particularly for typologies.
The Pinacate region's
tool collection has a simple technology, with most elaborate produc­
tive and preparatory tool methods being absent. Yet, just these attri­
butes aid development of a tool classification for typological pur­
poses.
To provide comparative information, I designed a tool typology
of 33 values using form and modification as classification criteria
(Chap. 4).
Quantitative data indicated that at most sites tools could
be placed in one of six types:
large unifacially or bifacially per­
cussion shaped forms; utilized or unifacially retouched flakes; and
bifaces or projectile points.
Attempts to isolate denticulation,
notching and burination, only suggested their insignificant presence.
Further, the irregular edge retouching suggests that a typology based
on flake modification has limited application because retouching may
be primarily a resharpening method.
Thus, the Pinacate study raises
the question, why employ an established typology when most classes are
insignificantly present?
Epstein (n.d.:6^) has commented on a similar
situation when discussing material from the San Isidro site.
The sample could be classed into more conventional niches,
but it would be misleading to use the conventional termi­
nology for this material, because such use would imply that
the artifacts are sufficiently patterned or consistent in form
to be typed according to patterns established elsewhere.
Like the San Isidro assemblage, the Pinacate's large percussion
flakes nodules and ejecta as well as the retouched flakes are highly
variable in shape and dimensions as well as in application of
155
manufacture methods, and in the range of artifacts that might be
classed as scrapers are few exclusively lateral or transverse forms.
Two possible reasons exist for this: retouching may be a resharpening
technique, or edge shapes may be unstandardized. The analytical re­
sult, however, is a highly variable group of unifacially retouched
tools which must be lumped in one category. An alternate approach to
that of typology, if finer distinctions cannot be made by employing
formal type concepts, is to employ the steps of tool-making and the
final form to develop a comparative system of classification.
The Pinacate collection demonstrates, for example, that between
the unifacially retouched obsidian and the basalt flake ("a scraper")
there is a major difference involving the flake selected and its thin­
ning prior to retouching. Yet, the scrapers may have had identical
functions. Therefore, it appears that in tool comparison the quanti­
tative occurrence of manufacture attributes and selection patterns is
more helpful than classifying the form of the tool itself, if, for
example, the unifacially retouched flake is the most common tool for
both prehistoric and historic periods.
CHAPTER 9
CONCLUSIONS
The Sierra Pinacate chipped stone collection is Hayden's selec­
tive sample of tools present at 5k of the 72 known sites in the region.
For analytical purposes it is a tool population consisting predomi­
nantly of either flake or nodular tools made from basalt, or less
commonly, other igneous stones. Selection of nodules or ejecta re­
quiring little reduction for shaping or of primary flakes requiring
substantial thinning prior to final modification characterizes the
collection. Four tool-making traditions each with distinct attributes
may be recognized.
They are:
1. Large basalt or rhyolite nodules, ejecta or cortical flakes.
The nodules are either unifacially or bifaciailly hard-hammer percussion
flaked and the flakes unifacially percussion retouched.
2.
Quartz primary flakes and cores with limited percussion reduc­
tion but with either unifacial pressure or percussion retouch.
3. Primary flakes of basalt, rhyolite and occasionally chert,
substantially percussion thinned by both hard-hammer and soft-hammer
(to a much lesser degree) percussion unifacially or bifacially, and
showing limited, randomly located serial pressure and percussion
retouch.
k.
Obsidian and chert secondary flakes mostly bifacially pressure
retouched.
156
157
A relative chronology based on the form and degree of surface
alteration suggests that tradition 1 is the earliest (San Dieguito);
traditions 2 and 3 are probably related (Amargosa); and tradition ^ is
most recent (ceramic and historic).
The study hypothesis — that if a single tool-making tradition
was present, then a site represented only one cultural tradition — was
partially supported at two sites. However, most sites appear multicomponent, having several distinct tool groupings.
The second hypothe­
sis — that if a uniform surface alteration was present, then a single
temporal horizon was represented by the sample — was partially veri­
fied, but alteration appears to be primarily material specific.
Typological problems exist in the collection due to the general
irregularity of edge shape and the lack of patterned edge retouch modi­
fication.
Future studies should incorporate flake selection and per­
cussion reduction techniques as well as final modification and shape
when designing a typology.
Within this collection material and flake
selection, and percussion reduction seem to vary more than other attri­
butes, and therefore, appear more indicative of cultural distinctness
than tool form.
This study's attempt to isolate and identify tool-making tra­
ditions among surface collections suggests that quantitative approaches
are useful but necessitate careful field observations to ascertain tool
groupings. Sampling of site areas with attention to numbers (minimum
sample size) and spatial location, so as to provide specific groups to
mathematically analyze would be helpful.
should be included in the collection.
Both tools and debitage
158
, The contingency table procedure provided data on attribute co­
occurrences and was most helpful, but the approach of combining it with
the chi-square statistic has some constraints. Simple comparison of
expected and actual frequencies provided more useful information than
the use of the chi-square statistic. Successful, acceptable chi-square
results could only be obtained by eliminating variable states with low
representation; combining attributes. In most situations this elimi­
nated variation and could not be justified. By inspection, however,
it could be determined that observed and expected frequencies deviated
greatly and, therefore, assumptions of strong attribute associations
could be made without the statistic.
Study results indicate that much more work must be done to de­
fine, classify and describe attributes preparatory to use of con­
tingency tables. Further work also is needed to refine the use of the
SPSS package for both nominal and continuous archaeological variables.
It must be reiterated that the success of studies like this lies in
the ability of the researcher to review and select the most informative
variables to record. This appears to improve with successive attempts
to describe tool attributes. Each investigation must build upon pre­
vious work, altering the codes to suit the collection; this study is
no exception. Future work must thus include rethinking and reorganiz­
ing the lithic code so that more successful analysis of the qualities
which distinguish one people's way of making tools from another's can
be achieved and usefully applied to the study of surface collections.
APPENDIX A
CODE DESCRIPTION
1. 1-3:
Survey Definition (HPS, WPS, EPS)
2.
5-8:
Site Number
3.
8-10:
k.
13-16: Coding Sequence Number
# Coding Spaces 1-16 for Provenience and Identification
5. 18-19:
1
2
3
k
5
6
7
8
9
10
11
12
13
Ik
6.
7.
15
16
17
18
19
20
21
22
23
2k
25
26
27
Diorite
Labradorite
Jasper in Rhyolite
Siltstone
Olivine-Basalt
Porphyritic Basalt
Arkose Sandstone
Porphyritic Rhyolite
Silicious Mudstone
Limestone
Basalt Porphyry
Welded Tuff
Ignimbrite
Category
Flake
Core
Worked Nodule or Ejectum
Utilized Stone
27-28:
1
2
3
k
Material
Fine Basalt
Scoriaceous Basalt
Andesite-Felsite
Rhyolite
Porphyry
Granite-Dacite
Obsidian
Chert-Jasper
Chalcedony
Quartz
Mica Schist
Pumice
Quartzite
Trap
21-22:
1
2
3
k
Catalog Number
Retouch (Pressure
Unifacial
Bifacial
Alternate
Reverse
Controlled Percussion)
5
6
7
8
Multiple
Platform
Oblique Blow (Burination)
Bifacial Reverse
159
160
8.
2k-23'
Percussion Reduction Flaking
1 Unifacial
2 Bifacial
3 Alternate
9.
30-31:
4 Reverse
5 Multiple
Surface Alteration
10.
33-3^:
1
2
3
k
5
11.
Amount of Alteration
Light
Moderate
Heavy
Partial
Heavy Oxidation,
Light Varnish
36-37:
21 ( 2 + 3 + 5 )
22 (5 + 19 + 1)
23 Orange Rust
2k (5 + 12)
25 (2 + 22)
26 (2 + 16)
27 (2 + 20)
28 (3 + 27)
29 (8 + 19)
30 ( 1 + 2 + 3 + 5 )
11 Green Salts
12 (3 + 11)
13 (3 + 5)
l*f Hematite
15 (11 + 2)
16 ( 1 + 3 + 5
17 (6 + 3)
18 ( 1 + 2 + 3 )
19 Black Oxide
20 (5 + 19)
1 Varnish
2 Groundstain
3 Caliche
if Eroded
5 Oxidized
6 Hydrated
7 (1 + 5)
8 (2 + 5)
9 Heat Treated
10 ( 1 + 2 + 5 )
6
7
8
9
10
Light Oxidation, Moderate Caliche
Heavy Oxidation, light caliche
Heavy Qroundstain, Light Oxidation
Moderate Oxidation, Light Groundstain
Moderate Caliche, Light Varnish and
Oxidation
Technological Attributes
8 (1 + k)
1 Platform Abraded
9 Perverse fracture
2 Platform Crushed
10 (3 + 4 + 9)
3 Step-fractures
11 Auraillar
k Hinge fractures
12 Transverse break
5 (2 + 3)
13 Longitudinal. Break
6 (3 + *0
Ik (12 + 13)
7 (2 + 3 + k)
12.
39-kO:
15 (3 + 12)
16 (9 + 12)
17 (6 + 9)
18 (2 + 13)
19 (13 + 9)
20 (15 + *0
21 (k + 12)
Affiliation
1 Malpais
2 San Dieguito
9 (3 + *0
10 (1 + 2)
17 San Pedro
3 Amargosa I
k Amargosa II-III
11 Late S.D.?
12 (2 + 9)
3
6
13
2A
Brownware
Sand Papago
(3 + 5)
(5 + 6)
7 Yuman II
8 San Dieguito
or
Amargosa I
15 Chiricahua
16 Trincheras
161
13.
k2-k3i
1
2
3
k
5
6
7
Flake Stage of Manufacture
cortical
cortex removal
primary (large non-cortical flake)
special primary (sidestruck and bifacial thinning)
secondary (small non-cortical flake)
blade
secondary trimming (small cortical flake)
8 (2 + *0
9 shatter
10 (2 + 6)
11 flake core
l*f.
k5-k6:
Core or V/orked Cobble Directionality and Preparation
1 unidirectional
2 bidirectional
3 multidirectional
4 prepared platform
5 (1 + *0
15»
48-^9:
6 (2 + *0
7 (3 + *0
8 (7 + exhausted)
9 (5 + exhausted)
10 unprepared,(3 + k)
Descriptive Tool Type
1 unifacial percussion reduced "chopper"
2 bifacial percussion reduced "chopping tool"
3 unifacial retouched "scraper"
abrader
5 burin blow
6 utilized blade
7 utilized flake
8 utilized core
9 projectile point
10 biface "projectile point preform"
11 retouched blade
12 (1 + 3)
13 converging reduction "perforators"
l*t denticulate
15 converging retouched piercing tool
16 bifacial retouched "knife"
17 notched
18 (3 + 17)
19 (8 + 17)
20 (1 + 17)
21 worn nodule
22 tip/base of projectile point
23 (2 + 8)
2k (3 + 16)
25 (1^ + 16)
26 (3 + 7)
27 eccentrically flaked
28 (16 + 17)
162
15. Continued
29 (15 + 16)
50 (1 + hammerstone)
51 tabular retouched
52 beaked
55 (1 + metate)
16.
51-52:
Location of Use Damage
1 planar core
2 accidental on flake
5 inner
k outer
5 transverse
6 lateral
7 (5 + 5)
8 (if both)
9 (5 + 6)
10 (4 + 6)
11 (if + 5)
12 inner/outer lateral
15 everything
5^-55:
1
2
5
4
18.
Edge Profile
Platform Preparation of Flakes
1 cortical
2 unifaceted
19.
5 (1 + 2)
6 (2 + 5)
7 (1 + 5)
concave
convex
straight
all
77-78:
56-58:
Ik 5 all
15 (5 + *0
16 on platform deliberate
17 (5 + ^ + 5)
18 (2 + 5 + if + 5)
19 (2 + 12)
20 (1 + 2)
21 (if + 6 + platform)
22 transverse alternate
25 on an edge
2b (1 + 15
25 both, lateral alternate
26 5, if alternate
5 dyhedral
k multifaceted
5 planar
6 reworked
Edge Working Angle
20. 60-62:
Length Perpendicular to Platform (or greatest)
21. 6if-66:
Width Parallel to Platfonn (or greatest)
22. 68-70:
Thickness greatest
25.
72-7if:
Rake Angle (planar inclination of edge)
APPENDIX B
ANALYSIS OF DEBITAGE
Provenience of Sierra Pinacate debitage.
Sit
Amount
Site
1
k
29
2
12
31
3
98
32
k
285
J>k
39
8
2
35
k
13
13
36
16
Ik
87
37
95
16
15
39
1^
17
302
ko
159
18
20
kl
37
19
1
kj>
22
20
5k
J»8
11
21
1
3k
1
22
k
56
22
23
60
59
72
2k
1
66
3
25
if
67
22
27
1
163
Debitage Total
1659
Classified
161*5
164
Rock Types Represented by Debitage
Raw Materials
1. Obsidian
11. Andesite Porphyry
2. Chert
12. Rhyolite
3. Jasper
13-
Rhyolite Porphyry
k.
Chalcedony
14. Granite-dacite
5.
Silicious Mudstone
15. Limestone
6. Quartz
16. Sandstone (Arkose)
7. Quartzite
17. Conglomerate
8.
Vesicular Basalt
9. Fine-textured Basalt
10.
Andesite
18. Micaceous Schist
19. Labradorite
20. Unknown
«-3 I\) H H H H H H H H H H
o O vO O0-O ONVJ1 -p-VjJ MMOvO OO-O Onvji -PVjJ I\) H
c+
&
H
O
VjJ H VX
vji on
h oo
"s3
ro
-O H
VJ1
on
ru
ro
ro
HHHrovji
ru
VjJ O
Ul
HH
•pU1 H Vj4
ro
• uj
ru
\J1
HH
H
-n3
ro
ro
vn ro ro -p- o
H
vjl
U1
-P" OO
VX
• -O
VJ1 ro
H
vO
-vl
NO -o
V>I
vx
ro vx Hui
ro VJ4 o VM
VjJ
ON
ON O
Nodular Core
•ph vn h ro co Flake Core
H
vo
>oJ H
H
ro
uihhv^iw wu
-O
Exhausted Core
(less 3 cm)
Fully Cortical Flake
(90-100&)
Partially Cortical
Flake (40-9056)
H Cortex Removal Flake
O
H N cor (0-ko£)
H
ro
H OO
O SI VX O-PVJI SI
H
rv»
ro
VfJ
vO
vx-ptovjion
ro
H
h
ru ru- p-vji ru vn ro
ro
HM
\D"0 H
H
-P
H
^3
ON
H
Kti 00
M
ro
3C U3
SO 9
cf <
v>i on ro vo
mo
Lge Non-cortical
Flake (greater than
3 cm in length)
Bifacial Thinning
Flake
Sidestruck Flake
Small Non-cortical
Flake
ro
Small Cortex Removal
Flake
ro
ro
Resharpening Flake
VJ1
vn
VjJ
>o4
VJI
-pON
-pU1
591
H 00
£
VJl
H
ON SI OV>4 ON H U4
ON CO
H
ro VX
\JJ b-> H -F H \J1
VJJ 00 ON ON O -0 OO^HL
H
Blade
Shatter
h3
O
ef-
P.
glossary
Much confusion exists in lithic technology about descriptive
terms. This glossary presents definitions used in the Pinacate study
relating to artifact description, measurement and typology. More
complete lists may be found in Bordes (1968), Crabtree (1972),
Epstein (n.d.), Marks (1976), and Tixier (197^).
1.
Alternate; Having both interior and exterior flaking from a
single edge; however, the scars are not in opposition along
the edge.
2.
Biface: Artifacts which have been reduced to a relatively thin
cross-section by controlled thinning techniques ("thinned
bifaces") ("preforms").
3. Bifacially Percussion Reduced Nodules or E.jecta:
Relatively
thick artifacts formed by hard-hammer percussion exhibiting
large conchoidal flake scars with deep negative bulbs of force
on two faces ("chopping tool") ("percussion biface").
if.
Bifacial Thinning Flake: A flake exhibiting a wide, lipped
platform and diffuse bulb, wider than it is long. These
attributes are often associated with use of a soft-hammerstone,
baton or billet during thinned biface manufacture ("billet
flakes").
5.
Blade; A flake twice as long as it is wide having scars from
previous parallel flake removals.
6.
Core; Remnant blocks of raw material which have been inten­
tionally flaked to produce either flakes or blades ("nucleus").
7.
Cortical Flake: A flake greater than 3 cms long, having natural
rind on 70% or more of its non-bulbar surface ("initial flake").
8.
Cortex Removal Flake; A flake greater than 3 cms long having
natural rind on 70% or 30% of its non-bulbar surface.
9. Debitage; Residual lithic material resulting from tool manufacture ("debris and cores") ("unutilized flakes and cores").
10.
Denticulate:
edge.
A flake having a series of notches along a single
166
167
11.
Desert Varnish: Glossy coating looking like a Bheen or liquid
plastic which covers tool surfaces composed of manganese and
iron rich sheet silicate (Hayden 1976:277; Potter and Rossman
1977:1V*8).
12. Dimensions:
Length — Maximum distance between the platform and the termi­
nation point of a flake. On a core or nodule/ejectum, the
maximum, flaked surface dimension to the nearest millimeter.
Width — Maximum distance measured parallel to the platform or
on a core or nodule/ejectum the maximum distance perpendicular
to the length.
Thickness — Maximum distance between the inner and outer sur­
faces of a flake, or on a core or nodule/ejectum and third
dimensional plane's greatest distance.
13. Ejectum: A basaltic pyroclastic stone formed, smoothed, shaped
and spewed forth during crater-forming eruptions (Hayden 1976:
28-281).
l*f.
Exterior: The artifactual surface opposite the bulbar face
having either cortex or negative scars from previous flake
removal ("outer") ("dorsal").
15. Flake: Any piece of stone removed from another by conchoidal
fracture thus displaying a positive bulb of force.
16. Green Lichen: A natural adherent of lichen found on the sub­
surface or surface contact portions of quartz pieces.
17.
Groundstaining: Accumulation of clays adhering to the subsurface
portion of toolB which becomes permanent and can only be
removed by sand-blasting or chemical solvents.
18. Hinge: A fracture which terminates a flake at right angles to
its longitudinal fracture plane, the break is usually round
or blunt.
19. Hydration: Obsidian weathering where a change from a clear
glasseous surface to an opaque, matte surface has occurred.
20.
Interior: The bulbar surface of an artifact ("bulbar"), ("inner")
("ventral").
21.
Lateral:
22.
Nodule:
23.
Notched Tool: An artifact having a concave edge portion due to
the removal of flakes by percussion or pressure notching tech­
niques.
The edge perpendicular to the striking platform.
Large tabular basalt block.
168
2k.
Oxidation: The weathering of a tool which results in a change in
the surface color of the specimen.
25. Percussion Reduction: The hard-hammer percussion method of thin­
ning and shaping. An initial tool-shaping activity.
26. Perverse Fracture:
tool's edge.
An abrupt angular break initiated along a
27. Primary Flake: The first non-cortical flake detached during core
reduction which is greater than 3 cos long ("secondary
element")•
28. Projectile Point: A thinned biface showing final manufacturing
evidence, production of a hafting area (base) or parallel
pressure flaking.
29. Retouch: Purposeful modification by percussion or pressure
flaking to the immediate edge of a tool for final shaping,
sharpening or resharpening ("secondary retouch").
30.
Reverse: Having interior flaking on one edge and exterior
flaking on a parallel or opposing edge.
31.
Secondary Flake: Small non-cortical flakes less than 3 cms long
indicating final manufacture or resharpening.
32.
Shatter: Fragmentary pieces of stone having bulbs or platform
remnants indicating their production during manufacture.
33. Sidestruck Flake: Flakes wider than they are long indicating
generally, hard-hammer percussion during biface manufacture.
Specialized Primary Flakes: Variable state which combines
"diverging" flakes (sidestruck and bifacial thinning).
35.
Step-Fracture: A flake or flake scar that has an abrupt right
angle break at its terminus.
36. Transverse:
37.
Edge parallel to the striking platform.
Hnifacially Percussion Reduced Nodule or E.jectum: Relatively
thick artifact formed by unifacial hard-hammer percussion,
exhibiting large deep conchoidal negative flake scars ("chopper")
("percussion uniface").
selected bibliography
BINFORD, LEWIS R.
1973
Interassemblage Variability — The Mousterian and the Func­
tional Argument. In The Explanation of Culture Change:
Models in Prehistory, Colin Renfrew, Ed., pp. 227-25*+.
University of Pittsburgh Press, Pittsburgh.
BINFORD, LEWIS R. and SALLY R. BINFORD
1966
A Preliminary Analysis of Functional Variability in the
Moustrian of Levallois Facies. American Anthropologist,
Vol. 68, No. 2, pp. 238-295.
BINFORD, SALLY R. and LEWIS R. BINFORD
1969
Stone Tools and Human Behavior. Scientific American, Vol.
220, No. 4, pp. 70-8^.
BORDES, FRANCOIS
1961
Typologie du Palaleolithique Ancien et Moyen. Publications
de L'Institute de Prehistoire de L'Universite de Bordeaux,
Memorire 1, Bordeaux.
1968
The Old Stone Age.
1972
A Tale of Two Caves. Harper and Row, New York.
1973
On the Chronology and Contemporaneity of Different Paleo­
lithic Cultures in France, in The Explanation of Cultural
Change: Models in Prehistory, Colin Renfrew, Ed., pp.
217-226. University of Pittsburgh Press, Pittsburgh.
Harper and Row, New York.
BRADLEY, BRUCE A.
1975
Lithic Reduction Sequences: A Glossary and Discussion.
In Lithic Technology: Making and Using Tools, E. H. Swanson,
Jr.V~EdT7ppiT5-l^ Aldine Press, Chicago.
BRYAN, KIRK
1975
Geological Interpretation of the Deposits. In The Stratig­
raphy and Archaeology of Ventana Cave, E. W. Maury. The
University of Arizona, Tucson.
169
170
CHRISTENSON, ANDREW L. and DWIGHT W. READ
1977
Numerical Taxonomy R-mode Factor, and Archaeological Classi­
fication. American Antiquity, Vol. k2. No. 2, pp. 163-179.
CLARK, GEOFFREY A.
1976
More on Contingency Table Analysis, Decision Making Criteria
and the Use of Log-linear Models. American Antiquity, Vol.
kl, No. 3, PP. 259-273.
CLARKE, DAVID L.
1971
Analytical Archaeology*
Methuen, London.
CLAY, A. BERLE
1976
Typological Classification, Attribute Analysis said Lithic
Variability. Journal of Field Archaeology, Vol. 3i No. 3t
pp. 303-311.
COLLINS, MICHAEL B.
1975
Lithic Technology as a Means of Processual Inference, In
Lithic Technology, Making and Using Tools,, Earl H. Swanson,
Jr., Ed., pp. 15-34. Aldine Press, Chicago.
CRABTREE, DON E.
1972
An Introduction to Flintworking. Occasional Papers of the
Idaho State University Museum #28. Pocatello, Idaho.
1975
Comments on Lithic Technology and Experimental Archaeology.
In Lithic Technology, Making and Using Tools, Earl H.
Swanson, Jr., Ed., pp. 105-11^. Aldine Press, Chicago.
CRABTREE, DON E. and B. R. BUTLER
1964
Notes on Experiments in Flintknapping I: Heat Treatment
of Silica Minerals. Tebiwa, Vol. 7» No. 5» pp. 1-6.
DORAN, J. E. and F. R. HODSON
1975
Mathematics and Computers in Archaeology. Harvard Univer­
sity Press, Cambridge.
EPSTEIN, JEREMIAH F.
1964
Towards a Systematic Description of Chipped Stone. Actas
y Memorias 351 Congreso Internacional de Americanistas,
Mexico, 1962, Mexico.
'
171
EPSTEIN, JEREMIAH F.
n.d.
The San Isidro Site on Early Man Campsite in Nuevo Leon,
Mexico. Anthropology Series #7, Department of Anthropology,
University of Texas, Austin.
EZELL, PAUL M.
1955
The Archaeological Delineation of a Cultural Boundary in the
Papagueria. American Antiquity, Vol. 20, No. 2, pp. 367-37^«
GLOVER, I. C.
1969
The Use of Factor Analysis in the Discovery of Artifact
Types. Mankind, Vol. 7» pp« 36-51.
GUMMERMAN, MELANIE
1976
Mechanical Models for the Analysis of Lithic Assemblages:
Preconditions Necessary for Use. Newsletter of Lithic
Technology, Vol. 5» Nos. 1-2, pp. 7-11.
GUNN, JOEL
1975
Idiosyncratic Behavior in Chipping Style: Some Hypotheses
and Preliminary Analysis. In Lithic Technology, Making and
Using Tools, Earl H. Swanson, Jr., Ed., pp. 35-62. Aldine
Press, Chicago.
GUTMAN, JAMES T.
1972
Eruptive History and Petrology of Crater
University Microfilm, Ann Arbor.
Elegante, Sonora.
HAURY, EMIL W.
1975
The Stratigraphy and Archaeology of Ventana Cave, Arizona.
University of Arizona Press, Tucson.
HAURY, EMIL W. and JULIAN D. HAYDEN
1975
Preface. In The Stratigraphy and Archaeology of Ventana
Cave, Arizona. University of Arizona Press, Tucson.
HAYDEN, JULIAN D.
1965
1967
Fragile Pattern Areas.
pp. 272-276.
American Antiquity, Vol. 31» No. 2,
A Summary History and Prehistory of the Sierra Pinacate,
Sonora, Mexico. American Antiquity, Vol. 32, No. 3« PP» 3353^.
172
HAYDEN, JULIAN D.
1976
Pre-altithermal Archaeology in the Sierra Pinacate, Sonora,
Mexico. American Antiquity, Vol. 4l, No. 3» PP» 274-289.
1977 Personal Communication. Research Associate, Arizona State
Museum, Tucson.
1978 Personal Communication. Research Associate, Arizona State
Museum, Tucson.
HAYNES, C. VANCE
1977 Personal Communication. Professor, Department of Anthro­
pology, University of Arizona, Tucson.
HESTER, THOMAS R. and ROBERT F. HEIZER
1973
Review and Discussion of Great Basin Projectile Points,
Forms and Chronology. Archaeological Research Facility,
Department of Anthropology, University of California,
Berkeley.
HODSON, FRANK ROY
1977
Quantifying Hallstatt: Some Initial Results.
Antiquity, Vol. 42, No. 3» pp« 394-412.
American
HOLE, FRANK and ROBERT F. HEIZER
1965 An Introduction to Prehistoric Archaeology. Holt, Rinehart
and Winston, Inc., New York.
IVES, RONALD I.
1964 The Pinacate Region, Sonora. Mexico. California Academy of
Science, Occasional Paper #47
JAHNS, R. M.
1959
Collapse Depression of the Pinacate Volcanic Field, Sonora,
Mexico. Arizona Geological Society, Southern Arizona
Guidebook #2, pp. 165-184.
JELINEK, ARTHUR J.
1965
The Lithic Technology Conference, November 23-28, 1964 at
Les Eyzies. American Antiquity, Vol. 31» No. 2, pp. 277-279®
1976
Form, Function and Style in Lithic Analysis. Reprint from
Cultural Change and Continuity. Essays in honor of James
Bennet Griffin. Compliments of the author.
173
JOHNSON, L. LEWIS
1975
Graph Theoretic Analysis of Lithic Tools from Northern Chile,
in Lithic Technology, Making and Using; Tools, Earl H. Swanson,
Jr., Ed., pp. 63-96. Aldine Press, Chicago.
1978
A History of Flintknapping Experimentation 1838-1976.
Current Anthropology, Vol. 19, No. 2, pp. 337-372.
KING, THOMAS JACKSON, Jr.
1976
The Archaeological Implications of the Paleobotanical Record
from the Lucerne Valley Region, Mohave Desert, California.
Master's Thesis, University of California, Los Angeles.
KLECKA, WILLIAM R., NORMAN H. NIB and C. HADLAI HULL
1975
SPSS Primer.
McGraw-Hill, New York.
KOWTA, MAKOTO
1969
The Sayles Complex: A Late Milling Stone Assemblage from
Cajon Pass and the Ecological Implications of Its Scraper
Planes. Publications in Anthropology #6. University of
California, Berkeley and Los Angeles.
KRIEGER, ALEX D.
19^
The Typological Concept.
pp. 271-288.
American Antiquity, Vol. 9» No. 3*
KROEEER, ALFRED
19^8 Anthropology. Harcourt, Brace, New York.
LEACH, B. F.
1969
The Concept of Similarity in Prehistoric Stone Flake
Assemblages. Anthropology Department, University of Otago.
LUMHOLTZ, CARL
1971
New Trails in Mexico. The Rio Grande Press, Glorieta.
MARKS, ANTHONY E. (Ed.)
1976
Prehistory and Paleoenvironments in the Central Negev,
Israel. Southern Methodist University Press, Dallas.
17^
MAY, LARRY A.
1973
Resource Reconnaissance of the Grain Dieierto Region, North­
western Sonora, Mexico. Master's Thesis, University of
Arizona, Tucson.
MUTO, GUY R.
1971
A Technologic Analysis of the Early Stages in the Manufacture
of Lithic Artifacts. Master's Thesis, Idaho State University,
Pocatello.
NEWCOMER, M. H.
1975 '^Punch Technique" and the Upper Palaeolithic Blades. In
Lithic Technology, Making and Using Tools, Earl H. Swanson,
Jr., Ed., p. 97. Aldine Press, Chicago.
NIE, NORMAN H., C. HADLAI HULL, JEAN G. JENKINS, KARIN STEINBRENNER
and DALE H. BENT
1975
SPSS Statistical Package for the Social Sciences. McGraw
Hill, New York.
POTTER, RUSSELL M. and GEORGE R. ROSSMAN
1977
Desert Varnish and the Importance of Clay Minerals.
Vol. 196, June 2k, pp. 14^6-1*148.
Science,
PURDY, BARBARA A. and H. K. BROOKS
1971
Thermal Alteration of Silica Materials: An Archaeological
Approach. Science, Vol. 173, pp. 322-325.
RANERE, ANTHONY J.
1974
Early Human Adaptation to New World Tropical Forests: The
View from Panama. Doctoral Dissertation, University of
California, Davis.
1975
Toolmaking and Tool Use Among the Pre-ceramic Peoples of
Panama. In Lithic Technology, Making and Using Tools, Earl
H. Swanson, Jr., Ed. Aldine Press, Chicago.
REDMAN, CHARLES L.
197^
Archaeological Sampling Strategies.
in Anthropology #55•
Addison-Wesley Module
175
RENFREW, COLIN
1973
The Explanation of Culture Change: Models in Prehistory.
University of Pittsburgh Press.
ROBINSON, W. S.
1951
A Method for Chronologically Ordering Archaeological
Deposits. American Antiquity, Vol. 16, No. k, pp. 293-301.
ROGERS, MALCOLM J.
1939 Early Lithic Industries of the Lower Basin of Colorado River
and Adjacent Desert Areas.
Diego.
Diego Museum Papers #3* San
ROGERS, MALCOLM J. and MALCOLM F. FARMER
19^8
A New Correlation of Early Lithic Complexes of the Far
Western United States. Manuscript in San Diego Museum of
Man.
ROSENTHAL, E. JANE, DOUGLAS R. BROWN, MARC SEVERSON and JOHN B. CLONTS
1978
The Quivotoa Valley Project. U. S. Department of Interior,
National Park Service, Tucson.
SACKETT, JAMES R.
1973
Style, Function and Palaeolithic Assemblages. In The Explana­
tion of Culture Changer. Models in.Prehistory, Colin Renfrew,
Ed., pp. 317-325. University of Pittsburgh Press.
1977
The Meaning of Style in Archaeology: A General Model.
American Antiquity, Vol. U2, No. 3» pp. 369-380.
SAYLES, E. B. and E. ANTEVS
19^1
The Cochise Culture.
Globe, Arizona.
Medallion Papers #29» Gila Pueblo,
SCHIFFER, MICHAEL B.
197^
Nomothetic Aspects of Chipped Stone Experiments. Newsletter
of Lithic Technology, Vol. 3i No. 3* PP» ^+6-^9.
SENTER, R. J.
1969
Analysis of Data, Introductory Statistics for the Behavioral
Sciences. Scott, Foresman and Company, Glenview, Illinois.
176
SHEETS, PAYSON D.
1973
The Two Chipped Stone Industries of Barriles, Western Panama.
Paper presented to the 38th Annual Meeting, Society of
American Archaeology, Sain Francisco.
1975
Behavioral Analysis and the Structure of a Prehistoric
Industry. Current Anthropology, Vol. 16, No. 3» PP» 361-391•
SHREVE, FORREST
1951
Vegetation of the Sonoran Desert, Vol. I. Carnegie Insti­
tution of Washington Publications 591. Washington.
SPAULDING, ALBERT C.
1953
Statistical Techniques for the Discovery of Artifact Types.
American Antiquity, Vol. 18, No. 3» pp* 305-313«
SPETH, J. D.
1972
Mechanical Basis of Percussion Flaking. American Antiquity,
Vol. 37» No. 1, pp. 3i*-60.
SPIEGEL, MURRAY R.
1961
Theory and Problems of Statistics. Schaum Publishing,
New York.
SWANSON, EARL H.
1975
Lithic Technology, Making and Using Tools.
Chicago.
Aldine Press,
THOMAS, DAVID HURST
1978
The Awful Trust About Statistics in Archaeology.
Antiquity, Vol. ^3» No. 2, pp. 231-2Mf.
American
TIXIER, JACQUES
197^
Glossary for the Description of Stone Tools with Special
Reference to the Epipalaeolithic of the Maghreb, translated
by M. H. Newcomer. Newsletter of Lithic Technology, Special
Publication #1, December.
TRINGHAM, RUTH, GLENN COOPER, GEORGE ODELL, BARBARA VOYTEK and ANN
WHITMAN
I97I+ Experimentation in the Formation of Edge Damage:
A New
Approach to Lithic Analysis. Journal of Field Archaeology,
Vol. 1, Nos. 1-2, pp. 171-196.
177
TSIRK, ARI
1974
Mechanical Basis of Percussion Flaking: Some Comments.
American Antiquity, Vol. 39, No. 1, pp. 128-130.
UNITED STATES DEPARTMENT OF THE INTERIOR
1976
Wilderness Proposal: Organ Pipe Cactus National Monument.
(Final environmental statement) National Park Service,
Denver Service Center.
VAN DEVENDER, THOMAS R.
1977
Holocene Woodlands in the Southwestern Deserts. Science,
Vol. 198, No. ^313, pp. 189-192.
VOEGLER, LAWRENCE
1977
Personal communication.
Tucson.
Archaeologist, Arizona State Museum,
WATERS, MICHAEL R.
in
The Diagnostics and Distribution of the Lowland Patayun
press Buffware Tradition 500 A.D.-Post 1900 A.D. San Diego
Museum of Man.
WATSON, PATTY JO, STEVEN A. LeBLANC and CHARLES REDMAN
1971
Explanations in Archaeology: An Explicitly Scientific
Approach. Columbia University, New York.
WEINBERG, GEORGE H. and JOHN A. SCHUMAKER
197*4-
Statistics:
An Intuitive Approach.
Brooks-Cole, Monterey.
WHEELER, MORTIMER (Sir)
195^
Archaeology from the Earth.
Clarendon Press, Oxford.
WHITE, J. P. and D. H. THOMAS
1972
What Mean These Stones? Ethno-taxonomic Models and
Archaeological Interpretations in the New Guinea Highlands.
In Models in Archaeology, David L. Clarke, Ed., pp. 275-308.
Harper and Row, New York.
WILMSEN, EDWIN N.
1968
Lithic Analysis in Paleoanthropology. Science, Vol. 161,
September 6, pp. 88-98.
YOUNG, GEORGE H. and PAYSON D. SHEETS
1975 A Test of Village Socio-economic Ranking Using Lithic
Debitage. Newsletter of Lithic Technology» Vol.
No.
pp. 31-37.
TRAIL (KNOWN SEGMENTS)
SITE
o
SITE
O
VOLCANIC
X
+ +
CURRENT
4 +
10
KM
CRATER
MINE
GRID
MC
OUGAb
OUGAb
BADILLA
CELAYA
BADILLA
U N/
t£d
*/ Co
CELAYA
2^KIN0
uN
t
^0
£s
*1c0
4/?,
So
CERRO COLORADO
I MOON
NORTH
©7
38
48
46
57
53
20
PI N A C
22
23
45
072
56
44
\I8
62
I
57
63
A67
61
PINACATE
072
48
26
59,
70
27
ELEGANTE] 5 9
63
54
A67
PINA C ATE
072
X.
59,
70 O
ELEGANTE/59
•54
49
40
39
50
55
I4Q)
I MOON
45
23
16
17
\I8
\
NORTH
5
J
0
L
KM
2
TINAJA MARIA
TINAJA BADILLA
3
TINAJA BADILLA
4
PAPAGO TANKS
TINAJA LAS OLL
1
F I G U R E I.
SITES
THE
OF T H E
SONORA.
ARCHAEOLOGICAL
SIERRA
PINACATE,
6
10
TINAJA
13
TINAJA DEL
17
18
SUNSET
SUNSET
19
TINAJA
DEL AGUIL
BOTE
GAMP
CAMP
DONA
Rl
d
03
A<
61
PINAC
072
56
44
\I8
\
\
62
\
5
J
KM
LOGICAL
PINACATE,
1
2
T I N A J A MARIA
TINAJA BADILLA
3
I
1
20
TINAJA DEL
TINAJA BADILLA
23
TINAJA
4
6
PAPAGO T A N K S
T I N A J A LAS. O L L A S
31
32
TINAJAS DE
TINAJA DEL
10
TINAJA
34
CARVAJALES
13
TINAJA DEL
36
JACAL RUIN
17
18
SUNSET
SUNSET
37
TINAJA
RIDGE 8
38
CARVAJALES
19
TINAJA
ANNA
39
TINAJA
DEL AGUILA
BOTE
CAMP
CAMP
DONA
DE
DEL
sir
61
A67
PINAC ATE
072
071
20
TINAJA DEL TULE
23
TINAJA
31
32
TINAJAS DE
TINAJA DEL
EMILIA
OJO
34
CARVAJALES
CAVE
36
JACAL RUIN
37
TINAJA
38
CARVAJALES
39
TINAJA
DE
LOS
DEL
SUV/*
CHIVOS
CUERVO
CORRAL
63
i54
^67
IC ATE
©
071
L TULE
LOS
E
L
)
TINAJA
CHIVOS
LAS
TINAJA
VEREDA
EMILIA
OJO
TINAJA
ONDA
TINAJA
DOBLE
CAVE
TINAJA
GALLETAL
TINAJA
EXTINTA
ALTA
M
L
CUERVO
TINAJA
S
CORRAL
SITIO
CELAYA
SITIO
LA
J VI
FIGURAS
PLAYA
TINAJA
LAS
FIGURAS
TINAJA
VEREDA
TINAJA
ONDA
TINAJA
DOBLE
TINAJA
GALLETAL
TINAJA
EXTINTA
TINAJA
ALTA
SITIO
CELAYA
SITIO
LA
PLAYA
Rosenthal, AK/THI?Opoi^osy
•Di-ss.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement