Ochre at 49-pet-408

Material Information

Ochre at 49-pet-408
Mrzlack, Heather
Publication Date:
Physical Description:
153 leaves : illustrations ; 28 cm


Subjects / Keywords:
Ocher ( lcsh )
On Your Knees Cave Site (Alaska) -- Antiquities ( lcsh )
Excavations (Archaeology) -- Alaska -- On Your Knees Cave Site ( lcsh )
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )


Includes bibliographical references (leaves 140-153).
General Note:
Department of Anthropology
Statement of Responsibility:
by Heather Mrzlack.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
55626368 ( OCLC )
LD1190.L43 2003m M79 ( lcc )

Full Text
Heather Mrzlack
B.A., Purdue University, 1996
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts

This thesis for the Master of Arts
degree by
Heather Mrzlack
has been approved
Chris Beekman

Mrzlack, Heather (M.A., Anthropology)
Ochre at 49-PET-408
Thesis directed by Professor Tammy Stone
Examination of ochre has the potential to provide archaeologists with
information concerning ancient procurement activities, trade networks, and
interaction patterns. Geochemical techniques such as particle-induced X-ray
emission, or PIXE, have been applied to ochre to trace individual source
deposits. By studying ochre found during excavations at 49-PET-408 (9,200
years BP), a multiple component cave site located in southeast Alaska, this
thesis expands the current body of knowledge regarding what materials
researchers can use from the archaeological record to understand ancient trade
and interaction.
The analysis of ochre from 49-PET-408 included macroscopic and low-
power microscopic analysis to determine usewear, and PIXE to determine the
geochemistry of the individual ochre specimens. The results from the PIXE
analysis were used to determine whether each specimen contained the same
geochemical signature. Results denoting the same geochemical signatures are
strong evidence suggesting that the ancient people who utilized 49-PET-408

obtained the ochre from one distinct source either through direct or embedded
procurement, or trade. Different chemical signatures would suggest the ochre
came from different source areas indicating people were procuring the ochre
from more than one source. In addition to the inter-site comparison of ochre,
this study statistically compares the PIXE results from 49-PET-408 to the
chemical signatures from eight ochre sources in Western North America to
determine if any of the specimens from 49-PET-408 match any of the eight
source areas.
The results from the study indicate that inhabitants of the site were
obtaining ochre from geologically similar sources, or sources which are close
together. This could be a result of the use of only one quarry. The results
combining the 49-PET-408 data and the data from the eight ochre sources in
Western North America show that nine specimens from 49-PET-408 match the
geochemical composition of two sources from Oregon. This may suggest that
people from 49-PET-408 were obtaining the ochre from Oregon.
Tracing discrete source deposits of ochre through geochemical analysis
can provide archaeologists with yet another line of evidence to reconstruct trade
and procurement activities. More ochre sources need to be identified and
geochemically analyzed to increase the database for comparisons.

This abstract accurately represents the content of the candidate's thesis. I
recommend its publication.

I dedicate this thesis to my husband, David, for his constructive criticism,
editing prowess, advice, and humor throughout this challenging process.
Without him, this would not have been accomplished.

My thanks to my advisor, Tammy Stone, for her tireless effort that demands the
highest academic standard. I also wish to thank Jim Dixon, for letting me work
with him at 49-PET-408, for his financial support for the ochre analysis, and his
guidance. Eric Parrish helped tremendously with photographs and maps-
thanks, Eric. Thanks also to Terry Fifield, for help with the background
information concerning NAGPRA issues. Special thanks to Jon Kent, my first
inspiration in archaeology. Finally, thanks to all those at RMC Consultants,
Inc.; their support and understanding has been incredible.

1. INTRODUCTION................................................1
2. RESOURCE PROCUREMENT.......................................9
Resource Distribution..................................15
Procurement Methods....................................19
Embedded Procurement..............................20
Direct Procurement................................23
Trade........................................... 26
Ochre Procurement......................................31
Conclusion........................................... 33
3. HISTORY OF OCHRE STUDIES..................................36
Utilitarian Uses of Ochre..............................38
Geochemical Analyses of Ochre..........................44
Symbolic and Ritualistic Uses of Ochre.................52

Environmental History of Bedngia
and the Northwest Coast..............................58
Models of New World Migrations.......................60
Prehistory of the Northwest Coast................... 67
Sites of the Northwest Coast Microblade Tradition..........69
Groundhog Bay II............................ 70
Hidden Falls................................. 72
5. METHODS AND RESULTS ....................................83
Specimen Recovery................................ 83
Macroscopic Ochre Analysis...........................86
Low-Power Microscopy.................................89
Macroscopic and Microscopic Results..................90
Specimen I: AN-2000-29.26 .....................90
Specimen 2: AN-2000-29.682................... 90
Specimen 3: AN-2000-29.71......................92
Specimen* AN-20G0-29.881..................!....93
Specimen5: AN-1999-49.299 .....................94

Specimen 6: AN-1999-49.388....................94
Specimen 7: AN-1999-49.533....................95
Specimen 8: AN-1999-49.668....................97
Specimen 9: AN-1999-49.669 ...................98
Specimen 10: AN-1999-49.670 ..................99
Specimen 11: AN-1998-73.180..................100
Specimen 12: AN-1998-73.228..................101
Specimen 13: AN-1998-73.486..................102
Specimen 14: AN-1998-73.496..................103
Specimen 15: AN-1998-73.499..................104
Specimen 16: AN-1998-73.606..................105
Specimen 17: AN-1997-134.60..................105
Particle Induced X-Ray Emission (PIXE).............106
Statistical Analysis...............................Ill
6. CONCLUSION.............................................122
A. Elemental Composition (PPM) of Ochre Specimens.........131

2.1 Map of Southeast Alaska and Locations
of 49-PET-408, Groundhog Bay, and Hidden Falls...............29
4.1 Map of Southeast Alaska and Locations
of 49-PET-408, Groundhog Bay, and Hidden Falls...............71
4.2 49-PET-408 Excavation Units and Stratigraphy...................78
4.3 Map of 49-PET-408 and Obsidian Sources
Mount Edziza and Suemez Island................................81
5.1 Provenience Location of Ochre
Specimens from 49-PET-408................................... 88
5.2 Specimen 1. AN-2000-29.269....:.,.............................92
5.3 Specimen 2. AN-2000-29.682................................. 92
5.4 Specimen 3. AN-2000-29.714 93
5.5 Specimen 4. AN-2000-29.881....................................93
5.6 Specimens. AN-1999-49.299.................................. 95
5.7 Specimen 6. AN-1999-49.388 ...................................95
5.8 Specimen 7. AN-1999-49.533 ...................................97
5.9 Specimen 7: Showing the Small Drill Holes
in the Un-worked Surface of the Artifact.....................97
5.10 Specimens. AN-1999-49.668 ....................................99

5.11 Specimen 9. AN-1999-49.669 ...............................99
5.12 Specimen 10. AN-1999-49.670..............................101
5.13 Specimen 11. AN-1998^73.180..............................101
5.14 Specimen 12. AN-1998-73.228..............................103
5.15 Specimen 13. AN-1998-73.486..............................103
5.16 Specimen 14. AN-1998-73.496..............................104
5.17 Specimen 15. AN-1998-73.499..............................104
5.18 Specimen 16. AN-1998-73.606..............................106
5.19 Specimen 17. AN-1997-134.60..............................106
5.20 Factor Analysis of 49-PET-408 Specimens...................115
5.21 Erlandson et al.'s Source Area Specimens..................117
5.22 Cluster Analysis of 49-PET-408 Specimens..................118
5.23 Cluster Analysis Combining Nine Specimens
from 49-PET-408 with Erlandson et al.'s
Source Area Specimens ....................................119
6.1 Map of 49-PET-408 and Location ofMinam Grade
and Blue Mountains Oregon Ochre Sources...................128

5.1 Range (Parts Per Million and Percentage)
of Each Element in 49-PET-4Q8 Specimens

The purpose of this research is to answer questions regarding
procurement and use of non-subsistence items by mobile hunter-gatherers.
Traditionally, archaeologists have used trace element analysis of tool stones,
such as obsidian, to reconstruct past exchange and interaction patterns. This
thesis expands the current body of knowledge regarding what materials
researchers can use from the archaeological record to understand ancient trade
and interaction.
We can look at this question by examining ochre found during
excavations at 49-PET-408 (9,200 years BP), a multiple component cave site
located in southeast Alaska. Red ochres were highly valued and traded over
long distances at least since the appearance of anatomically modem humans. If
ochre found from specific archaeological sites was traceable to the original
ochre source, could this possibly allow researchers to reconstruct ancient trade
and travel routes? Does ochre use fulfill a critical role of reinforcing social
relations through the material expression of meaning? What significance did
this material have for very early prehistoric inhabitants of southeast Alaska?
Examination of ochre has the potential to provide archaeologists with
information concerning ancient procurement activities, trade networks, and

interaction patterns. Traditionally, researchers have paid little scholarly
attention to this ubiquitous pigment found in various ancient and historic sites
worldwide, (barring some excellent studies by David et al. 1993; Erlandson et
al. 1999; Sagona 1994; and Tankersley et al. 1995, to name a few.) Recently,
however, geochemical techniques such as particle-induced X-ray emission, or
PIXE, and X-ray diffraction, have been applied to ochre to trace discrete source
deposits for the purpose of reconstructing trade and procurement activities
(David et al. 1993; Eiiandson et al. 1999; Tankersley et al. 1995).
Red ochre is an impure variety of the mineral hematite, one of the most
common minerals in the world (Tankersley et al. 1995). Hematite refers to
the vivid, blood red color produced when the mineral is in powder-form
(Tankersley et al. 1995). The color and luster of hematite varies from specular,
black, gray, and silver-gray to a non-specular dull earthy red, brown, reddish-
brown, and cherry-red (Chesterman 1978). When heated, the color of ochre
intensifies and darkens (Velo and Kehoe 1990). While most forms of hematite
occur as sedimentary deposits, hematite has also formed in igneous and
metamorphic rocks, resulting in its widespread availability (Tankersley et al.
This ochre study is based on excavations conducted at 49-PET-408
during the summers of 1997 through 2000 (Dixon et al. 1997). In addition to

numerous microblades, bifaces, cores, and flakes, investigations recovered
seventeen specimens of ochre weighing between. 1 gram and over 300 grams,
one obsidian biface with possible ochre residue embedded in the flake scars,
and a large basin-shaped siltstone palette with apparent ochre stains on the
concave surface. One of the ochre specimens appears to have cultural
modifications, with striations across one facet. Several of the specimens range
in color from red and reddish brown, while four of the specimens are orange, or
reddish-yellow, and brownish-yellow. This may indicate heat treatment
because ochre does not occur naturally orange. The ochre users could have
incompletely fired yellow limonite or goethite to achieve the orange ochre.
I conducted PIXE analysis on the ochre from 49-PET-408 to identify the
trace elements of the specimens: If the results denoted the same geochemical
signatures in these specimens, it would be strong evidence suggesting that the
occupants of the site obtained the ochre from one distinct source either through
embedded procurement (the material is passively acquired while conducting
other transactions, such as hunting), direct procurement (trips are conducted for
tire sole purpose of obtaining the material), or trade. Different chemical
signatures, on the other hand, would indicate that the ochre was likely being
procured from different sources.
In 1999, Erlandson et aL conducted PIXE analysis on eight ochre

sources from Western North America, distinguishing twenty-one major and
minor trace elements among the specimens. The ochre sources included Gold
Harbor in southeast Alaska, Red Rock, Sulphur Bank, and Tajiguas, in
California, Minam Grade, China Creek, and Blue Mountains in Oregon, and
Sunrise Mine in Wyoming. Statistically analyzing the PIXE quantitative data
with an RQ-mode principal components analysis, the researchers found that
there are geochemical differences in the specimens that indicate researchers
may be able to separate out source areas in archaeological specimens.
This study statistically compares the PIXE results from 49-PET-408 to
the results from Erlandson et al. s eight sources. Do any of the specimens from
49-PET-408 match any of Erlandson et al.s sources? Could a positive result
imply that inhabitants from 49-PET-408 were indeed procuring ochre from one,
or more, of the eight source areas? Could an inter-site comparison of ochre
from 49-PET-408 show that the specimens were obtained from one source area,
or many?
Geochemically sourcing ten ochre specimens from 49-PET-408, the
results suggest that the occupants of the site likely procured the ochre from one
source. After statistically comparing the results to the chemical signatures of
eight known ochre sources (three California, three Oregon, one Alaska and one
Wyoming source), I found that nine of the ten specimens from 49-PET-408

display similar chemical compositions to two of the three ochre sources in
Oregon. Conducting geochemical techniques on ochre found in archaeological
sites provides a way for researchers to trace discrete source deposits. This, in
turn, can be used to understand such important subjects as exchange networks
and procurement activities.
. The following is a brief introduction to each chapter of this thesis. In
Chapter 2, "Resource Procurement," I discuss the processes of raw material
acquisition observed in the archaeological record. Included is a discussion of
three procurement practices: direct procurement, embedded procurement, and
exchange. Although the procurement studies reviewed are based on lithic raw
materials used to make stone tools,.my analysis explains how these models can
be used to understand the procurement of other important materials, such as
Chapter 3, "The History of Ochre Studies," reviews previous studies
conducted regarding ochre. I divide ochre research into three main groups:
studies based on utilitarian uses of ochre, studies of symbolic uses of ochre, and
analyses based on geochemical sourcing of ochre. Utilitarian uses include
medicinal uses, as a pigment, as a preservative, and as an abrasive for polishing
bone and ivory. Symbolic studies of ochre include examinations of the
symbolic meanings prehistoric people may have placed on ochre or the color

red (Foster andBotscharow 1990; Marshack 1981; Wreschner 1980). Research
based on geochemical sourcing of ochre includes identifying trace elements of
ochre to distinguish among different ochre sources (Erlandsoh et al. 1999).
After closely reviewing the previous studies, I found that examination of ochre
has provided archaeologists with important information concerning mortuary
practices, art, ancient procurement activities, trade networks, and interaction
In Chapter 4, "Archaeological and Environmental Background," I
discuss the environmental history of Beringia and the Northwest Coast of
Alaska from 25,000 years ago BP, until after the Pleistocene/Holocene
transition, 9,000 years ago BP. Three different theories for the peopling of the
New World are analyzed: the overkill hypothesis (Mossiman and Martin
1975), the three-migration theory (Turner 1992), and the coastal migration
model (Dixon 1993, 1999;Fladmark 1979; Josenhans etal. 1997). A prehistory
of the Northwest Coast is presented with focus on one of the three regional
variants of the American Paleoarctic tradition: the Northwest Coast Microblade
tradition. A brief discussion of two early sites of the Northwest Microblade
tradition, Groundhog Bay n and Hidden Falls, follows. I end the chapter with a
description of 49-PET-408, another site of the Northwest Coast Microblade
tradition, and the main focus of this thesis.

Methods of conducting the analysis of this thesis and the results of this
research are discussed in Chapter 5, "Methods and Results." Included is a
description of the excavations conducted at 49-PET-408, followed by a
discussion of the macroscopic and low-power microscopic analysis of the ochre
specimens, and the results of this analysis. Next, I introduce particle-induced
X-ray emission (PIXE) and discuss the analysis conducted on the ochre
specimens from 49-PET-408. A statistical analysis, including a principal
components factor analysis (PCFA) with varimax rotation and a cluster
analysis, follows. The results and an interpretation of the findings from the
statistical analyses conclude this chapter.
This thesis is an example of the ways in which researchers can use ochre
to deepen their understanding of archaeological sites. Tracing discrete source
deposits of ochre through geochemical analysis can provide archaeologists with
yet another line of evidence to reconstruct trade and procurement activities. If
inhabitants of 49-PET-408 in southeast Alaska were indeed procuring ochre
from such far away distances as Oregon, this implies several possible, although
tentative, conclusions. First, if the ochre was obtained directly or embedded in
other activities, such as maritime hunting, this suggests that people from 49-
PET-408 traversed an impressively large territory that included different
environments, both coastal and inland. Tins indicates a remarkable

understanding of different environments. Since the results suggest that the
occupants of the site likely procured the ochre from one source, this implies a
preference for ochre from one particular source. If obtained through exchange
networks, this denotes the existence of well-established trade relationships
between coastal and inland groups of people. Groups may have exchanged this
important material to establish and maintain social contact and reciprocity
This research could not have been possible without the analysis by
Eriandson et al. (1999). Although researchers are beginning to pay more
scholarly attention to this important material found in various ancient and
historic sites worldwide, geochemical analysis of ochre is still in its early
stages. In order for archaeologists to utilize ochre as another line of evidence
for understanding the archaeological record, geochemical analysis of the
material needs to be perfected. This would provide an effective tool as useful
as the trace element analysis of obsidian. Furthermore, more ochre sources
need to be identified and geochemically analyzed to increase the database for

The ways in which hunter/gatherers procure raw materials are affected
by a number of factors, two of which are mobility and resource distribution.
Within the constraints imposed by these two factors, material can be obtained in
three ways: direct procurement, embedded procurement, and through
exchange. This chapter explores how raw material distribution and acquisition
is tied to mobility, specifically during the Pleistocene/Holocene transition in
North America In archaeological research, procurement studies, which
examine the ways in which ancient peoples obtained important, life-sustaining
materials, have traditionally focused on the lithic raw material used for making
stone tools. I argue that procurement studies can, and should, be expanded to
include the acquisition of other material^ such as ochre. This chapter illustrates
how the procurement of ochre can be analyzed by reference to existing research
regarding the acquisition of lithic raw material.
Times of environmental changes, such as during the transition from the
Pleistocene to the Holocene that occurred in North America approximately
10,000 years ago, greatly impacted lifeways of hunter/gatherers. During this
transition, North America experienced immense environmental alterations,
including mammalian extinctions and changes in the composition of forests,

woodlands, and parklands south of the ice sheets (Pielou 1991).
Hunter/gatherers had to physically adapt, altering their mobility patterns,
subsistence patterns, social networks, and methods for procuring raw materials,
including trade patterns. Because site 49-PET-4Q8 (9,200 years BP) was
occupied during the Pleistocene/Holocene transition (Dixon et al. 1997), the
first step in understanding procurement of ochre found at 49-PET-408 is to
understand early mobility patterns in the area of study, and to focus on these
patterns during times of environmental changes.
Mobility refers to various strategies of movement and settlement in
relation to properties of the natural environment (Jones et al. 2003:8). Early
inhabitants of the New World were highly mobile foragers with a low
population density and a dependence, on hunting (although not exclusively on
mega-fauna) (Kelly and Todd 1988; Seeman 1994). Traveling frequently, they
behaved like foragers rather than collectors (Bettinger 1991). Collectors cache
resources in anticipation of future use and have discrete seasonal movement
with down-time (usually during the winter) when tools are made and
repaired, followed by intense hunting and gathering (Bettinger 1991). By
contrast, researchers attest that foragers, having a low population density and

being highly mobile, did not cache resources in anticipation of future use.
Frequent residential movement was a response to the stress of temporal and
spatial resource shortages in a changing terminal Pleistocene/Early Holocene
environment (Bettinger 1991). For example, collectors in the New World
would have to adjust their seasonal hunting locations as habitat for wild game
expanded, coinciding with ice sheet retreat. Because of the shifting
environment, no precise cache location could be predicted.
The difference between collectors and foragers is not always so distinct,
however (Seeman 1994). Most researchers now realize that terms like
"collectors" and "foragers" or "travelers" and "processors" are not dichotomous
states, but ends of continuums between which people move back and forth. The
decision when or when not to travel is often tied to the distribution of resources.
The changing and uncertain environment of the Late Pleistocene/Early
Holocene necessitated several adaptive responses regarding subsistence
strategies, tool technology, raw material acquisition, and social alliances.
While the followinganalysis describes early inhabitants of a site in the eastern
Midwest/lower Great Lakes region of North America, the circumstances were
similar to those in northwestern North America. During the terminal
Pleistocene, human population densities were low, and the hunting of large
mammals such as mammoth, bison, and elk was central to early Paleoindian

communities (Seeman 1994).1
Analyzing lithic raw material at the Nobles Pond site in northeast Ohio,
a large Paleoindian site with 11 distinct clusters of Paleoindian occupations,
Seeman (1994) discovered a stark contrast to the assumed characteristics of
early Paleoindian populations. Through refit analysis of utilized tools and
debitage and comparisons of unutilized flakes, Seeman (1994:281) discovered
that the Nobles Pond site was an aggregation of several small bands at a
particular point in time. The site was not recurrently visited as a special
purpose site. Aggregations of populations may. have occurred to obtain short-
term subsistence returns through communal efforts, because of the limited
number of suitable habitations, and to reinforce social alliances through social
interactions (Seeman 1994). Seeman argues further that the predominance of
curated tools in finished form made horn high-quality exotic material could be
the result of caching, or pooling the material for future use. Through caching of
material, which eases scheduling constraints during communal subsistence
activities, early inhabitants of Nobles Pond portrayed characteristics of
collectors rather than foragers. Furthermore, explained in detail below, the
1 It should be noted, however, that early Paleoindians did not rely exclusively on big game.
Kelly and Todd (1988) suggest that while Paleoindians hunted large fauna, such as mammoth,
bison, and elk, easily acquired foods such as berries, seeds, roots, or small mammals were not

acquisition of lithic raw material was not embedded in subsistence activities,
but rather a specialized and direct activity required by particular demands of
band aggregation (Seeman 1994). Seeman's analysis illustrates how
characteristics of early Paleoindians may differ depending on environmental
and/or social issues.
A number of factors, including environmental conditions, resource
distribution, and social alliances, influence mobility decisions of early New
World inhabitants. The complex issue of mobility during the Late Pleistocene/
Early Holocene on the Northwest Coast of North America is further
complicated by a lack of early sites. Despite this, evidence exists suggesting
that early coastal people survived using a maritime technology (Dixon 1999;
Rick et al 2001).
This thesis focuses on a site of the Northwest Coast Microblade
tradition (explained in detail in Chapter 4). Sites of this tradition extend from
the Kodiac Archipelago southeastward along the Pacific Rim through Southeast
Alaska, British Columbia, Washington, and Oregon (Dixon 1999:178).
Josenhans et al. (1997) demonstrate that sea levels were higher on the Prince of
Wales Islands during the Late Pleistocene/ Early Holocene, cutting off
connections between many of the islands in the region and the mainland.
Dixon (1999) contends that two important factors indicate a maritime

economy for sites of the Northwest Coast Microblade tradition. The sites share
a coastal ecological setting, and preserved faunal remains indicate maritime
subsistence. Because of the coastal setting, mobility in the region during the
Late Pleistocene and Early Holocene may have required the use of watercraft.
Rick et al. (2001) argues that New World people along the Pacific Coast
effectively used watercraft between 11,500 to 8,500 BP. Analyzing fish bones
from Late Pleistocene/Early Holocene sites on Channel Island in southern
California, the researchers found that marine resources attracted people at least
12,000 years ago (Rick et al. 2001). People fished for a diverse array of aquatic
creatures and exploited a variety of marine habitats and resources (Rick et al.
2001). Because the Channel Islands are separated from the mainland, the use of
watercraft is indicated.
Dixon argues that discoveries at 49-PET-408 inferentially demonstrate a
maritime existence. Evidence supporting this was obtained from analyzing
human remains found at the site. This suggested that the inhabitant found at the
site subsisted on a marine diet. The coastal environment supplies another
indication of a maritime existence. The site was cut off from the mainland,
therefore island-hopping" with the use of watercraft may have been practiced.
If maritime movement occurred during this time, the early inhabitants who
utilized 49-PET-408 may have practiced the same procurement and trade

activities to obtain valued ochre as they did to obtain lithic material.
Resource Distribution
Because the way that raw material is distributed has an impact on how
the material is obtained, the availability and quality of raw material also affects
mobility. Local patterns of lithic resource availability place fundamental
constraints on technology and social organization (Bamforth 1986). Bamforth
argues that resource distribution is such an important factor in understanding
the archaeological record that it actually shapes the technological organization
of the societies in question.
Looking at evidence of tool .maintenance and recycling from sites in
coastal California and the southern. High Plains, Bamforth (1986) addresses the
importance of raw material availability to technology. The first example is the
San Antonio Terrace, an area covering 30 kilometers adjacent to the Pacific
Ocean in northern Santa Barbara County, California. The sites discussed are
located in two distinct areas of a dune field: one is its southern edge in
stabilized dunes along the San Antonio Creek and the other is in more recently
active dunes near the center. Lithic resources consist of high-quality chert
outcrops focused along the ocean five kilometers south of the Terrace. Low
quality, but usable, stone is found in other areas nearby.

Tool types found from recent sites indicate two types of sites on the
Terrace: more intensely occupied base camps near its edges and temporary,
task-specific camps in the central dunes (Bamforth 1986:45). The differences
between the two sites are represented by the different tool technologies found in
the archaeological record. In the recent sites of the central dunes, flake
densities are low, indicating that little tool manufacture was occurring. Also,
no tools used for tasks other than resource procurement and processing are
found. Another aspect of the recent central dune sites is a narrow range of
represented source material. Terrace edge sites, on the other hand, have lithic
material from a wide range of sources. Also, projectile point bases are over-
represented at the intensely occupied base camps indicating probable projectile
repair. Tools used to procure and process game dominate collections from
these sites, suggesting that recent sites on the terrace are behaviorally specific
(Bamforth 1986).
Early sites, oh the other hand, are more homogeneous, suggesting a
simpler site typology with little or no differentiation in the activities carried out
at different sites (Bamforth 1986). Early central dune sites tend to have higher
debitage densities than those in similarly located recent sites. While quarries
are exploited similarly during both early and recent periods with a dominance
of material from the coastal outcrops and less material from other sources,

recent terrace edge sites have a wider range of sources represented than die
central edge sites. Earlier sites* on the other hand, have a diverse selection of
material in both locations. Other evidence from early sites, including the lack
of projectile point base distribution pattern, absence of clear contrasts in the
distribution of tools, point bases, and debitage, suggest that these sites were not
behaviorally specific. Special use camps were weakly developed during the
early period occupation with settlement pattern becoming more complex
through time.
This example illustrates how the way that resources are distributed
affects technology and use. While high-quality chert is available in the region,
shortages occur on the Terrace because of limits on how much can be carried on
one trip. Once people are in the dunes, there is no immediate access to more
material, so material must be conserved The mobility patterns of the early
period short-term occupations required a portable toolkit and high curation.
Increased complexity of hunting techniques at the more recent occupation
improved the hunters' access to chert, perhaps through caching material at
staging areas near the edge of the Terrace. While this method may have
reduced the need for curating tools, difficulty of traveling in the dunes still
necessitated an easily transportable toolkit.
The Lubbock Lake Site in Lubbock County, Texas was intermittently

occupied over the past 12,000 years. The collection from the site contains three
basic types of raw material: high-quality microcrystalline stone available no
closer that 80 kilometers away, non-microcrystalline, but durable, cherts
located relatively close by, and low quality silicified caliche located at the site
(Bamforth 1986). Analysis of the tools from the site indicates that tools made
from non-local, high-quality material are more highly maintained and recycled,
and are used for a wider range of tasks than tools made from local materials.
Bamforth argues that Paleoindian hunters at the site relied on a toolkit with two
distinct components: a basic section of more curated tools made from high-
quality non-local material, and a specialized and more expedient component of
more local material prepared more specifically for the needs of a given trip.
Different distributions and the nature of different material affect how toolkits
are organized.
Technological organization is shaped by lithic resource availability and
settlement patterns of societies in question. The two examples provided by
Bamforth illustrate how the nature and distribution of lithic resources critically
affect technological efficiency (Bamforth 1986). These cases show how two
aspects of curation, maintenance and recycling of tools, are responses to raw
material shortages, caused by regional geological conditions and from behavior
patterns restricting access to raw material in certain contexts.

Procurement Methods
Raw material is acquired through different strategies, including direct
procurement, embedded procurement (Binford 1979), or through trade systems.
Direct procurement refers to specialized trips conducted with the exclusive
purpose of acquiring the needed material (Seeman 1994). Through embedded
procurement, individuals obtain the raw material during subsistence activities,
and not during specialized trips specifically for the raw material (Binford 1979;
Jones et al. 2003; Seeman 1994). Finally, individuals may acquire material
through trade networks (Hayden 1982). As explained below, Hayden (1982)
argues that Paleoindian trade patterns were used to cement economic and social
alliances during times of climatic changes.
Researchers determine the type of procurement strategy utilized through
various sources, such as direct analysis of quarries where the raw material was
obtained, from regional studies, through analysis of lithic material at the site
under investigation, and through environmental reconstructions. The type of
procurement strategy practiced depends on a variety of factors, such as
technology, the environment, the availability and quality of the raw material,
trade network alliances, traveling distance required, and mobility. Most, if not
all, of these factors are interrelated.

Embedded Procurement
Based on ethnographic research of the Nunamiut Eskimo, Binford
(1979) observed that the acquisition of tool-making raw materials is embedded
in the subsistence-settlement system, thus minimizing the direct costs
associated with procurement of the materials. Foragers or hunters would obtain
the material during routine subsistence trips, which would minimize search and
travel costs. Binford (1979:259) observed that only when things have gone
wrong does one venture into the environment for the sole purpose of obtaining
raw materials for tools. For groups whose range covers a large and changing
environmental zone, obtaining raw materials during other activities proves
more efficient and less costly than conducting specialized trips specifically for
the acquisition of the needed material.
Binford assumes that the quest for food is the most important activity,
and is the primary factor used in making other decisions regarding the different
modes of procurement, manufacture, use, and discard of tools. He views the
Nunamiut as an "extreme" example, however, because of their extreme
conditions: more than 70 percent of their yearly subsistence is obtained during
30 days: 15 days during spring caribou migration and 15 days during fall
caribou migration (Binford 1979). However, other groups may determine ways
to obtain lithic material differently. This may depend on several factors,

including the distance to high or adequate-quality lithic material, scheduling
patterns, and perhaps how tasks are divided among group members (older
children may be sent out to specifically to obtain raw material). For other
groups, it may have been just as efficient to have the extraction of several
resources embedded in each other, not only embedded in subsistence patterns.
One important aspect Binford fails to mention is the fact that the Nunamiut no
longer rely solely on lithic material, but use guns for hunting. Using an
ethnographic analogy to explain how lithic material was obtained
archaeologically when subsistence strategies are so drastically different makes
the analogy much weaker.
The nature of the embedded system depends on the frequency of moves
and the size of the seasonal round. When the size of that seasonal round and
mobility pattern changes because of environmental changes, the procurement of
other material is also affected. A study from the Great Basin during the
terminal Pleistocene/ Early Holocene illustrates this process.
Using source characterization of obsidian and dacite tools, Jones et al.
(2003) determined the geographic or territorial scale of terminal Pleistocene/
Early Holocene population mobility in the central Great Basin. The researchers
argue that subsistence is a primary factor affecting Paleoarchaic people (11,500
to 8,000 years ago). Groups traversed large subsistence areas, ranging over 400

kilometers north to south, to key into resource-rich wetlands (Jones et al. 2003).
As Paleoarchaic travelers, these small groups emphasized use of a narrow
suite of high quality resources, had brief residential stays at settlements, and
traveled great distances (Jones et al. 2003). They invested effort to move
between resource-rich patches and obtained high quality lithic material along
the way using embedded procurement.
Jones et al. (2003) looked at aspects of Paleoarchaic mobility through
source provenance information, and patterns of tool manufacture, use, and
discard. X-Ray Florescence (XRF) was used to determine the chemical
characterization of obsidian quarries. The researchers determined that 75
percent of the artifacts from the Great Basin study came from four chemical
types (sources). Using XRF on dacite, or andesite, seven different sites were
determined. The order in which geologic sources were exploited can be
determined from stages of lithic reduction. Assemblages with early production
stages and related flake debris occur near the geologic source (Jones et al 2003).
Moving farther away from the source, the number of tool-using events
increases, and so does the number of assemblages with expended or broken
tools and associated resharpening debris. Far from the lithic source,
assemblages will be late-stage bifaces broken in the final stages of manufacture,
hafted bifaces broken, smaller flaking debris, and evidence of tool refurbishing.

Using XRF and the sequence of production and maintenance events of
lithic material, Jones et al. determined that Paleoarchaic foragers were traveling
at least 450 kilometers in a north to south direction, and 150 kilometers east to
west (Jones et al. 2003). However, the warming and drying period of the Early
Holocene caused a reduction in resource abundance within the favored
wetlands. The former travelers began to incorporate seeds into their diet,
introduced a wider range of prey, and increased their diet breadth (Jones et al.
2003). In other words, their residence time increased and their source diversity
decreased. This also affected obsidian source use; as residence time increased,
source diversity decreased. Jones et ui, argue that exchange increased during
this time, as populations attempted to cope with changes in the environment,
and the decreased opportunity to obtain high-quality lithic material through
embedded procurement.
Direct Procurement
Raw material also may be obtained during a specific pursuit directly
from the source. Unlike Binford, who argues that because the food quest is the
most important transaction, all other activities, such as raw material
procurement, are carried out during subsistence activities, others (Seeman
1994) argue that time and labor decisions are case-specific. Each circumstance

should be viewed in context to appreciate how and why technologies are
structured the way that they are.
Groups may obtain raw material directly for several reasons, including
the location of raw materials relative to other resources, or social aspects such
as aggregation. Returning to Nobles Pond in northeast Ohio, Seeman argues
that the acquisition of lithic raw material was not embedded in subsistence
behavior, but rather a specialized activity (Seeman 1994). The particular
demands of band aggregation required procurement if they gathered in locations
far from sources of acceptable lithic material.
Inhabitants of Nobles Pond lived far from two high-quality lithic
quarries (70 and 110 kilometers, respectively). Seeman argues that if the
occupants of the site were passively obtaining raw material, such as during
settlement rounds, the archaeological record would demonstrate a high degree
of variability in raw material choices. Bands had access to high-quality
material within 200 to 250 kilometers in all directions of the Nobles Pond site,
thus one would anticipate a variety of materials due to bands moving their
quarry regions during settlement rounds. On the other hand, strong intercluster
similarities with few material types would suggest a more systematic, direct,
and less passive approach (Seeman 1994). This would imply that the band
chose to systematically travel through certain lithic supply zones and not others.

A section of Nobles Pond called South Field contained four different
clusters of high-quality material eight to ten meters apart. Seeman believes that
the clusters represent not four different small-scale occupations, but rather an
aggregation of several bands. These aggregations may occur for several
reasons: to obtain short-term subsistence returns through communal efforts,
because of limited areas for habitation due to a hostile environment, or for
establishing social interactions (Seeman 1994).
Contrary to Binford, who argues that procurement of raw material is
embedded in subsistence activities, Seeman suggests a situation in which
subsistence may be embedded in lithic exploitation. From lithic analysis of the
Nobles Pond site, researchers found a predominance of nonlocal raw material
(70 to 110 kilometers southwest from the site) (Seeman 1994). This is despite
the availability of lesser quality cherts in the immediate vicinity. Those who
frequent aggregation locations may prefer exotic material to more easily
obtainable, lesser quality material for several reasons. It may represent a
constraint on group mobility; bands are forced to a fixed point in the warm
season to obtain a specific resource (Seeman 1994). Bringing exotic materials
to an aggregated site also may be a display of status. Whatever the reasons for
preferring non-local, high-quality material, it appears the inhabitants of Nobles

Pond were not obtaining raw material passively. The exotic materials were
obtained directly from the quarries to bring to the aggregated sites.
This example illustrates how groups may have procured important,
high-quality material to bring to aggregated base camps. One may postulate
that other important, non-lithic materials, such as ochre, also may be directly
procured and brought to aggregated sites. Ochre is an important mineral with
various uses: as a pigment, a preservative for hides, for medicinal purposes,
and for camouflage (see Chapter 3). The material may have been viewed as a
status symbol and also brought to aggregated sites.
The third way in which raw material can be obtained is through
exchange. Exchange is seen in the archaeological record as an irregular
continuance of one raw material relative to others over a given distance (Lee
2001; Torrence 1989). As explained below, obsidian source data suggests that
long distance trade may have been occurring on the Northwest Coast 9500 to
9000 years ago.
Analyzing the transition from Paleoindian periods (>10,000 to
approximately 8000 BP) to Archaic periods (8000 BP to 3000 BP) in North
America, Hayden (1982) argues that Paleoindian exchange of materials and

services was undertaken to maintain social contact and reciprocity structures
during times of impoverished and changing resource bases. These alliances
would have been crucial to early hunter/gatherers during the climatic changes
of the Late Pleistocene/Early Holocene. Hayden suggests that the networks
were not established for the exchanged materials themselves, but rather to form
a far-reaching alliance network to fall back upon in times of resource failure
(Hayden 1982:118).
After climatic changes improved, the reliability of food resources
improved, thus negating the need to maintain reciprocal relationships with other
bands. This is Seen in the changing archaeological record from the Paleoindian
to the Archaic stages. Paleoindian stages are identified in the archaeological
record by the presence of exotic lithic materials, fine quality craftsmanship of
tools, and low stylistic diversity (Hayden 1982). With Archaic traditions,
especially after 5,000 BP, stylistic regionalization occurred, in addition to
increased variety of technological elements such as grinding stones, chipped
adzes, and borers. A change in emphasis occurred from using exotic raw
materials to more local raw materials, and high-quality craftsmanship of
Paleoindian tools declined.
In sum, with increased resource diversity due to an improved climate,
the need for strong social alliances diminished, and people of the Archaic

became more independent economically. The need for economic alliances was
reduced and the maintenance of these economic bonds during times of climatic
instability was abandoned. Hayden's analysis demonstrates that the
procurement behavior practiced depends heavily on, and is an important aspect
of, social and economic contexts of the society in question.
Researchers can determine potential geographic extent of group
mobility or trade networks using geochemical source characterization. In
southeast Alaska, several researchers attest that obsidian found in early sites of
the Northwest Coast Microblade tradition demonstrates an early trade network
(Ackerman 1992; Ackerman et al. 1979; Davis 1980; Dixon 1999). While
exchange is best determined where site densities are high, the current body of
evidence of early Northwest Coast sites is quite small. Only two other sites can
be compared with 49-PET-408 based on time period and proximity: Ground
Hog Bay II and Hidden Falls (Figure 2.1). All three sites date between 8,200
and 10,200 years BP, and all are located within the Alexander Archipelago. In
addition, all three sites have obsidian from two different sources: Suemez
Island and Mount Edziza.

Figure 2.1 Map of Southeast Alaska and Locations of 49-PET-408, Groundhog Bay, and
Hidden Falls. (Used with permission from E. J. Dixon).
At the Groundhog Bay II site, X-ray fluorescence analysis (Carlson
1994) suggests that the obsidian was obtained from Mount Edziza, located 220
kilometers up the Stikine River in northern British Columbia (Ackerman 1996).

Other obsidian may have been obtained from Suemez Island on the west coast
of Prince of Wales Island. Ackerman believes that the "evidence of trade or
dispersion of peoples from the Stikine River Valley to the coast is supported by
dates in excess of 9000 years at the Groundhog Bay II site" (Ackerman et al.
At the Hidden Falls site (Davis 1980), which dates to circa >9000 BP,
obsidian was also sourced to Mount Edziza and Suemez Island sources. Lee
(2001) notes that the presence of obsidian in these early southeast Alaskan sites
supports the hypothesis that long distance transportation and/or trade occurred
within the Alexander Archipelago during the Early Holocene (Ackerman 1996;
Davis 1980; Dixon 1999). To procure obsidian from Suemez Island, located in
the heart of the Alexander Archipelago, inhabitants of southeast Alaskan sites
might have "island hopped" by watercraft (Lee 2001). However, due to its
location 90 miles inland and over 250 miles away at the elevation of over 8000
feet, obtaining obsidian from Mount Edziza would have been more difficult.
People may have adopted a pedestrian mode of travel over rough terrain, or
perhaps they traded with inland people. More early sites need to be identified
to test these hypotheses.

Ochre Procurement
Although procurement studies conducted by Binford (1979), Hayden
(1992), Jones et al. (2003), and Seeman (1994), are based on lithic raw
materials used to make stone tools, these models also can be used to understand
the procurement of other important materials, such as ochre. However,
individuals may have conducted ochre procurement differently than lithic
procurement for several reasons. Depending on how individuals at the site
utilized ochre, the amount of the material needed may have been less than the
amount of lithic material needed for subsistence activities. Also, since ochre
may have had both utilitarian purposes (for hide preservation, medicine, etc.)
and ritualistic uses (mortuary contexts), individuals probably valued the
material greatly and expended more energy for its acquisition, either through
direct or embedded procurement, or trade. Researchers analyzing lithic
materials can use a variety of approaches, such as refitting flakes, distributional
analyses to identify changes in intensification, reduction analyses of local and
nonlocal materials, etc. However, by virtue of its more fragile nature,
researchers must approach ochre analysis differently.
By analyzing the distribution of artifactual ochre, researchers have
reconstructed decisions of early Paleoindians concerning which ochre sources
to procure, how to procure the sources, and how to use the ochre (Tankersley et

al. 1995:193). Ochre sources can be determined through geochemical source
characterization to reconstruct the procurement range, or distance traveled to
obtain the resource, and through comparison of ochre found at other sites.
Similar to geochemical source characterization of obsidian, sourcing ochre
enables researchers to address the potential geographic extent of group mobility
by showing the distance individuals may have traveled to obtain the raw
Binford, for example, studied ochre distribution at the Pomranky site, a
Red Ochre cemetery of the late Archaic period in the Great Lakes region
(Binford 1972). He compared sites in the region to determine if the material
had been widely traded. By comparing sites, Binford identified recurrences of
ochre varieties that provided information about cultural continuity and
processes of cultural change (Binford 1972).
Through geochemically identifying ochre from the Hell Gap site in
Wyoming, Tankersley et al. (1995) reconstructed mining activities at a hematite
outcrop site in Sunrise, Wyoming. The physical, chemical, and biological
properties of Sunrise red ochre corresponded with the characteristics of red
ochre found at the Hell Gap site, indicating procurement of this hematite
source. This reconstruction, comparable to the analysis of 49-PET-408 ochre,
provided researchers with associated economic and ideological aspects of the

early inhabitants at this site (See Chapters 3 and 5).
To reconstruct ancient trade and travel routes, this research first
identifies the geochemistry of the ochre at 49-PET-408. After establishing the
signature of the ochre, it is possible to identify recurrence of varieties by
comparing the geochemical characteristic to the signatures of known ochre
sources. This procedure will assist in determining if ochre from 49-PET-408 is
attributed to one or more particular sources. Will geochemical data derived
from PIXE analysis reveal that inhabitants procured the ochre from one or more
distinct ochre sources? If so, then can that source be identified from an existing
comparative database of known ochre sources (Erlandson et al. 1999)? If the
results from the chemical analysis of ochre at 49-PET-408 denote the same
geochemical signatures in these specimens, it would be strong evidence
suggesting that the occupants of the site obtained the ochre from one distinct
source either through embedded procurement, direct procurement, or trade.
Different chemical signatures, on the other hand, would indicate that the ochre
was likely being procured from different sources. Results that signify
procurement from one specific source are then available to interpret socio-
economic relationships through trade or direct procurement.

This analysis illustrates how researchers can use ochre to deepen their
understanding of archaeological sites. Tracing discrete source deposits of ochre
through geochemical analysis can provide yet another line of evidence to
reconstruct trade and procurement activities. Just as Hayden (1982) maintains
that Paleoindian exchange of materials and services was undertaken to maintain
social contact and reciprocity structures, 1 argue that groups may have
exchanged ochre to establish and maintain social alliances. These alliances
would have been crucial to early hunter/gatherers during times of impoverished
and changing resource bases, such as the climatic changes of the Late
Pleistocene/Early Holocene.
Artifacts from early hunter-gatherer sites often reflect subsistence
strategies; ochre analysis augments this knowledge by providing insight into
other important aspects of early cultures, which is discussed thoroughly in the
next chapter. For example, prehistoric peoples may have used ochre for
medicinal purposes (Peile 1979; Velo 1984) or for hide, food, or wood
preservation (Keeley 1980). Previous investigations associate ochre use with
symbolic, or ritualistic behavior (Bednarik 1994; Davidson and Noble 1989;
Marshack 1981; White 1992; Wreschner 1980) often depicted in burials and
used in cave paintings. Thus, regarding the incorporation of ochre for symbolic
transmission, what significance did this material have for very early prehistoric

inhabitants of southeast Alaska? This analysis seeks to understand ochre uses
in relation to socio-economic organization, symbolic and ritualistic behavior,
and procurement and trade activities of the early inhabitants who utilized 49-
PET-408, and to provide insights into the importance of ochre in prehistoric
sites along the Northwest Coast Of North America.

The seemingly ubiquitous nature of ochre in archaeological contexts
worldwide is startling. Ochre has been found in association with human
activity from the Lower Paleolithic to present-day art and rituals. Striking are
the cross-cultural similarities in the uses of ochre that demonstrate a variety of
purposes, including use as a pigment (Sagona 1994), a preservative for hide
(Keeley 1980), for medicinal purposes (Cordwell 1985; Peile 1979; Velo 1984),
and for camouflage. Examination of this important material potentially
provides archaeologists with information concerning mortuary practices, art,
ancient procurement activities, trade networks, and interaction patterns.
Red ochre is an impure variety of the mineral hematite (i.e., Fe203,
ferric oxide) that is one of the most common minerals in the world (Tankersley
et al. 1995). Hematite, derived from the Greek word haimatites meaning
blood-like (Sinkankas 1964), refers to the vivid, blood-red color produced when
the mineral is in powder-form (Tankersley et al. 1995). The color and luster of
hematite varies from specular, black, gray, and silver-gray to a non-specular
dull earthy, red, brown, reddish-brown, and cherry-red (Chesterman 1978).
Because of its brittleness, fresh fractures on specular hematite will quickly
weather to an earthy, or ochreous, red hematite. When heated, the color of

ochre intensifies and darkens (Velo and Kehoe 1990). In fact, yellow goethite,
a mineral low in iron oxide and high in silica, can be transformed to red
hematite when heated between 260 and 280 degrees Fahrenheit (Weinstein-
Evron and Hani 1994). While most forms of hematite occur as sedimentary
deposits, hematite also is formed in igneous and metamorphic rocks, resulting
in its widespread availability (Tankersley et al. 1995).
Traditionally, researchers have paid little scholarly attention to ochre
found in various ancient and historic sites worldwide (Erlandson et al. 1999).
For example, ochre is found in several late Archaic burials in the Great Lakes
region of North America (Ritzenthaler and Quimby 1962), but the presence of
the ochre at these sites is only briefly mentioned and largely ignored. Recently,
however, researchers have begun to acknowledge ochre and its importance in
prehistoric, historic, and contemporary societies worldwide, although the
literature is still relatively sparse.
Ochre studies are divided into three main groups: studies based on
utilitarian uses of ochre, analyses based on geochemical sourcing of ochre, and
studies of symbolic uses of ochre. Utilitarian uses of ochre include hide, food,
and wood preservation (BCeeley 1980), medicinal uses (Cordwell 1985; Peile
1979; Velo 1984), as a pigment (Miller 1980; Sagona 1994), and as an abrasive
for polishing bone and ivory (Tankersley etal. 1995). Research based on

geochemical sourcing of ochre includes identifying trace elements of ochre to
distinguish among different ochre sources (David et al. 1993; Erlandson et al.
1999; Tankersley et al. 1995; Weinstein-Evron and Ilani 1994). Having the
geochemical identity of different sources known gives archaeologists an
important tool. Ochre found in archaeological contexts at sites can be
geochemically analyzed and compared to the trace elements of particular
sources. Ochre that is found to match source areas can shed light on
procurement activities, trade, and interaction. Symbolic studies of ochre
include examinations of the symbolic meanings prehistoric people may have
placed on ochre or the color red (Bednarik 1994; Foster and Botscharow 1990;
Marshack 1981; Wreschner 1980). The primary focus of symbolic studies is
the widespread occurrences of ochre in burials. In this analysis, symbolic
studies of ochre are discussed minimally due to the fact that no burials were
found at 49-PET-408.
Utilitarian Uses of Ochre
The vivid color of ochre and its widespread utilization in mortuary
practices leaves many scholars to conclude that the use of this mineral served
only symbolic or ritualistic behaviors (Wreschner 1980). However, some
researchers contend that ochre served many important utilitarian purposes as

well. These include as a preservative for hide; for medicinal purposes, as paint,
and for camouflage (Cordwell 1985; Keeley 1980; Peile 1979; Sagona 1994;
Velo 1984).
Keeley (1980) examined 10 ochre-stained Magdalenian endscrapers
from La Madeleine, Abzac, and Laugerie-Basse housed at the Pitt-Rivers
Museum. Using high-magnification techniques, Keeley found traces of dried
hide polish on these same endscrapers, as well as traces of red coloring in the
cracks of the tools working edges. Keeley concluded that the Magdelanians
were probably rubbing ochre (combined with fat or grease) into their prepared
hides (Keeley 1980). Red ochre stains appear frequently on the floors of
Magdelanian houses. Keeley suspected that the surface stains might be traces
of decay left by ochre-treated hides that were used as floor coverings.
Additionally, Keeley speculated that the red ochre associated with Magdelanian
burials might be the remains of rotted, red ochre-impregnated hides (Keeley
Magdelanians rubbed ochre on hides for two possible reasons: as a
preservative and for aesthetic purposes. Hides colored red were perhaps more
aesthetically pleasing than light-brown, untreated hides. Additionally, applying
iron oxide in the form of ochre inhibited the decay of the hide. As Keeley
explained, hide is composed of collagen. Collagen decays when attacked by

bacteria-producing collagenase, an enzyme that destroys collagen (Keeley
1980). Since metal ions, like those of iron, inhibit the action of collagenase, the
application of red ochre may inhibit the decay of hides. In addition to using
ochre as a preservative, some researchers argue that ochre may have been used
to promote healing when ingested or applied to the skin. Iron compounds act as
an astringent to neutralize the enzymes produced by bacteria (Cordwell 1985).
The Australian Aborigines and Tasmanians used ochre for a variety of
purposes: for decoration, ceremonial markings of their bodies, paintings for
caves, bark, or sacred boards, camouflage, and as medicine (Peile 1979; Sagona
1994). When Europeans first encountered Tasmanians in the early 1800s, they
documented how the Tasmanians used ochre mixed with grease to color their
hair and bodies. Depending on the nature of the occasion, Tasmanian men and
women smeared their bodies with charcoal and red paste, plastered the ringlets
of their hair with a paste of ochre and grease, or shaved their heads completely
and rubbed red ochre onto their scalps (Sagona 1994). European traveler
George Robinson reported that the women in Tasmanian society mined ochre
from Toolumbunner, a large ochre outcrop that the Europeans called the City
of Ochre (Sagona 1994). The women also procured and prepared the ochre for
use by the entire community. Once mined, the ochre was rendered down in two
stages. First, the ochre was pulverized at the mine site with a pounder and a

ballywinne stone, which served as a mortar. Then the women loaded their bags
with the mineral and joined the men at camp. Once at camp, the men would
produce the ochre-and-fat paste with the fat of penguins or other birds (Sagona
Researchers of Tasmanian studies, Robson and Plomley, endeavored to
rediscover Toolumbunner by consulting journal entries by the explorer
Robinson. They wanted to locate the ochre deposits, study ochre use, and
reconstruct trade and relationships between the tribes. The journal entries
proved invaluable as Robson and Plomley were able to locate the mine and
conduct excavations. The excavations, which commenced in 1982 and 1984,
revealed five centuries of ochre mining and processing from the scatters of
stone tools. Mining and processing tools included various end- and side-
scrapers, blades, retouched flakes* and ballywinne stones (Sagona 1994).
Utilizing both ethnographic literature and excavation, examinations at
Toolumbunner have provided new and substantial data on Tasmanian ochre
mining. Although Robson and Plomley examined the major and minor
elements of the Toolumbunner ochre, geochemical analysis to identify the trace
elements of the minerals would be helpful for future work in reconstructing
trade activities and movement.
Father Anthony R. Peile conducted linguistic work, collected traditional

remedies, and examined the Australian Aborigines concepts of health,
nutrition, and disease at the Balgo Mission, south of the Kimberley Ranges in
the northern part of Western Australia (Peile 1979). Peile described the use of
traditional ochre remedies of the Aborigines (Peile 1979). Medicinally, they
would cover a wound with warm leaves, chew ochre and spit it out over the
wound. The wound would dry, the people would sing for healing, and soon the
wound would form a scab and fall off (Peile 1979); The Aborigines would also
mix ochre with fat, cold ash, and water and apply it to sore areas on the body or
on bums. The fat ensures that the ochre remains on the wounded area. The
patient is then placed in the midday sun, to promote sweating. Peile suggested
that minerals in the ochre, such as iron, magnesium, and zinc, constitute the
healing properties of the ochre (Peile 1979).
Velo (1984) states that the healing properties of ochre were known long
before ochres appearance in Mousterian funeral rites as at La Chapelle-aux-
Saints, where bones, jasper, and ochre were placed around the head of a corpse,
and in Upper Paleolithic art beginning with the Aurignacian, when Home
sapiens sapiens carved bone objects and painted them with red ochre. Iron
oxides contain a property that converts oxygen into ozone, which promotes
healing. The astringent and styptic nature of iron compounds arrest
hemorrhages and also deodorizes (Velo 1984). Velo argues that early humans

must have learned about the healing aspects of ochre through experience,
speculating that the presence of ochre in burials may not be from funeraiy
ritual, but from efforts to save the persons life (Velo 1984). In fact, ochre has
been found in association with Homo erectus activities at Olduvai Gorge in
East Africa, Tenra Amata in France, and the site of Becov in Czechoslovakia
(Marshack 1981).
While the healing properties of ochre are known, there is currently no
evidence to support the claim that ochre was used for this purpose in the
archaeological record. In addition to healing properties, Cordwell suggested
that Homo erectus may have discovered that ochre, when ground into a powder,
mixed with water and applied to the skin, acts as an effective antiseptic and
body deodorant (Cordwell 1985). Cordwell has focused on the early use of
cosmetics and the art of human transformation through clothing and adornment
(Cordwell 1985). Although confirming her hypothesis is nearly impossible,
one wonders when Homo sapiens predecessors discovered that applying ochre
to animal hides prevents bacteria from generating odor, and applying it to ones
own skin also prevents odor and effectively prevents various wounds from
becoming infected (Cordwell 1985).
Speculating on other possible utilitarian uses of ochre, prehistoric
people may have drawn on cave walls or other objects to communicate

information or warnings. Perhaps games were played in a way that children
today use chalk to draw on sidewalks. While some approaches can be tested,
such as Keeleys microwear analysis on tools, many hypotheses regarding
utilitarian ochre uses cannot, such as ochres possible medicinal uses, unless
coprolites are found. Comparing utilization by contemporary and historically
based people is an important way to help understand possible uses by
prehistoric people. However, one cannot assume that the past will resemble the
present, or that all utilitarian purposes that ochre were used for in the past are
still used today.
Geochemical Analyses of Ochre
Red ochres were highly valued and traded over long distances at least
since the appearance of anatomically modem humans (Erlandson et al. 1999).
Thus, knowing the trace elements of individual source areas allows researchers
to reconstruct ancient trade and travel routes if ochre from specific sites can be
traced to the original ochre source (Erlandson et al. 1999).
Recently, researchers have made exciting progress by sourcing discrete
ochre source deposits which aids in reconstructing trade and procurement
activities (David et al. 1993; Erlandson et al. 1999; Tankersley et al. 1995).
Geochemical techniques such as PIXE (particle-induced X-ray emission)

(David et al. 1993; Erlandson et al. 1999), X-ray diffraction (Tankersley et al.
1995), and Inductively Coupled Plasma-Atomic Emission Spectroscopy (ICP-
AES) (Weinstein-Evron and Ilani 1994) have effectively been applied to ochre
for the purpose of tracing discrete source deposits (David et al. 1993; Erlandson
et al. 1999; Marshack 1981; Tankersley et al. 1995; Weinstein-Evron and Ilani
At el-Wad Cave, one of the Mount Carmel Caves in Israel, researchers
Weinstein-Evron and Ilani (1994) found an Early Natufian occupation dating
from circa 13,000 to 11,000 BP: Ochre fragments and basalt pestles stained by
red and yellow ochre were found in the inner portions of the cave. The
researchers conducted geological surveys of the ochre outcrops in the area and
mineralogical and geochemical comparisons to provide information on
exploitation patterns of the prehistoric population. Seven ochre outcrops were
found near el-Wad ranging at distances of one to 10 kilometers away
(Weinstein-Evron and Ilani 1994).
Weinstein-Evron and Ilani used x-ray diffraction analysis to determine
mineralogical types and Inductively Coupled Plasma-Atomic Emission
Spectroscopy (ICP-AES) to determine the chemical composition, or trace
elements, of the ochre (Weinstein-Evron and Ilani 1994). The x-ray diffraction
analysis showed that most of the fragments were composed of hematite (36.5

percent), although goethite (15 percent) and jasperoid (8.5 percent), which is
rock composed of low iron oxide content and high silica content, were also
abundant. Twenty-two percent of the pieces were burnt clay. The inhabitants
of the cave were probably extracting red pigment from the hematite and
goethite. Jasperoid is too hard to grind and too poor in iron oxide to be used as
an effective pigment.
The mineralogical and chemical composition of the iron oxide
fragments at el-Wad Cave was similar to the mineralogical and chemical
compositions of the local sources. The ICP-AES analysis showed that the
ochres from Mount Carmel and el-Wad cave have similar ranges of elemental
contents. Mineralogically, the ochres from the two areas also have similar
percentages of hematite, goethite, and jasperoid. Thus, prehistoric inhabitants
of el-Wad Cave collected ochre from outcrops and alluvium approximately 10
kilometers from the cave.
Interestingly, Weinstein-Evron and Ilani found that hematite was rare in
outcrops surrounding Mount Carmel. The authors suggested that the
inhabitants may have heated the goethite to obtain hematite. Yellow goethite,
when heated between 260-280 degrees, transforms to red hematite (Weinstein-
Evron and Ilani 1994). However, aside from burnt flints and bones in the
Natufian layers of the cave indicating use of fire, the researchers found no

direct evidence for the burning of ochre at el-Wad Cave.
Using PIXE (proton induced x-ray emission) on eight ochre sources
from California, Oregon, Wyoming, and Alaska, Erlandson et al. (1999)
distinguished twenty-one major and minor trace elements among the
specimens. Statistically analyzing the PIXE quantitative data with an RQ-mode
principal components analysis (PCA), the researchers found that there are
geochemical differences in each specimen that indicate researchers may be able
to separate out source areas in archaeological specimens (Erlandson et al.
Erlandson et al. also looked at intrasource variation at Sunrise Mine,
Wyoming, and Tajiguas in Santa Barbara, California. The Tajiguas specimen
was subdivided into four subsample splits to test the internal consistency of
the results (Erlandson et al 1999:523). Two Sunrise Mine specimens were
obtained from a core only three meters apart. While the results from the
Tajiguas specimen were consistent with each other, this may be because
running a test multiple times on a subdivided specimen is really an examination
of instrument precision, not intersource variability. Because significant
variation was detected between the two Sunrise Mine specimens, more research
is suggested to evaluate internal variation in discrete ochre deposits (Erlandson
et al. 1999). To test intersource variability, specimens should be obtained from

different areas at the same source, not from a single, subdivided sample.
In Australia during the late Holocene, the Australian Aborigines
underwent extensive changes in their social systems, including the beginning of
large scale, intergroup ceremonial networks (Lourandos 1983 as cited in David
et al. 1993). Thusfar archaeologists have directed little attention to past inter-
regional interactive networks due largely to a lack of sourceable archaeological
material. This is because the structure and chemistry of most stone raw
materials is not region-specific, and that Australian stone artifacts lack the
regional typological variability common to many other lithic industries
throughout the world (David et al. 1993:50); Ochre is also more plausibly
linked to ceremonial networks than lithics.
David, Clayton, and Watchman aspired to alleviate this problem by
sourcing various ochres from different archaeological sites in northern
Australia. Ochre has been a widely traded material among the Aboriginals, at
least at the time of initial contact (David et al. 1993). The researchers therefore
set out to determine whether the ochre sources had distinguishable element
concentrations by conducting PIXE on 60 ochre specimens (David et al. 1993).
David et al. wanted to ascertain whether the red ochres from sources in different
geological settings were differentiable, and if ochres from sources within a
single geological setting were differentiable (David et al. 1993). The specimens

examined included 33 from 5 different source locations 150 kilometers
southwest of Katherine, Northern Territory, and 36 specimens from Cape York
Peninsula. The PIXE results were subjected to a Multi-Dimensional Scaling
statistic and a Cluster Analysis to identify whether Cape York ochres could be
identified as chemically distinct from Northern Territory ochres. Results
showed that the Cape York specimens could indeed be differentiated from the
Northern Territory specimens, although there was some degree of overlap
(David et al. 1993). The Northern Territory specimens, however, could not be
differentiated from each other.
The researchers concluded that PIXE is a useful method to differentiate
ochre specimens from different geological formations. However, further
research needs to be conducted to determine different sources within the same
geological formation. While the authors discuss reconstructing past movement
of ochres within and between regions, there is no mention of direct or
embedded procurement involved to initially obtain the ochre. To determine the
type of procurement strategy utilized, David et al. could analyze the quarries
where the ochre had been obtained, analyze the lithic material at the sites under
investigation for evidence of ochre mining, such as staining on the blades of
tools, and conduct environmental reconstructions.
Several researchers collaborated on an important analysis to examine

Early Paleoindian mining of Sunrise red ochre at the Powars II site in Platte
County, Wyoming (Tankersley et al. 1995). Early Paleoindians used red ochre
for possible utilitarian or domestic purposes such as for pigments, as a
preservative, and as an abrasive for polishing bone and ivory (Tankersley et al.
1995). These uses are suggested based on occurrences of ochre on Paleoindian
habitation floors, in caches, on grinding stone surfaces, on game animal bones,
and on tools (Tankersley et al. 1995). Archaeologists speculate that red ochre
also served ceremonial needs of Early Paleoindians. Human skeletal remains
and associated grave goods have been found covered with red ochre at Early
Paleoindian sites such as the Anzick site in Montana (Lahren and Bonnichsen
The Powars II site (or Sunrise Mine site) is currently the only known
Early Paleoindian red ochre mining site (Tankersley et al. 1995). In addition to
a variety of specular and earthy hematite exposed at the surface, researchers
found hundreds of Paleoindian artifacts exposed and in situ deposits.
Tankersley et al. (1995) argued that these artifacts were used as mining tools
based on an examination of the chipped stone and bone artifact assemblage by
Stafford in 1990. Stafford showed that the tools, which include artifacts from
the Clovis, Goshen, Folsom, Midland, Agate Basin, and Hell Gap cultural
complexes, reflect the mining and processing of red ochre as well as an

ideological element of Paleoindian lifeways (Stafford 1990:64). The
Paleoindian artifacts, including projectile points, bifaces, and flakes, were
exposed in an in situ deposit of mine tailings. Some of the projectile points had
been used to dig into the ochre deposits (Tankersley et al. 1995).
The study included analyzing the physical and chemical characteristics
of the red ochre at the Powers n site in order to determine where Paleoindians
were using the ochre and why it was procured (Tankersley et al. 1995). The
researchers conducted macroscopic analysis to identify general physical
characteristics using Mohs scale and Munsell color charts, microscopic analysis
using scanning electron microscopy to identify mineral fabrics and
bioinclusions, and chemical analysis using x-ray diffraction to determine
overall mineralogical compositions. Next, artifactual red ochre from the Hell
Gap site in Goshen County, Wyoming was compared to the ochre at the Powers
II site. Results from the study conclude that the Hell Gap site material matched
the physical and chemical properties of the Sunrise ochre from Powers II site
located 10 kilometers away (Tankersley et al. 1995). Macroscopically and
microscopically, the material tested from the Hell Gap site was the same earthy
variety of hematite found at the Powars n site. The results from the x-ray
diffractometry test showed that the mineralogical composition of the specimen
corresponds to Sunrise red ochre.

Because the physical and chemical properties of the two sites match,
this demonstrates that Early Paleoindians from the Hell Gap site obtained this
mineral from the Powars II site. Procurement of ochre from Powars II may
have occurred directly, through trade, or during other journeys, such as hunting
expeditions. This important analysis shows how researchers can utilize ochre
to reconstruct Paleoindian procurement patterns and determine how and why
they were using this valuable resource.
Symbolic and Ritualistic Uses of Ochre
While the study of red ochre provides archaeologists with information
concerning ancient procurement activities, trade networks, and interaction
patterns, it also yields potential evidence regarding the symbolic and artistic
value of this prized pigment for cultures over time and throughout the world.
Use of this material has functioned as an integral part of many societies for the
transmission of social and cultural processes through ceremonies and initiation
Symbolic studies of ochre have focused mainly on burials, color theory,
and ethnographic analyses of ceremonial customs. Findings of ochre in burials
is widely discussed in early Homo sapiens sapiens sites in Europe (Wreschner
1980) and across the globe in North America, where people of the Late Archaic

used red ochre extensively in mortuary practices (Binford 1972; Laughlin 1967;
Orr 1962; Ritzenthaler and Quimby 1962; Snow 1980; Tuck 1976). Snow
(1980) explains that red ochre became a singularly important substance in
prehistoric mortuary practices in the Northeast, including the Red Paint
Cemeteries of Maine and the burials at Port Au Choix in Newfoundland. Both
have numerous burials that contain vast amounts of red ochre and associated
grave goods (Tuck 1976).
Many researchers have attempted to bring color-research issues into
archaeology (Turner 1966; Velo and Kehoe 1990; Weitman 1973; Wreschner
1980). Velo and Kehoe explain that archaeologists can examine the physiology
of human vision, which allows recognition of the red portion of the light
spectrum (Velo and Kehoe 1990). This is a universal trait for humans with
normal vision. Perhaps for this reason, red is the most commonly named color
in the human language. Multitudes of color tests and theories have shown a
predominance for red and a striking consensus about the meaning of red in
different cultures, such as the color red on national flags to symbolize wars,
courage, blood, revolt, struggle for independence, and revolution (Weitman
1973). Therefore, color behavior should be evaluated.
While many researchers have expressed excitement about bringing
color-research issues into archaeology, the concept is also met with

apprehension and skepticism (Bolton 1980; Butzer 1980; Delporte 1980;
Jacobson-Widding 1980; Masset 1980; Miller 1980; Solecki 1980). Both
Bolton (1980) and Masset (1980) pointed out that perhaps the color red and
ochre use itself is simply more salient and durable than other pigments.
Therefore, red survives in the archaeological record whereas other pigments,
which may have been organic and biodegradable, do not. Others cautioned that
the use and significance of the color red might have been multiple (Delporte
1980; Jacobson-Widding 1980). Red may symbolize different things when
used in different situations or contexts. Therefore, archaeologists need to look
at the ochre in relation to other artifacts or symbols (Jacobson-Widding 1980).
Researchers have also surveyed ochre uses among contemporary and
historically- based people. Ceremonies and initiation rites are often connected
with the color red based on studies of recent societies, such as the Australian
Aborigines, the Ndembu in Africa, Druze communities in Israel, Yemenite
Jews, and Mari Baluch in Baluchistan. With the Ndembu and the Australian
Aborigines, red is connected with birth, life, and death (Wreschner 1980).
Wreschner viewed this modem usage of ochre as repetitive patterns in symbolic
actions through more than 500,000 years. However, no current data suggests
red was used in the same ways throughout time.
While color-theory research and ethnographic studies may help shed

light on the importance of ochre, neither will truly answer why people in the
Paleolithic used red ochre. Both of these approaches are a form of
uniformitarianism, or assuming that past events were produced by the same
processes observed in the present (Velo and Kehoe 1990). Archaeologists
should keep in mind that Paleolithic people may have used ochre for purposes
not seen today or with historically based people.
Marshacks (1981) research of Paleolithic ochre and early uses of color
and symbol suggests that there cannot be a preconceived general rule for
possible meaning or intent regarding ochre use. Coloring probably represented
different modes of symbolic use, but archaeologists have no way of knowing if
the ochre represented aspects relating to blood, life, or death. Unlike
Wreschner, Marshack argued that the uses and importance of ochre might have
been multiple. Symbolic use of ochre and color were early, widespread, and
variable (Marshack 1981:190). Color could have been used in contexts
relating to life, death, or blood, as Wreschner argued, or ritual body decorations
to mark rank, status, age, sex. Ochre also could have been used to decorate
artifacts, to color floors or walls, such as in symbolic spaces (ritual areas, etc.)
or simply as a marking material. According to Marshack, ochre use and the
symbolism contexts cannot be reduced to a single universal meaning or intent.
Archaeologists can only really see the modes of use (Marshack 1981), rather

than decipher meaning or intent.
Researchers can, to some extent, use knowledge of the present and
historical past to understand Upper Paleolithic uses of ochre. However, as Velo
and Kehoe cautioned, Artifacts may have been colored with red ochre for
hundreds of thousands of years, but no amount of archaeological data will tell
us why. Its meaning in prehistoric society is ambiguous (Velo and Kehoe
Thusfar, geochemical sourcing of ochre proves the most promising
avenue of research regarding this mysterious mineral. As sourcing techniques
become more refined, archaeologists will have another important line of
evidence for understanding procurement activities and trade. Utilitarian studies
of ochre also hold promise. However, some uses may never be proven, while
others may simply never be known. Finally, while symbolic archaeological
studies provide researchers with an opportunity to make archaeology more
holistic, broader, and well-rounded, symbolic approaches alone are difficult to
study and confirm. Researchers should not deny the thoughts, beliefs, and
cognitive decision-making strategies of prehistoric people, but unfortunately,
this approach alone is difficult to test and study scientifically.

This chapter briefly discusses the environmental history of Beringia and
the Northwest Coast of Alaska from 25,000 years ago BP, until after the
Pleistocene/Holocene transition, 9,000 years ago BP. Included is a discussion
of three different theories for the peopling of the New World: the overkill
hypothesis (Mossiman and Martin 1975), the three-migration theory (Turner
1992), and the coastal migration model (Dixon 1993,1999; Fladmark 1979;
Josenhans et al. 1997). Next, a prehistory of the Northwest Coast is presented
with focus on one of the three regional variants of the American Paleoarctic
tradition: the Northwest Coast Microblade tradition. A brief discussion of two
early sites of the Northwest Microblade tradition, Groundhog Bay n and
Hidden Falls, follows. This analysis describes 49-PET-408, another site of the
Northwest Coast Microblade tradition, and the main focus of this thesis.
Methods of excavations conducted at 49-PET-408 are discussed thoroughly in
Chapter 5. The results of the excavations conducted, along with the results
from obsidian sourcing analysis from 49-PET-408 (Lee 2001), are presented in
this chapter.

Environmental History of Beringia
and the Northwest Coast
The last glaciation, the Late Wisconsin period, occurred from circa
25,000 to 13,000 years BP. During this time, glacial ice advanced southward
from the Arctic, covering vast areas of Alaska and Canada. Approximately 30
percent of Alaskas land area, in addition to sizeable portions of the continental
shelf in the Gulf of Alaska, was buried under ice (Ager 2002). Glaciers
expanded and retreated along the Bering Land Bridge and the Pacific Northwest
Coast of North America as a massive frozen tide moved across the continent in
rhythm with natural global warming and cooling. While glacial ice was
extensive during this time, the Late Wisconsin differed from the earlier
glaciation. Many more nunataks, or mountain summits that protrude through
ice sheets (Pielou 1991), were exposed and larger areas of the continental shelf
were free from glacial ice. The height of the Wisconsin glaciation,
approximately 18,000 years ago, exposed the continental shelf as dry land,
more commonly described as a land bridge that skirted the western border of
the glacial ice sheet (Dixon 1993).
The arid climate during the middle Wisconsin established a landscape
dominated by herbaceous tundra, rich in grasses, sedges, Artemisia, willows,
and forbs (Ager 2002). Late Pleistocene mammals, such as mammoths, horses,

caribou, bison, saiga antelope, cave lions, wolves, and grizzly bears, traveled
over the exposed land bridge. The temperatures in the region of the land bridge
were probably four to six degrees Fahrenheit cooler than today (Ager 2002). In
southeastern Alaska, ice covered the coastal and high mountains of the
Alexander Archipelago islands. Many of the western areas of the Archipelago
were free from glacier ice. Glaciers slowly melted and sea levels began to rise
by circa 16,000 years BP. Icebergs were abundant in the waters along the shelf
until approximately 13,000 years BP (Ager 2002).
During the Pleistocene/Holocene transition, around 10,000 years BP,
North America experienced immense environmental changes, including
mammalian extinctions and abrupt changes in the composition of forests,
woodlands, and parklands south of the ice sheets (Pielou 1991). Over 70
genera of animals including mastodons and mammoths, and species of bison
and horse species, became extinct in North America. Warmer temperatures
caused changes in marine and terrestrial ecosystems in southeast Alaska.
Despite the vast extinctions throughout North America, the changing
ecosystems in southeast Alaska triggered more plants and animals to colonize
the area (Ager 2000). Caribou and brown bear remains, dating from 12,500 to
10,000 years ago (Dixon 1999), suggest a hospitable area for animal and human
populations alike. Sites in Nenana and Tanana River Valleys of eastern

Beringia, such as Dry Creek, Moose Creek, and Walker Road, indicate human
occupation at 11,800 years BP (Powers and Hoffecker 1989).
The transition from the Pleistocene epoch to the current Holocene,
approximately 10,000 years ago, marks the end of the last glaciation and the
beginning of the present interglacial (Peilou 1992). A rapid warming occurred
following the transition from the Wisconsin to the subsequent interglacial, and
lasted for 3,000 to 4,000 years. The warming of the Holocene brought forests
and grasslands northward all across the western half of North America, and
upward into the hills and mountains. Geographical ranges of numerous plant
and animal species extended farther north in the hypsithermal (the warmest
time interval during the Holocene) than they do today. For example, fossil
evidence from the Seward Peninsula, Alaska, shows early signs of warmth,
with fossils of beaver dams, beaver-gnawed wood, and fossil logs of poplar,
birch, and spruce (Peilou 1992).
Models of New World Migrations
Three main theories for the peopling of the New World have emerged
throughout the twentieth century: the overkill hypothesis (Mossiman and
Martin 1975), the three-migration theory (Turner 1992), and the coastal
migration model (Dixon 1993,1999; Fladmark 1979; Josenhansetal. 1997).

For many years, the theory that hunters traveled into the New World in search
of big-game governed anthropological thinking (Mossiman and Martin 1975).
The theory suggests that human movement into North America caused the
Pleistocene extinctions. The animals, supposedly naive to human predation,
quickly succumbed to the sophisticated human predators. Mossiman and
Martin (1975) simulated a migration model in which humans colonized the
Americas in 1,000 years and killed off large Pleistocene mammals along the
way. However, colonization of the New World most likely took longer than
1,000 years because the new populations needed to gain familiarity with the
environment and resources. Also, the human population would not have been
large enough to cause the massive extinction, especially if humans hunted
smaller animals instead of big-game, like mammoths and mastodons. Hunting
smaller mammals would have been safer for human survival and humans may
have merely scavenged for larger game.
Despite scholars lack of agreement concerning early migration routes, most
archaeologists agree that the first people to inhabit North America arrived from
eastern Asia (Ames and Maschner 1999; Matson and Coupland 1995; Turner
1985). The theory of an Asian ancestry has been supported through various
studies, including the study of dental morphology (Turner 1985), linguistics
(Greenberg et al. 1986), and from the study of human genetics (Williams et al.

1985). Proponents of the three-migration theory (Greenberg et al. 1987; Turner
1992) postulate that the research also supports three separate migrations into the
New World from Asia. According to Turner (1992), three late Pleistocene
migrations are supported by language, dentition, genetics, and archaeology.
These three traditions are the Nenana, Denali, and Angula Island complexes,
that later developed into Paleoindian, Paleoarctic, and Aleut/Eskimo.
Turner (1985) has shown that living Native Americans have one single
dental characteristic, called Sinodonty (shovel-shaped incisors). This dental
trait is also found in north China, Mongolia, modem Japan, and all of pre-
Russian eastern Siberia. Turner suggests that the lack of dental evolution
proves that Native Americans, Arctic Aleuts, and Eskimos all have a northeast
Asian ancestral origin (Turner 1992). Using dental characteristics alone,
current New World populations can be divided into Aleut-Eskimos, Northwest
Coast and Canadian and Alaskan interior Indians, and all other North and South
American Indians. These divisions stand when looking at language differences
and genetic studies (Greenberg et al. 1986). Research shows that there are three
divisions of immunoglobulin G variants (antibodies) in the New World.
Geneticists believe that this is the result of three separate Siberian migrations,
and did not occur after one single migratory event (Greenberg et al. 1986).
In 1960 the coastal migration model was first proposed by Heusser to

explain the spread of Clovis, the earliest definite New World culture (Matson
and Coupland 1995). The model, further developed by Fladmark (1979), and
revived by Dixon (1993,1999) and Josenhans et al. (1997), suggests an
economy based on marine hunting and fishing and the use of watercraft.
Supporters of the coastal migration theory (Dixon 1993,1999; Fladmark 1979;
Josenhans et al. 1997) propose that the first people to inhabit the New World
arrived from Asia or Europe possibly as early as 16,000 years ago. The early
inhabitants traveled southward from Beringia along the Gulf of Alaska using
watercraft and continued to migrate southward along the Northwest Coast of
North America (Dixon 1999). Paleontological evidence of various land and sea
mammals, birds, and fish from this period indicates sufficient resources for
human consumption. Using Global Information System (GlS)-based analysis
Of elevation data and late glacial location of ice sheets, Anderson and Gillman
(2000) suggest possible migratory corridors Used by colonizing populations of
the New World. Their research suggests that populations spread and diversified
rapidly along coastal and riverine settings, and on plains (Anderson and
Gillman 2000).
Much controversy surrounds the issue of when and how the first humans
arrived in North America due to the incomplete nature of the archaeological
record; research is ongoing. If coastal occupation occurred first, archaeological

evidence should appear at multiple sites along the western coast of North and
South America. Unfortunately, few sites exist that pre-date 11,500 BP. Many
scholars (Butzer 1991; Collins 1991; Toth 1991; Wright 1991) have expressed
concern regarding the lack of evidence.
^Several factors, however, may explain the paucity of early coastal sites.
Researchers traditionally have thought that the regions along the Northwest
Coast had been glaciated prior to 10,000 years ago. Thus, little archaeological
work has been conducted in this region to explore possible coastal migration
routes. In addition to a lack of archaeological attempts, researchers may also
have difficulty locating coastal sites dating prior to 4,500 BP. Flooding caused
by post-glacial rise of water and the presence of thick rainforests may conceal
many sites (Fedje and Christensen 1999). Rainforests and caves may also hide
sites. Fedje and Christensen explain that in southern Haida Gwaii, shorelines
dating from 13,000 to 9,500 BP are underwater while those dating from 9,200
to 3,000 BP are located in the rainforest up to 15 meters above modem levels
(1999). In order to discover and investigate early archaeological sites, Fedje
and Christensen recommend modeling paleo-shorelines since sea-level
fluctuations may have submerged the history of coastal occupations.
Further confusing the issue are early interior sites, such as Meadowcroft
Rockshelter, pre-Clovis sites in Colorado, and sites from the Nenana Complex

in interior Alaska. The Meadowcroft Rockshelter site in southwestern
Pennsylvania (Adovasio et al. 1978) contains eleven major stratigraphic units
spanning the past 16,000 C-14 years, to possibly 19,000 C-14 years, making
this the site oldest in North America. Accepting the validity of the dates
recorded at Meadowcroft, humans migrated into the New World at least before
16,000 years ago, and possibly before 19,000 years ago. Using the coastal
migration model, which suggests that early inhabitants occupied coastal regions
first and eventually moved inward, how does one account for the early dates at
Meadowcroft, an interior site?
Little Salt Springs, located near Charlotte Harbor in southwest Florida,
holds the remains of organic Paleoindian and archaic artifacts that range in age
from 12,000 to 9,000 and 6,800 to 5,200 years ago (Clausen et al. 1979).
Excavators at the site have uncovered the shell of an extinct giant land tortoise,
Geochelone crassiscutata, impaled with a sharp wooden stake. The tortoise has
been C-14 dated to 13,450 years BP. The stake, dated to 12,030 C-14 years,
apparently caused the animal's death. Carbonized bone and fire-hardened clay
fragments suggest that the tortoise has been killed with the stake and then
cooked upside-down in its shell. Evidence recovered at Little Salt Springs
indicates that humans inhabited the site between 12,000 and 9,000 years ago.
Sites in Colorado, including Dutton and Selby (Graham 1981; Stanford

1983) in eastern Colorado and Lamb Spring (Stanford et al. 1981), in Douglas
County, Colorado, contain faunal remains that may be related to pre-Clovis
occupations in North America (Cassells 1997). According to Stanford (1983),
evidence from Dutton and Selby suggest the presence of humans in North
America prior to 11,500 BP. At the Lamb Spring site, near the Denver suburb
of Littleton, Colorado, excavations yielded the bones of extinct Pleistocene
mammals in conjunction with possible flaked chalcedony specimens. The
mammoth bone dates to 13,140 + 1000 BP, which is older than Clovis sites
(Dixon 1999).
The Nenana complex, found at several sites in interior Alaska, spans
between at least 11,600 BP to 10,000 BP (Dixon 1999). Archaeologists have
discovered artifact types of the Nenana complex throughout Alaskas Nenana
River Valley and the upper Tanana River Valley. Artifacts that define the
complex include triangular and tear-drop shaped projectile point and knives,
straight or concave lanceolate projectile points, perforators, end- and side-
scrapers, burins, hammer and anvil stones, and unifacial knives and scrapers
(Dixon 1999). The antecedents of the Nenana complex remain a mystery to
Perhaps the peopling of the New World actually occurred over a variety of
routes with a number of separate migrations. Lacking adequate archaeological

information from early sites, conducting more underwater archaeology along
the Pacific Coast seems wise. Paleo-shoreline modeling could indicate the
presence of possible underwater sites. Further excavation of caves along the
Northwest Coast of North America will solidify the archaeological foundation
necessary to develop and expand upon early migration theories.
Prehistory of the Northwest Coast
The three major cultural developments occurring in Alaska and the
Pacific Northwest prior to 8,000 years BP include the Nenana Complex of
interior Alaska, the Northern Paleoindian tradition, which is considered a
variant of western North Americas Paleoindian tradition, and the American
Paleoarctic tradition, which concerns this analysis (Dixon 1999).
The American Paleoarctic tradition is divided into three regional
variants: the American Paleoarctic tradition (circa 10,500 to 8,000 BP), the
Denali complex (10,500 to 8,000 BP), and the Northwest Coast Microblade
tradition, which first appears about 9,500 BP and continues into the mid-
Holocene, < 8,500 BP (Dixon 1999). The American Paleoarctic tradition
subdivision is found along the coastal margins of the Bering and Chukchi Seas
and the Arctic Ocean and adjacent terrestrial environments (Dixon 1999:173).
Distinctive artifacts associated with this tradition include microblades, blades

and blade cores, core bifaces, wedge-shaped microblade cores, and slotted
arrow points made from antlers for the insertion of microblades. Dixon
contends that people of the American Paleoarctic tradition probably hunted
marine mammals and exploited adjacent non-coastal areas to fish and hunt for
terrestrial mammals (Dixon 1999).
The Denali complex, found throughout interior Alaska and the Yukon
Territory, is an early microblade culture (Matson and Coupland 1995).
Common artifacts of this complex include microblades and microblade cores,
small prismatic blades, burins, large blades, and bifacial biconvex knives
(Dixon 1999). Economy of this tradition included, but is not limited to,
terrestrial hunting of large mammals and freshwater fishing.
The Northwest Coast Microblade tradition dates to approximately 9,500
BP in Alaska, but its roots go back to 30,000 years ago in North China or
Siberia (Carlson 1976). This tradition persists until sometime well after 8,000
BP (Dixon 1999). Also called the Paleomarine tradition by West (1996), the
Early Coast Microblade complex by Fladmark (1975), the Archaic period by
Ames and Maschner (1999), and the Microblade tradition by Carlson (1979),
this tradition marks the earliest evidence for human occupation in Southeast
Alaska. Sites from the Northwest Coast Microblade tradition extend from the
Kodiac Archipelago along the Pacific coast southward into Southeast Alaska,

British Columbia, Washington, and Oregon (Dixon 1999).
People that utilized microblade technology prepared micro-cores from
which to remove sharp-edged, parallel-sided flakes called microblades. The
microblades were then struck or pressed from the cores and presumably inset
into wooden or bone point to create a piercing implement or knife (Carlson
1976). Other artifacts that distinguish this complex include blocky microblade
cores, flake cores, utilized flakes, notched and waste flakes, bifaces with
constricting bases, leaf-shaped bifaces, and tools possibly used for scraping and
chopping (Dixon 1999).
Sites of the Northwest Coast
Microblade Tradition
There are several early sites of the Northwest Coast Microblade
tradition in southeast Alaska, including Groundhog Bay II (Ackerman 1996;
Ackerman et al. 1979), Hidden Falls (Davis 1980), Chuck Lake (Ackerman et
al. 1985; Okada et al. 1989), Thome River (Holmes 1988; Holmes et al. 1989),
Arrow Creek 2 (Fedje et al. 1996), Namu (Carlson 1996), Lyell Bay (Fedje et
al. 1996), Richardson Island (Fedje et al. 1996), and 49-PET-408 (Dixon et al.
1997). Of these sites, only two are of similar antiquity to 49-PET-408 (dating
to within 1,000 years) and located within the Alexander Archipelago (Figure

4.1): Groundhog Bay II and Hidden Falls (Lee 2001).
Groundhog Bay II
Groundhog Bay II is a multiple component site located in the Chilkat
Peninsula, about 65 kilometers west of Juneau, Alaska (Ackerman 1996;
Ackerman et al. 1979). The site is located on a glacio-marine terrace set back
from the present beach (Ackerman 1996). The elevated terraces at the site
reflect the general deglaciation of the region at the end of the Pleistocene
(Ackerman et al. 1979). Within the ancient beach sand and gravels of terrace
m, researchers found a series of cultural occupations dating between 9,200 and
4,200 years ago (the sites lower component). Basal dates (using two sigmas)
for Ground Hog Bay II site are 9130 130 BP (1-6304) and 9220 80 BP (SI-
21 12). One radiocarbon determination for Component HI dates to 10,180 + 80
BP (WSU-412) (Jordan 1992). However, due to the large standard deviation,
this date is not considered a reliable basal date (Dixon 1999).

Figure 4.1 Map of Southeast Alaska and Locations of 49-PET-408, Groundhog Bay, and
Hidden Falls. (Used with permission from E. J. Dixon).
Thick forest soil dating between 4,000 and 1,500 years ago covered the
lower component and represents a period when the site was not occupied.
Inhabitants reoccupied Terrace III at Ground Hog Bay II around 900 years ago
(Ackerman 1996). Groundhog Bay II is a very important discovery for several

reasons. It extended the chronological record for habitation in southeast Alaska
back at least 9,200 years, and established the existence of an early microblade
and core industry with bifaces during the early Holocene (Davis 1980).
In fact, the artifact assemblage from the lower component of Ground
Hog Bay II, including wedge-shaped microblade cores, microblades, a burin,
bifacial fragments, cobble choppers, hammerstones, and side scrapers
(Ackerman 1996), may suggest a possible pre-microblade assemblage for
southeastern Alaska (Jordan 1992). Most of the obsidian found at the site was
procured from Mount Rdziza (Jordan 1992), one of the two locations that
occupants of 49-PET-408 obtained their obsidian (Lee 2001). Excavators
found no evidence of hearths, structures, or other cultural features in the lower
two components, nor evidence of ochre.
Hidden Falls
The Hidden Falls site is a stratified, multiple component site located on
the northeastern shore of Baranof Island 30 kilometers northeast of Sitka,
Alaska (Davis 1980,1996). Hidden Falls was the second prehistoric site to be
excavated in southeast Alaska, following the excavation of Ground Hog Bay II
(Davis 1980). Hidden Falls contains three archaeological components:
Component I, dating to circa 10,000 BP, Component 13, dating between circa

4,620 + 110 BP [Beta-7442] and circa 3010 + 40 BP [SI-4505], and Component
III, which contains the most recent prehistoric occupation at the site, dating
between circa 3015 + 55 BP [SI-3974] and circa 1315 + 105 BP (Davis 1980).
Component I, the oldest component, has evidence of a microblade technology,
but no evidence of a transition to a ground-stone-and-bone industry (Davis
1996). Component II is considered within the Developmental Northwest Coast
Stage with a tool assemblage consisting of a ground-stone-and-bone industry,
including single-edged tools, abraders, hammerstones, ground-stone-and-bone
points, adzes, beads, labrets, and a predominately bipolar reduction flaked-stone
industry (Davis 1996). There is no evidence of microblades or microblade
cores from this component. Component III has more emphasis on the ground-
bone technology, and is considered in the middle phase of the Developmental
Northwest Coaist Stage.
Artifacts from Component I closely resemble those at Ground Hog Bay
II (Davis 1980). The assemblage consists of microblades and microblade cores,
scraping tools, split-cobble and split-pebble tools, choppers, hammerstones,
gravers, an abrader, burins, a unifacial blade, and utilized flakes (Davis 1980,
1996). Results of neutron activation analysis revealed that the obsidian from
Component I was recovered from Mount Edziza in northern British Columbia,
and from Suemez Island at the southern end of the Alexander Archipelago, west

of Prince of Wales Island. Inhabitants of 49-PET-408 procured obsidian from
both of these locations as well (Lee 2001).
Site 49-PET-408 (Dixon et al. 1997) is a multiple component cave site
located in the northwest portion of Prince of Wales Island in southeast Alaska.
As part of an extensive temperate rain forest expanding from northern
California to Kodiak Island, Alaska, Prince of Wales Island is characterized by
heavy rainfall and steep, thickly forested terrain (Dixon et al. 1997). The cave
is located on United States Forest Service land in a region unglaciated during
the late Pleistocene. Site 49-PET-408 lies approximately one kilometer from
the Pacific Ocean. Although the early Holocene paleoecology, glacial and sea
level history, and cultural chronology is largely unknown, 49-PET-408 provides
a unique opportunity to begin reconstructing the ecological and cultural history
of the area.
After the caves discovery in 1993, paleontologist Timothy Heaton from
the University of South Dakota sampled it for mammalian remains in 1994 and
1995 (Heaton et al. 1996). In 1996, Heaton revisited the cave with a preparator
from the Smithsonian Institution, Fred Grady. Grady found a stone tool during
the last week of June. One week later, on July 4th, Heaton discovered two bone

tools and several human bones.
The discovery of human skeletal remains prompted an immediate visit
to the site and immediate tribal consultations by Terence Fifield, United States
Forest Service (USFS) archaeologist for Prince of Wales Island, under
provisions of the Native American Graves Protection and Repatriation Act
(1990) (NAGPRA). Fifield immediately met with the Craig and Klawock tribal
councils, and described what had been discovered and the possible implications
of the discovery. In a joint tribal council meeting, arranged with the help of the
Central Council of Tlingit and Haida Indian Tribes of Alaska, representatives of
three tribes debated whether they were comfortable with scientific study of the
remains and further excavations at the cave. The tribes decided that study of
the remains and knowledge about this person was important to their people.
Two of the tribes passed resolutions supporting the studies within certain
conditions. The principal concern was that, should the site prove to be a burial
cave, excavations would cease and further consultation would ensue. In
further discussion it was determined that burial cave meant a site containing
the remains of more than one person or evidence of ceremonial internment.
The tribes also requested that any scientific information be shared with them
first, before being released to the public or the media.
Fifield notified E. James Dixon, then the Curator of Archaeology at the

Denver Museum of Natural History, about the human remains discovered in the
cave within days of the discovery. Dixon had been examining caves in
southeast Alaska for several years hoping to find evidence supporting the
coastal migration hypothesis (Dixon 1996) as a model for peopling of the
Americas. At the initial tribal consultation meeting Fifield presented an offer
from Dixon of the Museum s curation and laboratory facilities. Dixon visited
the cave during the summer of 1996. Archaeological excavations began in
1997 with permission and cooperation of the Klawock Cooperative Association
and the Craig Community Association and funding from the National Science
Foundation (NSF) along with support from the USFS (Dixon 1999).
Paleontological investigation led by Dr. Timothy Heaton of the
University of South Dakota ran concurrently with Dixons fieldwork.
Researchers conducted investigations at 49-PET-408 during the summers of
1997 through 2000 (Fifield, October 2002, personal communication). My work
at the site commenced during the summers of 1999 and 2000.
Site 49-PET-408 is assumed to be an Early Period site of the Northwest
Coast Microblade tradition (Dixon 1999; Lee 2001). Excavations of this multi-
component stratified cave site yielded eight major stratigraphic units (Figure
4.2). From the surface downward, the units are defined as: 1) Upper Forest
Organics, 2) Middle Forest Organics, 3) Lower Forest Organics, 4) Microblade

Horizon, 5) Diamicton, 6) Brown Silt, 7) Glacial Till, and 8) Bedrock
(limestone) (Dixon 1998). The majority of the cultural material found at 49-
PET-408 has been found in unit four (the Microblade Horizon). The material
cultural remains include bifaces; flake cores; blocky, poorly prepared cores
exhibiting sided microblade-like flake scars; true microblades and microblade
like flakes; waste flakes; blocky sandstone pebbles and cobbles; and ochre
(Dixon 1998). In addition, human remains from one individual were found
inside the cave.

Figure 4.2 49-PET-408 Excavation Units and Stratigraphy. (Used with permission from E.
J. Dixon)

Scientists have found that isotopic analysis conducted on animal
remains varies based on the animals diet. For example, the bones of animals
that relied solely on plants throughout their lives have different isotopic values
than bones of animals that relied on marine resources. Isotope analysis yielded
delta 13-C values of -12.5 o/oo (parts per million) for the human mandible
from 49-PET-408 and -12.1 o/oo for the pelvis. These isotopic values fall
within the same values of marine carnivores, which range from circa -15 o/oo
and -9 o/oo. Researchers from 49-PET-408 determined from these results that
the individuals diet consisted almost entirely of marine foods (Dixon 1999).
Samples of bone extracted from the human mandible and pelvis resulted
in AMS C-14 dates of 9,730 + 60 BP (CAMS-29873) and 9,880 50 BP
(CAMS-32038), respectively. Together, the dates suggest a C-14 date of 9,800
BP. However, since isotopic analysis has shown that the individual relied on a
marine diet, the C-14 determination must be adjusted (Dixon 1999:118).
Josenhans et al. (1995) have shown that in the Queen Charlotte Islands, there is
a 600 year C-14 difference in the regional marine and atmospheric carbon
cycles based on comparisons of C-14 determinations run on wood and bone.
Thus, the 9,800 BP C-14 date obtained from the human remains must be
adjusted by subtracting 600 C-14 years (Dixon 1999). The corrected date, circa
9,200 years BP, is contemporaneous with charcoal from unit four radiocarbon

dated to 8,760 + 50 BP (CAMS-43991), 9210 50 BP (CAMS-43990), and
9,150 50 BP (CAMS-439899).
The stone artifacts from 49-PET-408 were made from a variety of material,
including obsidian, quartzite, chert, and sandstone. Visually and through trace
element analysis, Craig Lee (2001) concludes that the color and translucency of
obsidian artifacts from 49-PET-408 correspond with specific source areas in
British Columbia and southeastern Alaska (Lee 2001). Lee categorized the
obsidian found at 49-PET-408 into two groups based on geographically distinct
source areas. He concludes that obsidian from 49-PET-408 was procured from
two separate locations: Mount Edziza (at least 200 air miles northeast of 49-
PET-408) and Suemez Island (approximately 65 air miles south of 49-PET-
408) (Figure 4.3). Lee suggests that differences between the obsidian artifacts
found at 49-PET-408 may suggest that the two materials were obtained or used
differently, and may have been procured by two distinct groups of people.
Although tentative, Lees analysis supports researchers longtime speculation of
prehistoric trade networks established in the Alexander Archipelago (Carlson
1979; Fladmark et al. 1990).

Figure 4.3 Map of 49-PET-408 and Obsidian Sources Mount Edziza and Suemez
Island (adapted from Moss 1998:89).
Several fragments of ochre also were found at the site. The ochre
artifacts are discussed thoroughly in the following chapters. Other analyses,
including spatial distribution analysis of flake types and material types, are
currently underway.

Archaeological sites such as 49-PET-408 provide archaeologists with
important information regarding early occupation of the New World. In the
Northwest Coast of North America, specifically southeast Alaska, little
attention has been paid to the Pleistocene/Holocene transition in archaeology
and paleontology (Dixon et al. 1997). This is primarily due to the lack of sites
reliably dated to earlier than 10,000 BP. Because of the scarcity of early
Holocene human remains in North America, research at 49-PET-408 provides a
rare opportunity to analyze and understand the material culture and physical
characteristics of early Northwest Coast peoples (Dixon et al. 1997). Dixon
utilized this opportunity by conducting archaeological excavations and analyses
to better understand this rare, early Holocene site. Research of the
archaeological material obtained from 49-PET-408 is ongoing.

The ochre specimens used in this study are housed at the Institute for
Arctic and Alpine Research (INSTAAR) at University of Colorado, Boulder,
Colorado. The ochre specimens were collected during summer excavations of
1997, 1998, 1999, and 2000 at 49-PET-408. Macroscopic and low-power
microscopic analysis was conducted at the University of Colorado at Denver
and the University of Colorado at Boulder to determine usewear. The
macroscopic and microscopic results are presented following the
methodological description of the low-power microscopic analysis in this
section. PIXE analysis was conducted at The Element Analysis Corporation,
Lexington, Kentucky on 10 of the 17 ochre specimens collected. Results from
the PIXE analysis follow the methodological description of PIXE analysis in
this section.
Specimen Recovery
Site 49-PET-408 (Dixon et al. 1997) is a multi-component site located
in the northwest portion of Prince of Wales Island in southeast Alaska.
Excavations began at 49-PET-408 following the discovery of human remains
inside On Your Knees Cave in 1996. Researchers at the site used a transit to

set up the one-meter by one-meter excavation units and unit datum. Each unit
was subdivided into quarter units (northeast, northwest, southeast, and
southwest). Workers used trowels to carefully excavate each quarter unit one at
a time. They also took measurements and constructed maps at each natural
level (as opposed to 10 centimeter levels.) All artifacts found in situ were
carefully mapped in their exact provenience measured from the southern edge
of the unit and the eastern edge of the unit for a positive provenience, x
centimeters north, y centimeters west. Depth of the artifacts also was recorded
using a unit datum. Researchers noted the direction and placement of the
artifacts, and the relation of the artifact to other artifacts found in the units by
carefully sketching diagrams and writing notes in their fieldbooks regarding
artifact placement. Each researcher at 49-PET-408 kept detailed fieldnotes to
meticulously document their progress.
Excavations at 49-PET-408 yielded eight stratigraphic units (see Figure
4.2). From the ground surface downward, die levels are composed of: 1)
Upper Forest Organics, 2) Middle Forest Organics, 3) Lower Forest Organics,
4) Microblade Horizon (or E Horizon), 5) Diamicton, 6) Brown Silt, 7)
Glacial Till, and 8) Bedrock. C-14 dating of charcoal specimens from the
Microblade Horizon (where much of the cultural material was found) suggests
that artifacts from this level were deposited over a period of several hundred

years, probably between circa 9,300 and 8,700 BP (Dixon 1998:6).
Soil from the individual quarter units was collected in buckets and
labeled by the unit number, range of depth, level type, name of the excavator,
and date. The individual levels were kept separate from other levels (for
example, soil from lower organics was put in a different bucket as soil from the
cultural level.) Workers then screened each bucket using a large, three-screen
filtering system, similar to a sluice box, set at a 35-degree angle down the slope
of a hill. The Sarge had three nested screens of progressively smaller mesh,
each set approximately two feet apart. The first screen had a large, two inch by
one-inch mesh mainly used to break up the soil and allow researchers to remove
large, non-cultural rocks. The second mesh had a smaller, %-inch mesh and
was also used to further separate the soil. The third screens 1/16-inch window
mesh trapped any artifacts missed during excavation. The buckets of soil were
individually dumped into the top screen. A watering hose was used to wet the
soil for easier passage through the screens. Workers documented the artifacts
general provenience based on the provenience information in die bucket.
To plot the provenience data of each ochre specimen on the site map,
analysis employed GIS software. The site map plotted a northing and westing
location of each ochre specimen, in addition to depth (Figure 5.1 shows the
provenience of the ten ochre specimen sent for PIXE analysis.) Field notes

from excavations were thoroughly examined to determine the types of
archaeological material found in close proximity to the ochre specimens.
Macroscopic Ochre Analysis
Macroscopically, the 49-PET-408 ochre specimens were analyzed for
usewear and compared to Munsells soil color charts for differences in color,
lightness, and hue. The Munsell system matches the specimen in question
against a set of color specimens. This process leads to a three-part designator
for the appearance of a given specimen that characterizes both tint and hue.
Each specimen was examined under a Leica light source.
Following the Munsell analysis, each ochre specimen was placed on
individual paper squares and weighed using an Acculab digital scale. Of the 17
ochre specimens found during excavations, 14 are eligible for PIXE
examination based on weight.2 Only 10 specimens were sent for PIXE analysis
due to financial restrictions.3 The largest of each specimen was measured using
Mitutoyo Digimatic calipers. If unable to use tweezers to handle the ochre, the
examiner used rubber gloves.
Only specimens weighing .5 to one gram can be used for PIXE analysis.
Three of the four remaining specimens were not chosen for analysis based on their weight Specimens weighing only
0.5 grams or slightly above would have to be completely pulverized. This would destroy the entire specimen and
render any future ochre analysis impossible. The fourth specimen was not chosen because it is unclear whether the
specimen is actually ochre. Although reddish in color, the hard, flat material resembles a different type of rode material
rather than ochre.

Pictures were then taken to visually document the specimens prior to
sending them forPIXE analysis (Figures 5.2 5.19). All specimens were
photographed at INST AAR using an Olympus E-10 4.0 Mega Pixel and an
Olympus macro-lens. Adobe Photoshop software digitally formatted the
photographs for easy conversion to prints.
During the ochre analysis, great care was used to avoid cross-
contamination of the specimens. The paper trays used for weighing each
specimen were discarded after each use. The examiner also used new rubber
gloves to handle each specimen if necessary.