Tools Versus Cores Alternative Approaches To Stone Tool Analysis
Tools Versus Cores Alternative Approaches To Stone Tool Analysis
Tools Versus Cores Alternative Approaches To Stone Tool Analysis
Edited by
Shannon P. McPherron
All rights for this book reserved. No part of this book may be reproduced, stored in a retrieval system,
or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or
otherwise, without the prior permission of the copyright owner.
ISBN 1-84718-117-1
INVESTIGATING THE BEHAVIORAL CAUSES
AND ARCHAEOLOGICAL EFFECTS
OF LITHIC RECYCLING
DANIEL S. AMICK
Introduction
expended in quarry extraction but potential yields tend to be lower because surface
exposures of rock are more weathered and easily depleted of the best materials. In
contrast, the secondary recycling of stone tools and debris from can provide high
returns with relatively low investments of time and energy.
Figure 12-1 Hypothetical return curves for different kinds of lithic procurement
strategies (modified from Elston 1992:Figure 14).
see Gould et al. 1971:163) observed “a tendency for these people to pick up ancient
stone tools from the surface of sites where they are camped and reuse these
implements.” He concludes that, “Reutilization of already ancient materials may
have been a fairly common behavior among prehistoric peoples in many parts of the
world” and cautions that, “It can result in the discovery of early tools in much later
levels in an ancient site.” In fact, most archaeological concerns about lithic
recycling seem to be focused on the potential chronological confusion caused by
secondary recycling of artifacts that serve as discrete temporal markers.
Wandsnider (1989:430-436) reviews numerous ethnographic accounts of the
reuse of prehistoric arrowpoints by the Paiute, Ute, Yuma, Papago, Seri, Apache,
Navajo, Taos Pueblo, and several other historic aboriginal groups in the arid west
of North America. For example, Isabel Kelly (1934:141) notes that among the
Surprise Valley Paiute, “Arrowpoints found archaeologically were used if in good
condition.” She also mentions that among the Southern Paiute, “manufactured
points found archaeologically also [were] used” (Kelly 1964:75). Smith (1974:11)
reports among the Northern Ute, “Old arrow points, discovered when the people
were roaming, were picked up, sharpened, and used.” Concerning the Jicarilla
Apache, Opler (1946:84) states that, “Whenever a group camps near a site formerly
occupied by Pueblo Indians or other aliens, the children are sent out to look for flint
arrowheads.” The Honey Lake Paiute (Wadatkuht) often collected obsidian flakes
and debris from archaeological sites:
The Wadatkuht got their obsidian from ‘Flint Mountain,’ Dakakudak, a hill near
Gerlach, Nevada. Obsidian also was obtained at Pagushuhad, a village and
archaeological site on the east side of Honey Lake. It also was gotten at one of the
hot springs near the lake presumably from an archaeological site. That is to say, the
Wadatkuht picked up chips and nodules left by previous people as no natural
outcrop of obsidian is recorded for that region (Riddell 1960:50, emphasis added).
Similar behavior is reported among the Northern Paiute (Fowler 1992:106-109) and
the Western Apache who often scavenged Pueblo sites for flakes and pieces of
debris suitable for arrowhead manufacture:
The old men used to go around to ruins and pick up pieces of white flint there until
they had enough to fill a small buckskin sack. Then when they got ready to make
arrow points, they laid a blanket down and on this spread out their pieces of white
flint. Then they picked whichever one they wanted to work on (Basso et al.
1971:231).
Figure 12-2 Shaded relief map of the USA showing primary geographic regions
and site locations discussed in this paper (base map from Thelin and Pike 1991).
dated between 10,000 and 9,000 radiocarbon years BP at several stratified sites
throughout the southeastern US (Chapman 1985). However, these Early Holocene
diagnostics were co-mingled within an archaeological horizon that also contained
several stemmed projectile points classified as members of the Sykes-White
Springs Cluster known to date 3,000 radiocarbon years younger (Amick 1985).
Initially, this unexpected finding forced questions about security of the radiocarbon
age for these two temporal markers as well as possible mixing within the
archaeological deposits.
Extensive geomorphological investigations at this site and the surrounding
region indicated that the overbank sedimentation was relatively gentle with several
periods of surface stability and soil formation during the Holocene (Brakenridge
230 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
1984). However, there was no evidence to suggest that regional floodplain surfaces
could have lasted for more than several centuries during the Mid-Holocene.
Furthermore, alluvial deposits of Early Holocene age that could have contained
Kirk occupations were not known in the region. All the dated alluvial sequences in
the region begin around 7,000 radiocarbon years BP, although it is possible that
earlier deposits are deeper than backhoe testing was capable of exposing (limited to
about 250 cm). Comprehensive artifact refitting at a nearby mid-Holocene site
within the same alluvial regime has shown that archaeological materials can be
vertically dispersed about 30 cm through post-depositional movement but that
mixing of discrete occupational surfaces was probably limited to several centuries
at most (Hofman 1992).
Radiocarbon dating of charred hickory nuts from the Cedar Creek Site
produced an age of 6,375±215 BP (GX-8822), consistent with expectations for the
Sykes-White Springs artifacts. How could these Kirk artifacts become incorporated
into an archaeological deposit that was three thousand years younger? Fortunately,
our initial confusion was easily resolved because these Kirk points had been
patinated then retrieved and resharpened by later artifact scavengers revealing
younger marginal flaking over the patinated flake scars (often termed “repatination”
or what McDonald [1991] called “double patina”). Apparently, mid-Holocene
individuals must have collected these finished Kirk tools from exposed and
weathered archaeological deposits and reused them.
Using evidence of double patination, flake and flake scar morphology, and
flake refitting, Sassaman and Brooks (1990; also see Sassaman 1994:104) have
proposed considerable amounts of lithic recycling associated with Early Woodland
peoples (c. 2,500-3,000 BP) scavenging lithic debris from Archaic sites (c. 4,500-
9,500 BP) in South Carolina. Relatively heavy vegetation covers most ground
surfaces of the southeastern US, which probably limited most prehistoric
scavenging opportunities. Secondary lithic recycling does not appear to be
widespread or easily recognized in densely vegetated, depositional environments,
like the eastern US, but it seems to have been encouraged by periods of localized
surface erosion and reductions in the residential mobility of prehistoric peoples
living in areas with poor lithic resources (Amick 1987; Amick and Carr 1996;
Sassaman and Brooks 1990). Procurement of these finished tools probably
represented a substantial savings of energy in procurement and manufacture,
although this limited evidence suggests that secondary recycling was largely
opportunistic in these cases.
Although alternative materials for the production of stone tools occur in the
area, local obsidian sources were often preferred and served as important locations
in the context of general prehistoric land use. The reuse of these obsidian scatters
over several thousands of years resulted in a complex landscape of debris from
overlapping occupations and activities. This palimpsest pattern associated with
lithic recycling is not uncommon in North American desert regions (e.g., Bettinger
1989: 331-333; Camilli 1983, 1988a; Camilli and Ebert 1992; Kelly 1988a, 1988b,
2001; Wandsnider 1989) and appears to be particularly intense at the NTS obsidian
source areas. Obsidian procurement and reduction produced many redundant
byproducts that are ubiquitous throughout these surficial workshop areas.
232 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
Obsidian Source
Point Age Local Non-local Total
Early Holocene O=41, E=45 O=15, E=11 56
Late Holocene O=42, E=38 O=5, E=9 47
Total 83 20 103
Chi-square = 4.258, df = 1, p < .039062
Table 12-2 Obsidian sourcing results on a sample of projectile points from the NTS
in southern Nevada. Nonlocal sources are defined as those greater than 200 km
away including: Kane Spring, Brown’s Bench, Montezuma Range, Coso Volcanic
Field, and Fish Springs. Observed values (O) and expected values (E) are listed.
Figure 12-4 Obsidian biface fragments from Dead Horse Flat (26Ny4203) that
have been fractured radially (top row) and bipolarly (bottom two rows) during lithic
recycling. Smoked with ammonium chloride for photo enhancement.
Despite a general preference for obsidian in this region, it is possible that Late
Holocene inhabitants at the NTS showed less frequent use of local obsidian because
these resources had been degraded during the Early Holocene. In fact, obsidian
source locations associated with the desert pavements along Fortymile Wash
contain the densest concentrations of Early Holocene artifacts. Many land surfaces
in the southern Great Basin are distinguished by long-term stability for several tens
of millennia (Davis 1983), which has three important impacts on the character of
the archaeological record. First, these stable surfaces typically produce a palimpsest
record of multiple occupations. Second, shallow lithic resources are subject to
234 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
Archaeological studies at two large lithic procurement sites on the NTS have
provided notable evidence of the depletion of obsidian and chalcedony, two of the
locally preferred tool stones. At the Midway Valley chalcedony quarry
(26Ny4759), lithic resource depletion is indicated by evidence from lithic refitting,
site spatial analysis, and the scarcity of unmodified tool stone (Buck et al. 1994).
This surface site served primarily as a procurement station for large chalcedony
blocks, but 24 clusters of obsidian flaking debris were also collected from the
reduction of small, rounded nodules. Although all unmodified tool stone was
collected from the site during archaeological recovery, only two small nodules of
obsidian were recorded. This site illustrates severe prehistoric exploitation of a very
dispersed scatter of small obsidian nodules. The largest chalcedony also appears to
have been depleted because the unmodified chalcedony blocks remaining at the site
were usually smaller than the discarded cores.
At Buckboard Mesa (26Ny4892), there was a denser surface scatter of small,
subrounded obsidian nodules with several lithic workshops primarily associated
with initial through intermediate stages of core reduction (Amick et al. 1991).
Daniel S. Amick 235
Figure 12-5 Recycled artifacts from Buckboard Mesa (26Ny4892) illustrating later
flake production on older discarded cores. Note characteristic scar resulting from
splitting technique using direct, freehand hard hammer percussion on Reference
#264-14. Drawn by Sue Ann Monteleone.
Daniel S. Amick 237
Figure 12-6 Recycled artifacts from Buckboard Mesa (26Ny4892) illustrating older
artifacts for small biface production. Reference #556-1 is a broken Late Holocene
(Gatecliff) projectile point manufactured on an older weathered artifact. Drawn by
Sue Ann Monteleone.
Tsong 1980) and has been used to suggest patterns of artifact recycling at various
locations in western North America (Batcho 1984; Earls et al. 1989; Jackson 1985;
Kaufmann 1984; Raymond 1985). These efforts have usually focused on the
demonstration of chronological anomalies (deposits containing artifacts with a
238 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
Figure 12-7 Scavenged and recycled Great Basin Stemmed (Early Holocene)
projectile point (Reference #1428) from Midway Valley (26Ny4759). Stem exhibits
recent flake removed during the recycling effort. Percussion removal of this flake
probably caused the artifact to simultaneously break transversely because of
bending failure. Proximal and distal fragments can be refit along the surface of this
unweathered break.
Knife River Flint quarries by Root et al. (1999) indicate at least 20% show contact
wear on tips and edges.
On the other hand, there has been considerable debate about the interpretation
of bipolar recycling in Clovis assemblages. Some argue that certain bipolar
artifacts, sometimes called pièces esquillées, are tools used for slotting and wedging
in bone and wood (Lothrop and Gramly 1982; MacDonald 1968). Recently,
Bradley and Frison (1996:62-64) identified two pièces esquillées at Mill Iron, a
Goshen Complex bison kill in southeastern Montana that is contemporaneous with
late Clovis occupations on the High Plains. One is described as a tool with a
heavily battered poll that was split during use sustaining interior damage, which
they believe resulted because one piece was rubbing against the other while tightly
wedged in bone or wood. The second tool has a sharp point formed on one end,
while repeated blows to the opposing end produced several flakes (recovered and
refitted) during use. The bone bed context for these items is used to support
interpretation as tools, presumably during the butchery and processing of the bison
carcasses. However, use-wear examinations were not conducted on these purported
tools and contemporary replication work suggests that similar patterns of damage
can be produced on bipolar cores (Callahan 1987; Flenniken 1981).
Some argue such items in Paleoindian assemblages are simply bipolar cores
used as an economizing measure during states of toolkit exhaustion resulting from
high rates of residential mobility (Goodyear 1993; Shott 1989, 1999). In his
examination of eight eastern Clovis assemblages, Goodyear (1993) proposes that
pièces esquillées are best interpreted as bipolar cores because their frequency is
inversely related to the frequency of “potential cores.” He argues the lack of
“potential cores” indicates depletion of lithic supplies resulting in increased
frequencies of recycling through bipolar reduction. Distance from the lithic source
can provide an effective alternative for measuring the potential depletion of lithic
supplies. Comparison of pièces esquillée frequencies (from original published
reports) against distance to primary lithic source from a sample of 35 eastern Clovis
assemblages fails to show any clear patterning regarding distances from lithic
sources (after Meltzer 1989:Table 2.2). Low sample sizes and variable recovery
techniques limit effective comparisons but this generally poor relationship is well
illustrated by examining the three eastern Clovis sites with the highest frequencies
of pièces esquillées. There are 70+ at Shoop (Cox 1986:125), located 320 km from
its primary lithic source; there are 1,046 at Debert (MacDonald 1968:85), located
100 km from its primary lithic source; and there are 567 at Vail (Lothrop and
Gramly 1982), located only 25 km from its primary lithic source.
These substantial cases seem to contradict Goodyear’s “toolkit entropy”
hypothesis where demand on available raw material should increase with distance
from source, resulting in the greater frequency of tool stone economizing measures
like bipolar core reduction. In fact, significant numbers of pièces esquillées have
242 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
Figure 12-8 Exhausted Clovis (Gainey) projectile point (MH-9) from the Morrow-
Hensel site (far western Wisconsin) that shows laterally recycling with transverse
burin blows (bipolar) on lateral margins at the bending fracture on distal end.
Drawn by Sarah Moore.
dwindling. This assemblage also exhibits bipolar recycling of broken and exhausted
tools (Figures 12-8 and 12-9) as well as the bipolar reduction of locally available
small chert pebbles for flake blanks (Figure 12-9). Evidence of bipolar reduction
presents considerable ambiguity in this case where it seems to represent the “toolkit
entropy” of exhausted tools, production of pièces esquillées as wedges, and an
effective means of reducing the locally available small chert pebbles. To complicate
matters, the lithic assemblage from the Gainey type site in central Michigan is
dominated by Upper Mercer cherts from sources 380 km distant, yet this
assemblage shows little or no signs of lithic recycling or tool stone conservation
(Simons et al. 1984).
Figure 12-9 Variety of bipolar artifacts from Morrow-Hensel. Top, left to right:
Hixton pièces esquillée, Hixton bipolar core made from recycled biface, bipolar
core on Prairie du Chien chert pebble. Bottom, right: bipolar flake from Prairie du
Chien chert pebble.
In many cases, the high mobility of Early Paleoindian groups placed stress on
the amount of available raw material within the toolkit, requiring groups to employ
economizing strategies of lithic manufacture like the bipolar reduction of waste and
exhausted tools. Conversely, many cases are documented where tool stone
availability is not a constraint but bipolar splintered pieces occur (sometimes
abundantly) in the toolkit discards. These contradictions may suggest that pièces
esquillées served as wedges for splitting bone and wood in some Paleoindian
assemblages. The particular association of these artifacts with Gainey occupations
244 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
of the Great Lakes and Northeast implies a general functional association caribou-
hunting lifeways and high mobility.
Conclusions
Several general conclusions can be drawn from this review of lithic recycling.
Scavenging and reworking of discarded tools and debris as cores for tool and blank
production is very efficient economically although the resulting products may have
limited utility because of their small sizes. Generally, there seems to be an inverse
relationship between the access to lithic resources and the amount of lithic
recycling that occurs. However, there seem to be multiple and contradictory factors
that can limit access to lithic resources. Extraordinarily high levels of residential
mobility are argued to limit access to lithic resources among the various
Paleoindian groups
who practiced systematic lateral recycling. Conversely, low levels of residential
mobility are argued to limit access to tool stone resources and increase lithic
recycling among many Late Holocene groups in the Great Basin (and elsewhere)
who settle into lithic-poor locations. Because of this contradiction, it is clear that
the identification of lithic recycling alone is not capable of supporting inferences
about prehistoric group mobility.
Although these unresolved contradictions exist, residential mobility is
generally believed to strongly influence lithic recycling behavior in pre-industrial
societies. Alternative archaeological explanations for recycling often emphasize its
economic necessity during periods of scarcity, similar to the “waste not, want not”
approach of social and historical explanations for recycling behavior in modern
American industrial society (e.g., Strasser 1999). However, these assumptions
about the behavioral role of recycling remain inadequately supported in prehistoric
lithic technologies. Deliberate ethnohistorical and ethnoarchaeological research is
sorely needed to investigate the situational and organizational context of recycling
behavior in small scale pre-industrial societies and stone age economies (e.g.,
Kuznar 1995:116).
Additional problems are presented by lithic recycling behavior because of the
many different forms it can take. Many archaeologists rely on bipolar cores and
debris as the primary means of identifying lithic recycling despite the analytical
difficulty in distinguishing the lithic debris from bipolar reduction (Jeske and Lurie
1993). Furthermore, bipolar technology is frequently associated with the reduction
of small pebbles and cores and may not necessarily reflect the lithic recycling of
discarded waste. At the Morrow-Hensel site, Clovis peoples apparently used
bipolar reduction to exploit local chert pebbles as well as for recycling their
exhausted tools made of transported tool stone and the manufacture of wedges
(typically classified as pièces esquillées). Technological analysis of obsidian
Daniel S. Amick 245
exploitation at the NTS shows that direct freehand, hard hammer percussion rather
than bipolar percussion was the primary method used in lithic recycling.
Opportunistic scavenging and resharpening of projectile points from exposed
archaeological sites seems to have been the primary kind of lithic recycling
behavior practiced at the Cedar Creek site.
Evidence of scavenging at the NTS obsidian sources is also significant
because it contradicts Schiffer’s (1987:27-28) suggestion that recycling should be
unexpected at quarries: "Scarcely any reuse can be discerned in a lithic quarry-
workshop; as a result, the archaeological record contains the bountiful traces of
virtually every knapping act that took place." The analysis presented here
demonstrates that assumptions about the lack of recycling at lithic quarry-workshop
locations are unreliable. In fact, Early Holocene overexploitation of these obsidian
resources may have accelerated Late Holocene scavenging behavior at these
abandoned quarries.
It is important that archaeologists seek out and attempt to define and explain
the kind of observational and interpretive ambiguity demonstrated by lithic
recycling. But unless double patina or other unambiguous evidence of sequential
flaking (such as differences in obsidian hydration band thicknesses) can be
identified, it is often difficult to distinguish lithic recycling with confidence.
Consequently, recycled artifacts that do not exhibit these characteristics can be
easily overlooked. Because of these analytical obstacles, it likely that prehistoric
lithic recycling was much more frequent than most archaeologists recognize.
Identification of lithic recycling is important for rigorous interpretation of
archaeological chronology and the formation processes affecting sites and
assemblages. In addition, thoughtful recognition and understanding of lithic
recycling is important for accurate reconstructions of this prehistoric behavior,
which contributes to many current explanatory models of mobility and the
organization of technology. Finally, it is important to develop skills and reliable
methods to identify material recycling because processual approaches to lithic
analysis require a focus on understanding the dynamic life histories of artifacts and
assemblages.
Acknowledgments
Thanks to the patient encouragement of Shannon McPherron and John Lindly
who invited this paper and supported my prolonged efforts to finish the manuscript.
During the past several years, I am very grateful for discussions and comments on
these ideas from Peter Ainsworth, Bob Kelly, John Fagan, Ted Hartwell, Jack
Hofman, Eric Ingbar, Steve Kuhn, Dave Rhode, Mike Rondeau, Ken Sassaman and
Betsy Skinner - each of whom forced me to clarify my thinking on this thorny
topic. Many of the archaeological materials reported in this study resulted from data
246 Investigating the Behavioral Causes and Archaeological Effects
of Lithic Recycling
References Cited
Amick, D. S. 1985. Buried Late Holocene Terrace Site Testing in the Central Duck
River Basin, K. Guthe, T. Whyte, C. Boyd, and B. Riggs (eds.) Exploring
Tennessee Prehistory: A Dedication to Alfred, 23-38. Report of Investigations
No. 42. Department of Anthropology, University of Tennessee, Knoxville.
—. 1987. Lithic Raw Material Interassemblage Variability in the Central Duck
River Basin: Reflections of Middle and Late Archaic Organizational
Strategies. TVA Publications in Anthropology No. 50. Report of
Investigations No. 46. Department of Anthropology, University of Tennessee,
Knoxville.
—. 1990. Lithic Reduction and Artifact Recycling Strategies at an Obsidian Nodule
Source in Southern Nevada. Paper presented at the 22rd meeting of the Great
Basin Anthropological Conference, Reno.
—. 1992. Fire on the Mesa: Archaeological Investigations at the U19an Borrow
Pit on the Nevada Test Site. Technical Report No. 72. Quaternary Sciences
Center, Desert Research Institute, Reno.
—. 1996. Regional Patterns of Folsom Mobility and Land Use in the American
Southwest, World Archaeology 27: 411-426.
—. 1999a. Using Lithic Artifacts to Explain Past Behavior, in C. Beck (ed.) Models
for the Millennium: Great Basin Anthropology Today, 161-170. Salt Lake
City: University of Utah Press.
—. 1999b. New Approaches to Understanding Folsom Technology, in D. S. Amick
(ed.) Folsom Lithic Technology: Explorations in Structure and Variation, 1-
11. Archaeological Series 12. Ann Arbor, Michigan: International
Monographs in Prehistory.
Amick, D. S., R. Boszhardt, K. Hensel, M. Hill, T. Loebel, and D. Wilder. 1999.
Pure Paleo in Western Wisconsin. Report of Investigations No. 350.
Mississippi Valley Archaeology Center, University of Wisconsin, La Crosse.
Daniel S. Amick 247
—. 1988a. Lithic Raw Material Selection and Use in the Desert Basins of South-
Central New Mexico, The Kiva 53: 147-163.
—. 1988b. Interpreting Long-Term Land-Use Patterns from Archaeological
Landscapes, American Archeology 7: 57-65.
Camilli, E. L. and J. I. Ebert. 1992. Artifact Reuse and Recycling in Continuous
Surface Distributions and Implications for Interpreting Land Use Patterns, in
J. Rossignol and L. Wandsnider (eds.) Time, Space, and Archaeological
Landscapes, 113-136. New York: Plenum Press.
Chapman, J. J. 1985. Archaeology and the Archaic Period in the Southern Ridge-
and-Valley Province, R. S. Dickens and H. T. Ward (eds.) Structure and
Process in Southeastern Archaeology, 137-153. Tuscaloosa: University of
Alabama Press.
Close, A. E. 1996. Carry That Weight: The Use and Transportation of Stone Tools,
Current Anthropology 37: 545-553.
Collins, M. B. 1975. Lithic Technology as a Means of Processual Inference, E.
Swanson (ed.) Lithic Technology: Making and Using Stone Tools, 15-34. The
Hague: Mouton.
Cox, S. L. 1986. A Re-analysis of the Shoop Site, Archaeology Eastern North
America 14: 101-170.
Davis, J. O. 1983. Geological Reconnaissance and Chronologic Studies. Technical
Report No. 33. Quaternary Sciences Center, Desert Research Institute, Reno.
Dibble, H. L. 1991. Local Raw Material Exploitation and Its Effects on Lower and
Middle Paleolithic Assemblage Variability, A. Montet-White and S. Holen
(eds.) Raw Material Economies Among Prehistoric Hunter-Gatherers, 49-58.
Publications in Anthropology No. 19. Department of Anthropology,
University of Kansas, Lawrence.
Earls, A., C. R. Lintz, and W. N. Trierweiler. 1989. Chronology, J. B. Bertram et
al. (eds.) Report of Surface Collection and Testing at 18 Sites Near Abiquiu
Reservoir, Northern New Mexico, 305-357. Mariah Associates, Inc. Submitted
to U.S. Army Corps of Engineers, Albuquerque District. Copies available
from Mariah Associates, Inc., Albuquerque.
Ebert, J. F. 1986. Distributional Archaeology: Nonsite Discovery, Recording and
Analytical Methods for Application to the Surface Archaeological Record.
Ph.D. dissertation, Department of Anthropology, University of New Mexico,
Albuquerque.
Elston, R. G. 1992. Modeling the Economics and Organization of Lithic
Procurement, R. G. Elston and C. Raven (eds.) Archaeological Investigations
at Tosawihi, A Great Basin Quarry, Part 1: The Periphery, Vol. 1, 31-47.
Report prepared by Intermountain Research, Silver City, Nevada. Report
prepared for Bureau of Land Management, Elko, Nevada.
Daniel S. Amick 249