United States

(Topik et al. 1986:19). Representative timber types of this vegetation zone in the vicinity of the Packwood Lake site include Douglas-fir (Pseudotsuga menziesii), western hemlock (Tsuga heterophylla), and western redcedar (Thuja plicata). Assorted understory vegetation present near the lake include vine maple (Acer circinatum), dwarf Oregon grape (Berberis nervosa), sword fern (Polystichum munitum), bracken fern (Pteridium aquilium), huckleberry (Vaccinium spp.), and wild strawberry (Fragaria virginiana) (Honcharsky n.d.; Topik et al. 1986).

Fauna noted to occur in the vicinity of Packwood Lake include deer (Odocoileus hemionus), elk (Cervus elaphus), cougar, (Felis concolor), bobcat (Felis rufus), black bear (Ursus americanus), red fox (Vulpes vulpes), Douglas' squirrel (Tamiasciurus douglasii), osprey (Pandion haliatus), bald eagle (Haliaetus leucocephalus), golden eagle (Aquila chrysaetos), red-tailed hawk (Buteo jamaicensis), Swainson's hawk (Buteo swainsonii), Cooper's hawk (Accipiter cooperii) and the common raven (Corvus corax) (Honcharsky n.d.; Schamberger n.d., 1970; Taylor and Shaw 1927). A native species of rainbow trout, "Packwood Rainbow" (Salmo gairdneri) is also present within the waters of Packwood Lake (Lucas 1989; Honcharsky n.d.).

Portions of the Packwood Lake shoreline are bordered by the Goat Rocks Wilderness. Geological mapping of the wilderness and the adjacent Goat Rocks Roadless Areas has been

done by Swanson and Clayton (1983). In the Goat Rocks area, Cenozoic volcanic rocks overlie a pre-Tertiary basement (Swanson and Clayton 1983). The oldest rocks belong to the Russell Ranch Formation (Simmons 1950) of the Rimrock Lake pre-Tertiary inlier composed primarily of graywacke and argillite. Minor occurrences of pillow basalt, chert, and thin tuffaceous beds are present within this formation. The Ohanapecosh Formation of the early to middle Oligocene rests unconformably on the Russell Ranch Formation and consists mainly of altered basaltic and andesitic lava flows and bedded andesitic and dacitic volcanoclastic rocks (Swanson and Clayton 1983). Andesites, of late Pliocene and early Pleistocene age occur on many of the high ridges in the area. Late Pleistocene basalts are reported for the area near the southern boundary of the Goat Rocks Wilderness (Ellingson 1968, 1972; Swanson and Clayton 1983). Holocene volcanic eruptions of Mount St. Helens, Mt. Rainier, Mount Mazama, and Glacier Peak have deposited tephra throughout the area (Crandell 1969; Mullineaux and Crandell 1981; Mullineaux et al. 1975; Yamaguchii 1983, 1985).

Packwood Lake was originally considered to have been formed behind a terminal moraine left by the retreat of the Upper Lake Creek valley glacier and subsequently buried by debris from a late Pleistocene landslide from nearby Snyder Mountain (Harrison and Powell 1977). However, recent data suggest that the formation of the lake was much more recent

than had previously been suspected. Radiocarbon dates of 1,140 ± 60 B.P. (Beta-32674) and 970 ± 60 B.P. (Beta-32673) obtained from snags buried by the landslide indicate that Packwood Lake was impounded by the Snyder Mountain landslide approximately 1,000 to 1,100 years ago (Schuster 1990). This event presumably resulted in the entrapment of fish which established the native trout population present within the lake.

Ethnographic Background

Packwood Lake lies within lands assigned ethnographically to the Taidnapam (also Taitnapam or Upper Cowlitz), a northern Sahaptin speaking group that inhabited the upper Cowlitz River drainage (Curtis 1913; Gibbs 1855; Ray 1974; Smith 1953, 1964; Spier 1936; Teit 1928). There is some suggestion that the Taidnapam moved into the upper Cowlitz River area from east of the Cascade Mountain crest during the middle of the nineteenth century (Ruby and Brown 1986:234; Teit 1928). The ethnographer Melville Jacobs (1934) conducted extensive interviews with Taidnapam individuals during the period from 1926 to 1928. According to Jacobs' Taidnapam informants, the Taidnapam had lived within the upper Cowlitz River region for as long as could be remembered. Jacobs attributed the linguistic affiliation of the Taidnapam to Kittitas-Sahaptin

speaking groups east of the Cascades as the result of language drift rather than migration (Jacobs 1937).

The Taidnapam often were identified by neighboring groups such as the Klickitats located to the south, as "wild" or "wood" indians (Ruby and Brown 1986:234). Various myths--such as the stealing and eating of children and the ability to travel unseen (Gibbs 1855)--were also associated with the Taidnapam. The origin of these myths is unclear.

Ethnographically, the Taidnapam occupied several villages within the upper Cowlitz River valley. These villages were occupied mainly from October to May and were usually associated with fishing sites. The availability of seasonal resources at various locations within the upper Cowlitz River drainage and beyond resulted in the scheduling of procurement activities away from the villages (cf. Jacobs 1934; Smith 1964).

The Taidnapam made forays to Packwood Lake (Kiona 1964, 1965; McClure 1987) to procure a native species of trout that lives within the lake (cf. Anonymous 1909; Kiona 1964, 1965; Jacobs 1934). According to several sources, trout spawning occurred from late June to early July in at least four small creeks that flow into the lake (McClure 1987; PCSP 1954) as well as along the shoreline of the outlet bay in late May (Lucas 1989:166). Several Taidnapam informants have indicated that they or members of their family visited the lake specifically to obtain trout while the fish were spawning in

the small streams (Kiona 1964, 1965; Jacobs 1934; McClure 1987). Techniques utilized by the Taidnapam to catch fish involved damming and channeling water out of the streambeds. This resulted in the streambed below the channel to drain dry, effectively stranding the fish. The trapped fish were gathered into baskets, later to be dried over open fires. Some of the fish were roasted and eaten while processing was in progress (Kiona 1964; McClure 1987).

Additionally, huckleberries were obtained from the Packwood Lake area. Speaking through a translator, Mary Kiona, a Taidnapam informant, related that "They used to go to Packwood Lake and they used to pick huckleberries there a long time ago, and different ones from around here" (Kiona 1965).

The Packwood Lake region also provided access to resources within the Goat Rocks area. These resources included mountain goats and an obsidian source at Elk Pass, 45LE286 (McClure 1989), approximately eleven kilometers southeast of the Packwood Lake site. Available ethnographic and ethnohistoric literature indicate specifically that mountain goats were hunted by the Taidnapam within the lands now designated as the wilderness area (Jacobs 1934; Kehoe 1890; Kiona 1965; Sethe 1938, McClure 1989). William O. Douglas (1950:207) was apparently aware of an obsidian source within the Goat Rocks region. Douglas ventured into the Goat Rocks area as early as 1915 and found "...the obsidian rock the Indians used for arrowheads."

Ethnographic evidence suggests that the Taidnapam exploited the Packwood Lake area during late spring and summer. The Packwood Lake site apparently was used primarily for catching and processing trout and also as a base camp for gathering huckleberries and hunting mountain goats.

CHAPTER 3

PACKWOOD LAKE: LITHIC ANALYSIS

Methodology

The analytical approach used herein is based upon the principles of replicative systems analysis (Flenniken 1981, 1984). Replicative systems analysis relies upon data experimentally derived to demonstrate the entire lithic reduction processes. These experimental data are then used to elucidate the lithic reduction activities indicated by the archaeological assemblage (Flenniken 1981:3-5).

Lithic debitage from 45LE285 was analyzed utilizing technologically derived flake attributes denotative of specific experimental reduction stages (cf. Gilreath 1983; Flenniken and Ozbun 1988, Flenniken et al. 1989; Raymond 1986; Scott et al. 1986). Debitage attributes were correlated with experimental flintknapping analogs in order to propose a reduction model for prehistoric flintknapping activities. While the reduction of stone during the manufacture of tools is generally considered to be a continuum (cf. Collins 1975; Muto 1971:109; Flenniken 1984:192), stage analysis (cf. Gilreath 1983) has proven a useful tool for the study of prehistoric lithic assemblages (cf. Flenniken and Ozbun 1988; Flenniken et al. 1989, 1990; Ozbun 1987; Raymond 1986; Stanfill 1976).

A fundamental premise of the technological analysis of the Packwood Lake site lithic assemblage is that techniques of stone tool manufacture are the result of culturally learned behaviors (Collins 1975; Flenniken 1984, 1985; Young and Bonnichsen 1984, 1985) which are patterned within culturally constrained, or bound set of acceptable behaviors (Flenniken 1984). Therefore, debitage resulting from the production of stone tools would be expected to reflect this patterning within cultural limitations and to enable reconstruction of prehistoric lithic reduction activities through analyses employing a technologically oriented approach (Flenniken 1984, 1985).

Each piece of lithic debitage and formed artifact recovered from the Packwood Lake site was examined from a technological perspective. All artifacts were classified according to lithic raw material type and technological category. In order to make the text less cumbersome, the technological terms and their definitions used in the following discussions are explained in detail in Appendix A and B and are from Flenniken et al. (1990). Technological attributes such as dorsal flake scar configuration and frequency, evidence of heat treatment, fracture type, and presence of an original detachment scar also were noted during the analysis. This information was combined to identify the frequency of lithic reduction activities that occurred prehistorically at the Packwood Lake site.

Tabulations of technological categories were generated for each level and excavation unit within the site in an effort to determine the density distribution of the assemblage and reveal any detectable horizontal or vertical clustering indicative of individually distinct site components. Because no significant aberrations were noted in either the vertical or horizontal distribution and frequency of technological categories and raw materials, the assemblage was analyzed as unit representative of the normative activity remains existing at the site.

Assemblage Parameters

The analytical assemblage represents the prehistoric cultural material recovered from five one-by-one meter horizontal excavation units on the south side of Lake Creek (Figure 7) screened through three-millimeter screen. The use of this smaller screen size, rather than the standard six-millimeter screen, results in a higher recovery rate of technologically diagnostic debitage (cf. Kalin 1981).

Experiments have demonstrated that as much as ninety-eight to one hundred percent of the debitage associated with projectile point manufacture was lost as a result of the use of a six millimeter screen (Kalin 1981). Additional experiments have demonstrated that debitage associated with projectile point rejuvenation is also subject to the same loss

Figure 7. Location of August 1987 USDA Forest Service test excavation units on the south side of Lake Creek.

when the larger (i.e., 6 mm) screen size is used (Towner and Wharburton 1985). Comparisons of archaeological collections recovered from the Judd Peak site (45LE222) poignantly illustrate the differences in assemblage composition associated with the use of three millimeter versus six millimeter screen. One portion of the site, Judd Peak North, was excavated utilizing six-millimeter screens and the other portion, Judd Peak South, utilized three-millimeter screens. When compared to the use of six-millimeter screen, the three-millimeter screens resulted in a significant increase in the recovery of pressure flakes (6 mm screen, n=168 versus 3 mm screen, n=1563) and microblades (6 mm screen, n=21 versus 3 mm screen, n=1,103) from the two rockshelter areas of the site (Daugherty et al. 1987a:61).

The Packwood Lake Assemblage

The prehistoric lithic assemblage analyzed from the Packwood Lake site consists of 4,242 artifacts including formed artifacts (n=72) and debitage (n=4,170) mainly from excavated levels below a primary deposit Mount St. Helens Wn tephra which dates to A.D. 1479-1480 (Yamaguchii 1983, 1985) and on top of the Snyder Mountain landslide debris estimated to be have been deposited 1,000 to 1,100 years ago (Schuster 1990). Approximately 34.4% of the debitage assemblage was considered technologically diagnostic; 65.6% (n=2,737) of the

debitage was undiagnostic (i.e., the fragmentary nature of the debitage precluded assignment to any technological category). Prehistoric cultural materials recovered from previous excavations (McClure 1987) on the north side of the pedestrian foot bridge across Lake Creek (Figure 8) were not used for this analysis because most deposits represented fill material and analytical assemblages recovered from utilization of two different screen sizes at one site (cf. Daugherty et al. 1987a) created an interpretive dilemma. Artifacts associated with original bridge construction sediments (i.e., fill material) were also excluded from the analysis because of questionable provenience.

Raw Material Types. For the purpose of this analysis, granular varieties of microcrystalline quartz, including chert and jasper, were combined into a single analytical category, chert/jasper (cf. Klein and Hurlburt 1985). The only distinction made in this category is the differentiation of heat-treated chert and jasper (CTJHT) versus nonheat-treated chert and jasper (CTJ). Chert and jasper are opaque forms of microcrystalline quartz composed of numerous grains that form a granular crystalline structure (cf. Klein and Hurlburt 1985:439-445). Chert and jasper are chemically precipitated sedimentary rocks which are classed as microcrystalline but may contain sheaf-like aggregates which may include impurities such as clays, silts, carbonates, pyrites, or other organic materials. Chert and jasper also may contain several forms of

Figure 8. Location of April 1987 USDA Forest Service test excavation units on north side of Lake Creek.

silica such as opal, chalcedony, or microcrystalline quartz. Cherts range in color from white to light gray to black. Jasper, which is red, yellow, green, or brown, is distinguished from chert only on the basis of color (Klein and Hurlburt 1985).

Fibrous microcrystalline varieties of quartz, included heat-treated (CLHT) and nonheat-treated (CLT) chalcedony, and were analyzed as a separate group. Chalcedony (cf. Klein and Hurlburt 1985) is a translucent, fibrous microcrystalline variety of quartz with a waxy luster and ranges in color from clear to white, red, brown, and black. Agate is a variety of chalcedony with alternating layers having different colors and porosities (e.g., onyx and sardonyx). Colored varieties of chalcedony include carnelian (red), chryroprase (apple-green), and heliotrope or bloodstone (green, with small red spots of jasper). Often the clear varieties have dark inclusions in the form of lines or plant-like impressions called dendrites (e.g., moss agate) commonly composed of manganese oxide. Chalcedony is composed of microcrystalline sheaf-like bundles of radiating fibers and is deposited from aqueous solutions which are found lining or filling cavities in other rocks.

Obsidian is locally available from the Elk Pass quarry site, 45LE286, eleven kilometers to the southeast of the site (McClure 1989). Obsidian (cf. Skinner 1983) is an igneous glass that is non-crystalline and often has a bright vitreous luster. Obsidian has a specific gravity of 2.4 and is

slightly harder than common window glass (H=5.5). It forms as a result of the rapid cooling of extrusive magma, preventing any crystalline growth, or as a result of a viscous magma which was too rigid to support crystalline growth. Obsidian is usually black, but gray, red, brown, green and even blue are common in specific geological locations in western North America. The red colors are due to the inclusion of magnetite or hematite. Obsidian usually is translucent but can occur as transparent (e.g., obsidian from Glass Butte, Oregon) or opaque (e.g., that from McKay Butte, Oregon) varieties (J. Jeffrey Flenniken, personal communication 1990). Samples of archaeological obsidian from the Packwood Lake site that were submitted for x-ray fluorescence spectrometry analysis indicated at least a portion of the obsidian originated from the Elk Pass quarry (McClure 1989).

Andesite and basalt were combined for the analysis as andesite/basalt (AND/BA) since these materials could not be visually differentiated. Andesite and basalt are available as both primary and secondary sources throughout the area. Andesite is especially prevalent throughout the Packwood Lake and Upper Lake Creek basins (Shawn Jones 1987). Pyroxene andesite outcrops in the Upper Lake Creek basin (cf. Swanson and Clayton 1983) have been indicated as a possible material source for some archaeological specimens from the Packwood Lake site lithic assemblage (Rick Wooten, personal communication 1987).

The final category includes only two pieces of debitage characterized as representative of sedimentary rock (e.g., siltstone or mudstone). Because of the infrequent occurrence in the assemblage, the significance of this material in the lithic reduction technology is unknown.

The first task necessary for the production of flaked stone tools is the acquisition of suitable lithic raw materials (cf. Crabtree 1967). Flakeable stone is widely available in many dispersed localities of the upper Cowlitz River drainage. Microcrystalline varieties of quartz, including chert, jasper, and chalcedony, are found in many secondary depositional environments such as the Cowlitz River and its tributaries. Primary deposits of chert and obsidian previously have been noted to occur in the Goat Rocks area. Andesite and basalt are also widely available in many primary and secondary source contexts throughout the area (Shawn Jones 1987).

Heat Treatment. The thermal pretreatment of silica materials was an important process in the successful reduction of non-obsidian materials and is well-documented in the experimental literature (cf. Crabtree and Butler 1964; Draper and Flenniken 1984; Flenniken and Garrison 1975; Flenniken and White 1983; Mandeville and Flenniken 1974; Purdy 1974; Rick 1978), archaeological literature (cf. Cole and Leonhardy 1964; Irwin-Williams and Irwin 1966; Shippee 1963), and ethnographic literature (cf. Barnett 1937:169; Goldschmidt 1951:419;

Grinnell 1898:142; Smith 1940:323; Stewart 1941:289,383; 1942:264; McAdams in Streuver 1970:50; Powell 1875:27-28; Watson 1950; Webster 1889:602). For the purpose of this study, identification of intentionally heat-treated lithic materials from the Packwood Lake site was based upon visual inspection and comparison with both raw and experimentally heat-treated examples of chert, jasper, and chalcedony obtained from sundry locations within the upper Cowlitz River drainage. The most obvious difference between this sample of raw and heat-treated materials is an increase in surface luster of the latter exhibited on a freshly flaked surface. Color was not used an indicator for the identification of heat-treated materials within the assemblage because of the poor correlation with color and the effects of heat treatment (Flenniken 1979; cf. Johnson 1985). Furthermore, some locally available, and thermally unaltered materials, exhibit several different colors within a single cobble.

Lithic Reduction. Approximately 34.4% (n=1,433) of the lithic assemblage (n=4,242) from the Packwood Lake site exhibits technologically diagnostic reduction attributes that were used to assign debitage to reduction stages for a reconstruction of the lithic reduction activities that occurred (Appendix A and B). These attributes include platform type and method of preparation, evidence of heat treatment, fracture type, and presence of an original detachment scar. The relatively low frequency of

technologically diagnostic debitage within the assemblage apparently results from differential effects of fire or excessive heat to various lithic raw materials (cf. Purdy 1974). Examples of crenated fractures, "potlidding," and crazing of lithic debitage and formed artifacts are frequent in this assemblage, especially among the chert/jasper and very fine-grained andesite/basalt materials. The occurrence of potlids (PL-Potlid, n=44) seems to support this conclusion. The high frequency of heat damage observed in the assemblage may have been the result of cultural factors, such as the repeated reuse of fire at the site area through time, or the result of natural agencies such as forest fires.

In spite of the apparent uniformity of the lithic assemblage, it has been divided into two groups based on differences in raw material and end product. The first reduction trajectory includes obsidian, chert/jasper, and chalcedony (n=1911). Andesite/Basalt (n=2329) represents the second reduction trajectory. The material category "other" was represented by only two specimens; a secondary decortication, single facet, platform flake (SD-SFP=1) and a broken secondary decortication flake that is missing the platform (SD-PA=1). Therefore, its significance in the lithic technology of the site is indeterminate and will not be discussed further.

Data: Results of Analysis

Raw Material Types. Five major categories of toolstone were identified from the technological analysis of the Packwood Lake site assemblage as discussed above. The category occurring most frequently is andesite/basalt (n=2329) which accounted for 54.9% of the analytical assemblage. Chert/jasper (n=1,025) represented 24.2% of the assemblage. Obsidian (n=595) 14.0%, chalcedony (n=291) 6.9%, and other (n=2) 0.05%, comprised the remaining identified lithic materials (Figure 9).

The infrequent occurrence of cortex (n=74) on artifacts from the Packwood Lake lithic assemblage indicates most primary reduction activities occurred elsewhere (cf. Gould and Sagers 1985; Mathis 1976; Wharburton 1980; Stanfill 1976). Cortical types that are represented within the assemblage suggest both a primary (primary geological cortex) and a secondary source (incipient cone cortex) for acquisition of lithic raw materials. Debitage associated with the limited on-site decortication of toolstone obtained from primary source contexts suggests the materials were tabular in form as indicated by secondary decortication alternate flakes (SD-Alt, n=7: Table 1) associated with the removal of square edges (note: abbreviations used in both the text and Tables 1 and 2 are defined in Appendix A and B). These flakes have single facet platforms; an extreme left or right platform

Figure 9. Frequency of lithic material types (debitage) from the Packwood Lake site assemblage.

Table 1. Frequency of Debitage Categories^a from Packwood Lake

Category		Materials^b							Totals
	OB		CTJ	CTJHT	CL	CLHT	AND/BA	OTHER	T
PD-NP						1			1
PD-SFP
PD-MFP							1		1
PD-MFP/AP
PD-PA
PD-OP
PD-Alt
SD-NP	1						3		4
SD-SFP				1		3	8	1	13
SD-MFP							5		5
SD-MFP/AP							1		1
SD-PA	3						3	1	7
SD-OP
SD-Alt						1	6		7
SD-BulbR
I-NP	1						12		13
I-SFP	25		5	34		18	66		148
I-MFP				2			9		11
I-MFP/AP							2		2
I-PA	3			6			7		16
I-OP
I-Alt				1		1	7		9
I-BulbR
BT-BulbR
BT-Alt	1			2			8		11
BT-EdgeP	4			8		2	39		53
BT-MargR	5			5		2	14		26
BT-OP
BT-EPerc	47			69		19	238		373
BT-LPerc	8			7		1	30		46
BT-EPres	146			187		72	160		565
BT-LPres	22			44		23	31		120
BT-Notch				1					1
BT-DS	3			8		3	1		15
PL-Potlid				26		6	12		44
SH-Cort	1								1
SH-NoCort	3		1	5		9	1		19
UFF-Cort	7		1	3		1	23		35
UFF-NoCort	315		9	576		125	1613		2638
TOTALS	592		16	977		284	2299	2	4170

^aSee Appendix A for List of Abbreviations and Appendix B for Glossary of Terms.

^bOB = obsidian; CTJ = chert/jasper; CTJHT = heat-treated chert/jasper; AND/BA = andesite/basalt.

orientation, and a flat cortical surface along one lateral flake edge perpendicular to the platform.

Heat Treatment. The results of the analyses indicate that 98.7 % (n=1299) of the debitage and formed artifact assemblage of microcrystalline varieties of quartz (i.e., chert, chalcedony, and jasper) exhibit characteristics of heat treatment. Lithic raw materials appear to have been heat-treated prior to the completion of the decortication process as evidenced by primary and secondary decortication flakes (n=10) from heat-treated materials, probably in response to the lack of large pieces of high quality toolstone (cf. Daugherty et al. 1987a, 1987b). It is likely that flake blanks and exhausted cores were also heat-treated, but this does not appear to have occurred at the site.

Reduction: Obsidian, Chert/Jasper, and Chalcedony. A variety of prehistoric lithic reduction activities are suggested at Packwood Lake. Stage 1CR, core reduction (Figure 10), indicates the production of flakes from unprepared cores presumably for use as expedient cutting tools (cf. Binford 1973, 1977, 1979; Bamforth 1986), and is indicated by numerous interior single facet platform flakes
(I-SFP=82: Table 1) as well as exhausted flake cores (RM-Exh=4: Table 1). Flakes were also produced by a bipolar technique from smaller subrounded to rounded nodules of heat-treated toolstone (cf. Crabtree 1972; Flenniken 1981; Flenniken and White 1985) as evidenced by the recovery of two bipolar cores (BP-Core=2:

Figure 10. Analytical reduction stage frequencies for obsidian, chert/jasper, and chalcedony.

Table 2). These bipolar cores exhibited crushed platforms, flat flake scars with pronounced compression rings, and bi-directional flake scars. Debitage samples were examined microscopically at both 80x and 160x magnification, but no polish attributable to use was observed. Therefore, 1) utilization of flake edges for cutting did not result in significant edge damage at the site; 2) flakes that were use for cutting were not recovered in the archaeological sample; or, 3) unmodified flakes were not used as cutting implements

Larger flakes (i.e., flake blanks) were then reduced further. The Packwood Lake assemblage exhibited a low frequency of original detachment scars (n=16), which are defined as remnants of the original ventral surface of a flake blank (Scott et al. 1986:10). Also absent is debitage relating to flake blank preparation, such as debitage associated with the removal of the bulb of percussion from the flake blank (Flenniken 1987:21) and preparation of the flake blank edges or edge preparation flakes (Appendix B). The absence of these technological categories suggests the production and subsequent reduction of flake blanks was not major production trajectory at the site. It is likely that flake blanks were being produced at the site, fragments of those broken in production would be expected to occur within the assemblage (cf. Flenniken et al. 1989:40).

Table 2. Frequency of Formed Artifacts^a from Packwood Lake.

Category		Materials							Totals
	OB		CTJ	CTJHT	CL	CLHT	AND/BA	OTHER	T
RM-Unalt
RM-Test
RM-FC
RM-Exh	2			1			1		4
BP-Core				2					2
B-CompDS
B-FragDS
B-CompBT
B-FragBT
B-Comp
B-Frag							9		9
P-CompDS				1					1
P-FragDS				1		1			2
P-CompBT
P-FragBT				1					1
P-Comp
P-Frag				3					3
LP-Comp
LP-ComRej
LP-ComExh							2		2
LP-Prox
LP-Med
LP-Dist
DP-Comp
DP-ComRej
DP-ComExh				1					1
DP-Prox				1					1
DP-Med
DP-Dist				1					1
AP-Comp				1					1
AP-ComRej
AP-ComExh
AP-Prox
AP-Med
AP-Dist
Flake Tool							1		1
Perf/Drill-CompEx				1					1
UF-Comp				1					1
UF-Frag				4		2	2		8
UBF	1		1	12		4	15		33
TOTALS	3		1	31		7	30		72

^aSee Appendix A for List of Abbreviations and Appendix B for Glossary of Terms.

The bifacial reduction of flakeable toolstone is characterized by numerous examples of Stage 3EC (Figure 10) early percussion, bifacial thinning debitage (BT-EPerc, n=135: Table 1). Early stage bifacial thinning flakes exhibit multifaceted platforms, have few dorsal flake scars, and are slightly curved or twisted in long-section (Appendix B). The low frequency of Stage 4LC (Figure 10), late stage percussion bifacial thinning debitage (BT-Lperc, n=16: Table 1) suggests that bifaces were being initially shaped during Stage 2BFT biface preparation (BT-EdgeP, n=14; BT-MargR, n=12; BT-Alt,. n=3:Table 1). Final Stage 4LC (late stage bifacial thinning) thinning by percussion was not an important on-site activity. Late stage bifacial thinning flakes exhibit multifaceted platforms, numerous dorsal flake scars, and are almost flat in long-section (Appendix B). The resulting biface blanks, defined as bifaces manufactured by percussion, were probably removed from the site. No complete blanks were recovered from the site. Blank configuration, in some instances, did not necessitate final percussion thinning to alter the width-to-thickness ratio (i.e., make the biface thinner), therefore it is likely that some bifacial blanks were further thinned by pressure flaking.

Stage 5ES, early stage pressure flake frequency (EPres, n=405: Table 1), versus Stage 6LS (Figure 10) late stage (LPres, n=89: Table 1), reflects the initial preparation of preforms from prepared blanks. Early stage pressure flakes

(Appendix B) are the result of regularizing a biface by pressure, have multiple dorsal flake scars, and are twisted in long section. They are small, and their platforms form an oblique angle with the long axis of the flake. Late stage pressure flakes are small, parallel-sided, and have one dorsal arris. The sporadic occurrence of debitage associated with flake blank preparation (e.g., bulb removal flakes, n=0; and original detachment scars, n=16:Table 1 and 2) indicates most of this preparation occurred elsewhere. Flake blanks are defined as flakes that were further reduced by percussion and/or pressure for the manufacture of bifacial or unifacial tools. Successfully prepared preforms, defined as pressure flaked bifaces, were transported away from the site rather than reduced further into notched projectile points; only one flake that had been removed in formation of a notch (cf. Titmus 1985) was recovered (BT-Notch, n=1: Table 1). Preform fragments (P-Frag, n=6: Table 2), one exhibiting a perverse fracture (cf. Crabtree 1972; Titmus and Woods 1986), and one complete but discarded dart point preform (P-CompDS: Table 2), are indicative of a preform production trajectory.

Additionally, the frequency of early stage pressure flakes (Stage 5ES), as well as the infrequent occurrence of broken and/or discarded projectile points (LP-ComExh, n=2; DP-ComExh, n=1; DP-Prox, n=1; DP-Dist, n=1; AP-Comp,n=l:Table 2), indicate that pressure flaking activities were probably not associated with rejuvenation (Flenniken et al. 1989, 1990),

though doubtless some of this did occur. Rejuvenation debitage, especially pressure flakes associated with the repair of right-angle fractures, is characterized by an extreme right or left platform orientation, single faceted platforms, and a flat surface along one lateral flake surface perpendicular to the platform (Towner and Wharburton 1985). Pressure flakes diagnostic of rejuvenation were not observed in the assemblage. Instead, pressure flaking of chert, chalcedony, jasper, and obsidian at the Packwood Lake site was related predominately to preform manufacture, and to a limited extent, to tool manufacture and maintenance.

Debitage, produced at any stage of the reduction sequence and considered to be of sufficient size for further modification, was recycled into a reduction trajectory for the production of flake tools (e.g., unifaces). This is represented in the assemblage by the occurrence of unifaces (UF-Frag, n=8; UF-Comp, n=1: Table 2) and one complete specimen manufactured on an interior flake (Flake Tool, n=1: Table 2).

Reduction: Andesite/Basalt. The reduction of andesite/basalt varies somewhat from the reduction of other raw materials, in part because of differences in end products. Analogous to the reduction of obsidian, chert/jasper, and chalcedony, numerous single-facet platform flakes (I-SFP, n=66) indicate Stage 1CR (Figure 11) flake production likely for use as expedient cutting tools. The absence of debitage

Figure 11. Analytical reduction stage frequencies for andesite/basalt.

associated with the preparation of a flake (i.e., flake blank) for further reduction (e.g., bulb removal flakes and flake blank edge preparation flakes) and near absence of original detachment scars suggests a flake blank-to-biface reduction trajectory was not employed with any significant frequency.

The tabular shape of the raw material appears to have significantly affected the direction of the andesite/basalt reduction trajectory. Bifacial reduction of tabular material is evidenced by Stage 2BFT debitage particularly alternate flakes (I-Alt, n=7), and alternate flakes that exhibit cortex (SD-Alt, n=6) as well as several blank fragments (percussion biface fragments) manufactured from tabular material. These artifacts, exhibiting sinuous margins characteristic of alternate flaking (Crabtree 1972), were apparently broken while creating bifacial margins by use of an alternate flaking technique (Crabtree 1972). One specimen retains primary geological cortex on both faces and was prehistorically abandoned because of a production failure (i.e., fractured by end shock) having occurred early in the reduction sequence. This apparently happened when the flintknapper unsuccessfully attempted to remove an area of high mass resulting from a series of stacked step fractures (cf. Crabtree 1972). The size of the desired end product evidently resulted in the discarding rather than lateral recycling of this material. Additional biface fragments of similar size that were broken later in the reduction sequence tend to support this inference.

The frequency of Stage 3EC, early (BT-EPerc=238) versus Stage 4LC, late (BT-LPerc=30) bifacial thinning flakes is suggestive of the initial preparation of bifaces. The characteristics Stage 2BFT, edge preparation debitage (BT-EdgeP=39) is the result of "setting up" the margins for the detachment of flakes by turning the edge of the biface by the removal of small percussion flakes from one face to move the margin towards the side to be thinned. Turning the edge in effect prepares, or "sets up" the bifacial margin for further flake removal from the opposite face of the biface (Knowles 1953:37-42). Fragments of percussion flaked bifaces (bifacial blanks, n=9) suggests biface blank production was a primary focus of the reduction trajectory. Stage 4LC final percussion thinning was either not necessitated because the desired end product (e.g., biface blanks) was manufactured to be transported away from the site and/or the biface did not require additional thinning by percussion prior to pressure reduction in order to produce a desired end product (e.g., preforms and projectile points).

Pressure flake frequency also indicates early rather than later stages of reduction. Stage 5ES, early stage pressure flakes, account for 83.8% (BT-EPres=160) of the total pressure flake assemblage compared to 16.2% of Stage 6LS late stage pressure flakes (BT-LPres=31). The occurrence of more early than late stage pressure flakes is likely representative of the initial manufacture processes rather than the result of

rejuvenation (Flenniken et al. 1990). Rejuvenation pressure flaking debitage is often the result of the rebeveling of broken edges by alternate flaking of the squared edge. The debitage produced during this process often has a distinct shape including an extreme platform orientation, single faceted platforms, and a flat surface along one lateral flake surface perpendicular to the platform (Towner and Wharburton 1985:5-6), none of which was detected in the assemblage.

Formed Artifacts. The most frequently represented formed artifact (n=72) category identified from Packwood Lake are unifaces (UF-Frag=33; UF-Comp=1). All of these were manufactured from interior flakes, but the fragmentary nature of the collection precluded ascertaining the reduction stages during which each flake was produced. Lithic material utilized includes obsidian (OB=1), chert/jasper (CTJ=1; CTJHT=12), chalcedony (CLHT=4), and andesite/basalt (AND/BA=15). The tools probably represent an expedient technology, manufactured, used, and discarded at the site (cf. Binford 1973, 1977, 1979; Bamforth 1986). Examination under 80x and 160x magnification revealed edge polish on a number of specimens. This was the only criterion used for the determination of use-wear (cf. Flenniken and Haggarty 1979:213).

Projectile points (Figure 12a, 12b, 12c, and 12e) are represented by one chert (CTJHT) arrow point (AP-Comp, n=l:Table 2), one chert (CTJHT) dart point (DP-ComExh, n=1),

two chert (CTJHT) dart point fragments (DP-Prox, n=1; DP-Dist, n=1:Table 2), and two andesite/basalt (AND/BA) lanceolate points (LP-ComExh, n=2:Table 2). All complete projectile points appear to have been "worn out", as evidenced by incurvate, blunt tips, overlapping flake scar patterns, and thick lateral margins (cf. Daugherty et al. 1987:158-159; Flenniken and Wilke 1989:152-153). The massive potlid fractures observed on the dart point base fragment (DP-Prox) suggests that this artifact was fractured by fire. No evidence remained to indicate whether the artifact was broken prior to this damage.

The relative paucity of either complete or broken projectile points in this assemblage suggests rejuvenation and/or retooling associated with hunting was not a major focus of the lithic reduction activities at the Packwood Lake site. Experiments (Towner and Wharburton 1985; Flenniken 1985; Flenniken and Raymond 1986; Flenniken and Wilke 1989; Titmus and Woods 1986) have demonstrated that fracture of projectile points during use are mainly limited to basal areas. Most of these basal fragments from these experiments could not be rejuvenated into functional points because they were too small. According to Flenniken et al. (1989, 1990) hunting camps or kill sites contain these types of discarded

Figure 12. Bifacial artifacts from the Packwood Lake site.

projectile point elements. No basal fragments of projectile points were recovered from the Packwood Lake site.

Bifacial blanks (B-Frag=9, n=1:Table 2) are only represented by andesite/basalt raw materials. As discussed above, the reduction, especially of tabular raw material (Figure 13), was directed towards the production mainly of bifacial blanks (Figures 14 and 15) which apparently were removed from the site. These bifaces were probably manufactured for use as bifacial flake cores.

Only six preform fragments (P-Frag, n=5:Table 2) and one complete preform (P-CompDS, n=1:Table 2) were recovered. The complete preform exhibits evidence of resharpening, probably within a haft element, as shown by slightly incurvate margins that terminate near the base (Figure 16). Examination of the margins under 80x and 160x magnification showed some polish and minute flakes scars perpendicular to the margins indicative of edge damage from use in a sawing manner (Cleveland et al. 1976:30-48; Keeley 1980; Semenov 1964; Tringham et al. 1974). Four (P-FragDS, n=2; P-CompDS, n=1;
P-CompBT, n=1:Table 2) of these were manufactured on flakes, one of which could be identified as having been an early stage, bifacial thinning flake (BT-EPerc).

The final bifacially flaked stone artifact, a perforator/drill (Perf/Drill-ComExh, n=1:Table 2) of heat-treated jasper (Figure 12d), exhibits a significant amount of edge polish and has striations perpendicular to the margins

Figure 13. Tabular andesite/basalt biface blank fragment (B-Frag).

Figure 14. Andesite/basalt biface blank fragment (B-Frag).

Figure 15. Andesite/basalt biface blank. Note perverse fracture.

Figure 16. Preform (P-CompDS).

visible under 80x and 160x magnification. Numerous resharpening episodes resulted in a form that no longer could be rejuvenated for its intended use (Figure 12), therefore the artifact was discarded to become part of the archaeological assemblage as a complete, but exhausted tool.

Summary

The analytical assemblage from the Packwood Lake site reflects a reduction technology indicative of activities associated with the manufacture of bifacial blanks, preforms, expedient flake tools, and also the limited discarding of exhausted bifacial tools. Figures 17 and 18 model the entire lithic reduction technology (note: as presented above, not all of the illustrated activities occurred at the site). Lithic raw materials arrived at the site after most of the cortex already had been removed, indicating decortication had occurred elsewhere. While some toolstone was probably heat-treated at the site, the low frequency of stone not heat treated and debitage surfaces associated with preheat-treatment (i.e., dull luster on portions or entire dorsal flake surfaces as opposed to glossy ventral flake surfaces) indicate that the majority of the microcrystalline quartz toolstone was brought into the site after already having undergone heat treatment. As will be discussed in the following chapters, the assemblage does not appear to reflect

Text Box: Figure 17. Illustration of obsidian, chert/jasper, and chalcedony reduction technology.

Text Box: Figure 18. Illustration of andesite/basalt reduction technology.

lithic reduction activities or discarded artifacts indicative strictly of hunting or hunting related activities.

CHAPTER 4

PACKWOOD LAKE: COMPARISON WITH OTHER ASSEMBLAGES

Introduction

In this chapter I compare technological data from the Warehouse site, 35LA822, and the Diamond Lil site, 35LA807, with the Packwood Lake site, 45LE285. The implications for interpreting site function from technological studies are also discussed.

Analysis of the prehistoric lithic assemblages and interpretive results of the Diamond Lil, Packwood Lake, and Warehouse sites are primarily differentiated by the reduction trajectories, representative end products, and discarded artifacts. The Warehouse site is representative of a hunting camp where lithic reduction activities were predominantly associated with rejuvenation and manufacture of projectile points (Flenniken et al. 1989), whereas the Diamond Lil site represented a kill site whose lithic assemblage contained numerous projectile point fragments and evidence for lithic reduction activities associated with late stages of tool manufacture and projectile point rejuvenation (Flenniken et al. 1990). In contrast, the Packwood Lake site assemblage shows no evidence that it was primarily associated with activities related to hunting.

Because of the poor preservation of perishable artifacts in forested environments, most open-air archaeological sites in the western Cascade Mountains contain lithic artifacts, including formed artifacts and debitage, while other artifact classes are almost entirely absent. Because of this circumstance, as discussed in Chapter 1, many archaeologists have focused on the use of projectile point typologies, especially their application for the development of local and regional chronologies. Additionally, lithic assemblages have been described according to inferred tool functions and their relationship especially to activities related to hunting (Mack 1989). Technological considerations of stone tool manufacture is often conspicuously absent in the analyses of these sites. Sites characterized as lithic scatters have often been overlooked as sources of information regarding prehistoric lifeways because of the preconceived supposition that they manifested little information (Flenniken et al. 1989).

The Warehouse Site: 35LA822

Introduction. Test excavations by Flenniken et al. (1989) recovered over 3,000 lithic artifacts (formed artifacts and debitage) at the Warehouse site, 35LA822, in the western Oregon Cascade Mountains within the Willamette National Forest (Figure 19). While no specific chronological data pertaining to site occupation were obtained, interpretations of obsidian

Figure 19. Location of the Warehouse site, 35LA822, within the Willamette National Forest.

hydration data indicate that the site represents a single component in which the archaeological assemblage was deposited. Additional information from x-ray fluorescence spectrometry indicated a very homogeneous obsidian assemblage which supports the interpretation of a single component site (Flenniken et al. 1989:58-59).

Based upon the results of technological analysis of the lithic artifacts (debitage and formed artifacts), lithic reduction activities were stated to represent the latest stages of projectile point manufacture (secondary reduction subsystem), especially the rejuvenation (tertiary reduction subsystem) of projectile points (Figure 20), which were associated with a logistical hunting camp (Flenniken et al. 1989). A logistical hunting camp, or site, is defined as:

...a site associated with the manufacture (secondary reduction subsystem) and rejuvenation (tertiary reduction subsystem) of projectile points, the evidence which was left as the permanent archaeological record at that location. These sites are situated at middle elevations nearer the home bases or somewhat more permanent residential sites in the stream valleys. These locations are relatively close to permanent sites where large quantities of a meat resource may be procured at one time, processed, and easily transported back to the permanent home base. Logistical hunting site assemblages include lithic debitage associated with late secondary reduction, broken and/or exhausted projectile points, usually some evidence of animal processing tools, and in some instances, faunal remains. The major activities associated with a logistical hunting camp were, multiple kill episodes, initial butchery and processing, and refurbishment of hunting equipment. These activities may have

Figure 20. Schematic diagram of a lithic reduction technology analytically divided into subsystems (Flenniken et al. 1989).

occurred all at one location or in separate activity areas or separate sites [Flenniken et al. 1989:68-69].

Discussion. Cores requiring minimal platform preparation were reduced to manufacture flakes with effective cutting edges are suggested by interior, single-facet platform flakes (I-SFP). At the Warehouse site these flakes represent 1.3% (n=16) of the diagnostic debitage whereas they represent 10.3% (n=148) at Packwood Lake. This difference is attributed to the production of more flakes from unprepared cores (RM-Exh, n=4) at Packwood Lake than at the Warehouse site (RM-Exh, n=1).

An additional flake production method, via a bipolar technique which produces extremely flat flakes, is suggested by the recovery of two exhausted bipolar cores at the Packwood Lake site. Experimental replication and use of hafted bipolar flakes, or microliths, has demonstrated the utility of this technique for the production of a very efficient cutting instrument (Flenniken 1981).

Frequency of bifacial thinning debitage at both sites slightly differs for the earliest stages of reduction. Early stage bifacial thinning flakes (E-Perc) at Packwood Lake account for 26.0% (n=373) of the diagnostic debitage and 15.1% of the Warehouse site. Late stage bifacial thinning flakes accounts for only 3.21 (n=46) at the Packwood Lake site whereas the Warehouse site contains 9.2% (n=110) late stage bifacial thinning debitage. Even though flake blanks, which

tend to have an initial low length-to-thickness ratio, were being reduced at the Warehouse site, more final thinning by percussion was undertaken than at Packwood Lake.

The assemblage of bifacial thinning debitage at the Warehouse site is primarily associated with the reduction, or preparation, of flake blanks (Flenniken et al. 1989:41-45) which were then further reduced to manufacture projectile points (Flenniken et al. 1989:45). Biface percussion debitage at Packwood Lake represents biface preparation as well as reduction of previously prepared blanks which were then removed from the site. This determination could not have been made without information pertaining to the entire range of reduction stages represented at each site as well as the detailing of the reduction trajectories.

The characteristics of the pressure flakes from the Warehouse site (n=771; 64.5% of the total diagnostic debitage assemblage), principally early stage pressure flakes (BT-EPres, n=224; 18.6%), were considered to be indicative of rejuvenation activities. Rejuvenation of squared or broken edges was evidenced by alternate pressure flakes (cf. Towner and Wharburton 1985). Late stage pressure flakes (BT-Lpres, n=547; 45.6%) appear to be related to the restoration of the width-to-thickness ratio necessary for hafting and use (Flenniken et al. 1989:47). The frequencies of early and late stage pressure flakes vary considerably from the pressure flaking debitage of the Packwood Lake site; no obvious

rejuvenation pressure flakes were observed (i.e., alternate pressure flakes) and early stage pressure flakes (BT-EPres, n=565; 39.4%) were much more prevalent than late stage pressure flakes (BT-LPres, n=120; 8.4%). The primary difference in the pressure flake assemblage of both sites reflects differences in the resultant end product. Pressure flaking at the Warehouse site was primarily associated with projectile point production and rejuvenation whereas preform manufacture was the principle activity at the Packwood Lake site.

Lithic reduction activities at the Warehouse site appear to have been much more focused on maintenance aspects associated with hunting equipment than at the Packwood Lake site. Data to which the investigators of the Warehouse site refer in support of this conclusion are especially debitage frequencies and their association with the rejuvenation and manufacture of projectile points, and to a lesser extent, the attributes of exhausted and discarded points (Flenniken et al. 1989). The most fundamental difference between the assemblages from the Warehouse and Packwood Lake sites is the realization that the reduction trajectories employed at each site differ not only in the desired end products, but especially in the behaviors associated with their ultimate use.

Diamond Lil: 35LA807

Introduction. The Diamond Lil site, 35LA807, is located in the on the west side of the Oregon Cascade Mountain Range in the Middle Fork of the Willamette River drainage within the Willamette National Forest (Figure 21). Data recovery excavations conducted at the Diamond Lil site resulted in the recovery of over 20,000 lithic artifacts, which included both flaked and ground stone tools, along with a faunal assemblage consisting of 800 bone fragments. Analyses included that of faunal remains, site structure, site context, obsidian x-ray fluorescence spectrometry, obsidian hydration, all of the formed artifacts, and a 65% sample of the debitage. Interpretation of the site data as well as ethnographic literature (cf. Anell 1969; Steward 1943, 1970; Ray 1942; Barnett 1937; Voegelin 1942) and archaeological literature (cf. Frison 1974, 1978; Wilke 1986) concerning multiple kill events, led the researchers to conclude that the Diamond Lil site operated as a kill, butchering, and meat processing locality (Flenniken et al. 1990). Site assemblages associated with intensive hunting may exhibit the following diagnostic, though not exhaustive, list of features:

1) impact fractured projectile points and fragments indicating the kill location; 2) artifacts diagnostic of weapons maintenance and rejuvenation such as pressure, alternate, edge preparation, and notching flakes; 3) tools and debitage evidencing manufacture and maintenance

Figure 21. Location of the Diamond Lil site, 35LA807, within the Willamette National Forest.

of butchering tools (i.e., flake core, bipolar, and/or microblade technologies); 4) ground stone or heavy industry tools and fire-cracked rock related to meat or bone processing. Traces of animal remains should also be present, although preservation of organic materials is dependent on micro-environmental conditions at the site and the age of the deposit [Flenniken et al 1990:143-144].

The frequent occurrence (n=95) of projectile points and point fragments that exhibited fractures associated with impact (cf. Flenniken 1985; Flenniken and Raymond 1986; Titmus and Woods 1986) suggested that hunting was the primary activity. The researchers argued that the occurrence of projectile point fragments, especially non-basal elements, in such frequency is not the exclusive result of being returned to the site within the meat, but rather the consequence of breakage on-site during use, specifically during the actual kill of game at the Diamond Lil site (Flenniken et al. 1990:156).

Discussion. Microblades and bipolar flakes were interpreted as the cutting tools used during the processing of carcasses. No other cutting implements were identified in the assemblage. Even microscopic inspection of lanceolate-shaped projectile point margins failed to reveal evidence of use as a cutting tool (Flenniken et al. 1990:159). The inference that flakes also were being produced for use as unmodified and expedient cutting edges is suggested by the recovery of eleven flake cores (pieces of lithic material from which these flakes were detached) from the site (Flenniken et al. 1990:65-68).

The use of a microblade technology at the Diamond Lil site was probably an important means for the production of straight cutting edges for processing meat. Complete microblades and fragments of microblades (n=184) along with microblade cores (n=4) recovered from the site (Flenniken et al. 1990:186-187) suggest a technique often employed to fully utilize small pieces of high quality toolstone (cf. Daugherty et al. 1987a, 1987b:109). No indication of the use of a microblade technology at the Packwood Lake site was encountered.

Though a microblade technology was in evidence as early as 7,000 B.P. within the upper Cowlitz River drainage at sites such as Layser Cave (Daugherty et al. 1987a) and continued until as recently as 200 B.P. at Judd Peak (Daugherty et al. 1987b:226), microblades appear not to have been used at Packwood Lake. Their utilization at Layser Cave and Judd Peak may be related to the types of meat processing activities that occurred at these sites. Inferred activities at the Diamond Lil site are comparable to those at both Layser Cave (Daugherty et al. 1987a) and Judd Peak (Daugherty et al. 1987b); these sites were associated with processing large quantities of game. Judd Peak also contained a significant quantity of salmonid bones, though it also should be noted that the use of the microblade technology at Judd Peak was not limited to levels associated these remains (Daugherty et al. 1987b:102). Apparently, the lithic assemblage at Packwood

Lake does not reflect the processing en masse of large quantities of meat resources such as that obtained from multiple kill events.

Principal lithic reduction activities at the Diamond Lil site were associated with the manufacture and especially the rejuvenation of bifacial tools (Flenniken et al. 1990). Early stage bifacial thinning flakes (BT-Eperc, n=558; 11.3% of total diagnostic assemblage) and late stage bifacial thinning flakes (BT-Lperc, n=130; 2.6%) are well represented in the Diamond Lil site lithic assemblage (Flenniken et al. 1990:82). These data percentages are comparable to bifacial thinning flakes from the diagnostic debitage of early stage bifacial thinning (BT-Eperc, n=373; 26.0%) and late stage bifacial thinning debitage (BT-LPerc, n=46; 3.2%) at Packwood Lake. The significant difference in the debitage sub-assemblages between the two sites is not in the actual frequencies of those categories, rather it is the dissimilarity in the associated end product resultant from the reduction trajectories. At the Diamond Lil site, biface thinning debitage is primarily associated with the reduction of flake blanks and their subsequent reduction for use in the production of projectile points, whereas the Packwood Lake biface reduction predominately indicates the production of biface blanks and preforms.

Debitage characteristic of reduction by pressure at the Diamond Lil site is represented by early stage pressure (BT-

Epres, n=1973; 40.0%) and late stage pressure flakes (BT-Lpres, n=1716; 34.8%). Replication experiments were conducted for the obsidian assemblage of the Diamond Lil site, and the analytical results were used to identify the specific on-site reduction stages represented by the debitage assemblage. Results of these experiments indicated that the Diamond Lil site assemblage exhibited characteristics of "preform-to-finished point, point-though-rejuvenation" reduction stages (Flenniken et al. 1990:83-96).

The Packwood Lake assemblage of pressure flakes is represented by both early stage pressure flakes (BT-EPres, n=565; 39.4%) and late stage pressure flakes (BT-Lpres, n=120; 8.4%). As presented in Chapter 3, characteristics of the pressure flaking debitage primarily reflect the manufacture of preforms rather than rejuvenation of artifacts, such as projectile points that have been broken during use. Characteristics of the pressure flakes in the debitage assemblage do not appear to be similar to those of the Diamond Lil site (e.g., no rejuvenation pressure flakes were observed among the pressure flakes from the Packwood Lake site).

One of the most obvious differences between the two assemblages is the frequency and characteristics of the projectile points from each site. A total of ninety-five projectile points and fragments were recovered from Diamond Lil compared to six at Packwood Lake. The numerical difference is not nearly as meaningful as the comparison in

the frequency and type of projectile point damage between the two sites. The Diamond Lil site projectile point assemblage exhibited numerous examples of impact damage and impact fractures. All of the projectile points from Packwood Lake are representative of exhausted and discarded projectile points, none of which exhibited characteristics of impact damage.

Summary

Differences in site assemblages reflected by the types of lithic reduction activities that occurred on-site can be used to differentiate site function. However, data alone (i.e., comparisons of specific debitage category frequencies) cannot be used to infer site function, but interpretations of these data which describe the whole range of reduction activities, reduction trajectories, resultant end products, and discarded artifacts may allow for inferences of some of the behavior which in part, resulted in the formation of the archaeological record. It is not possible from the technological analysis of a lithic assemblage to identify all activities that may have occurred at a site.

CHAPTER 5

CONCLUSIONS

The Packwood Lake site (45LE285) prehistoric lithic assemblage primarily represents utilization of the lakeshore area approximately between 500 and 1,100 years ago. The technological analysis determined that flintknapping activities were concentrated on the production of bifacial blanks, preforms, and expedient flake tools. These data do not suggest a site associated with hunting. The ethnographic record for the site vicinity and the absence of a hunting site assemblage suggests that other activities occurred there. These may have included the procurement of non-hunting related resources, such as fish and huckleberries, as suggested by the ethnographic record. Therefore, the Packwood Lake lithic assemblage may be indicative or characteristic of non-hunting sites or multi-resource acquisition sites within upper elevation areas of the Cascade Mountain Range.

The comparison of the Warehouse site lithic assemblage, which represented a hunting camp, and the Packwood Lake site lithic assemblage indicated differences in reduction activities. At the Warehouse site primary flintknapping activities were oriented towards maintenance aspects of the hunting toolkit. This was evidenced by the occurrence of rejuvenation debitage and the manufacture of projectile points

as well as the attributes of the exhausted and discarded projectile points.

The Packwood Lake site reduction activities do not indicate refurbishment or maintenance of the hunting toolkit. Instead flintknapping activities at Packwood Lake are the continuation of the reduction continuum of the entire toolkit. That is, some artifact classes, such as bifacial blanks and preforms, were manufactured at the site with the intent that they be further modified or reduced elsewhere on the landscape when replacement tools were required. Other lithic materials were reduced at the site to provide expedient tools.

A comparison of the Diamond Lil site lithic assemblage, which is characterized as representative of a deer kill site, and the Packwood Lake site also emphasizes the differences between hunting and non-hunting site assemblages. The Diamond Lil site contained abundant evidence for on-site use of projectile points as evidenced by numerous impact fractured points and point fragments. The Packwood Lake site contained no such evidence. The Diamond Lil site reduction activities were primarily associated with the rejuvenation of projectile points as well as the manufacture of replacements. Indications of this activity were virtually absent from Packwood Lake. Finally, Diamond Lil contained evidence for the use of a microblade technology for processing meat obtained from on-site multiple deer kills. There is no indication from the analysis of the Packwood Lake site lithic

assemblage that the site represents intensive hunting or processing activities.

The Packwood Lake site appears to have been a multi-resource acquisition site. It is likely that seasonal use of the lake area began with the onset of trout spawning sometime during late spring to early summer. Inhabitants of villages within the Upper Cowlitz River valley travelled to Packwood Lake to begin procurement and processing fish at this time. As the season progressed, additional resources such as huckleberries, became available in the surrounding hillsides. These were probably gathered and returned to the site for processing. Later when access to the upper elevations of the Goat Rocks area was possible, additional forays were made to procure mountain goats as well as other game. Meat processing probably occurred at or near kill sites within the Goat Rocks at which time toolkit maintenance also occurred. Obsidian was obtained while in the Goat Rocks area and probably represents an embedded strategy (Binford 1979). The procurement of resources in the vicinity of Packwood Lake probably ended at about the time the fall fish runs began in the upper Cowlitz River, at which time the prehistoric peoples returned to the villages within the Cowlitz River valley.

Other conclusions which can be elicited from results of the analysis of the Packwood Lake site and subsequent comparison to lithic assemblages of functionally dissimilar sites have implications for the interpretations of site

function utilizing a technologically oriented approach. A prerequisite for this approach has been the comparison of lithic assemblages from sites whose functions have been inferred from the same or similar analytical criteria and methodological approach derived from interpretations of experimental data.

Subsistence activities, especially those which are primarily concentrated or focused on the procurement and/or processing of faunal resources at a specific site, are especially apt to be reflected by the distinctive nature of the lithic assemblage such as those from the Warehouse and Diamond Lil sites. Hunting related site assemblages are represented by well defined lithic reduction strategies principally involving late stages of tool manufacture, and particularly rejuvenation of tools damaged either during the most recent hunting episodes, or during an interlude in preparation for those to follow.

Sites such as the Packwood Lake site that exhibit lithic artifact assemblages more generic in character apparently do not reflect the manufacture, use, or repair of distinctive stone tool categories for effective completion of various tasks; that is, interpretation of the data does not portray an emphasis on a well-delineated set of activities that likely occurred at this site. Therefore, sites should not be ascribed a specific function without consideration of all plausible aspects of human behavior likely to have led to the

nature of the archaeological record. This should also include an assessment of natural processes which may have affected the character of these remains.

The research presented in this thesis demonstrates the utility of an approach that uses data, and interpretations thereof, to describe technological aspects of logistical resource procurement organization (cf. Binford 1980). While no presumption is made that all subsistence activities can be identified using this approach, it nevertheless establishes a framework within which the researcher must work toward an examination of all aspects of the site and its context in order to contemplate a more likely hypothesis regarding site function.

No single assemblage attribute, such as technologically diagnostic lithic artifacts, is indicative of the limits of site activities, or range of diversity therein, for interpretations of potential activities that occurred prehistorically at a site. The whole of the assemblage--and not just its constituent parts--represents a portion of a behavioral system which enables the elucidation of a more precise definition of site function.

REFERENCES CITED

Anell, B.

1969 Running Down and Driving of Game in North America. Studia Ethnographica Upsaliensia XXX. Berlingska Boktrycheriet. Lund, Sweden.

Anonymous

1909 Story of Packwood Lake Interesting One to Anglers. Bee-Nuggett, 12/10/09, p.10. Chehalis, Washington.

Bamforth, Douglas B.

1986 Technological Efficiency and Tool Curation. American Antiquity 51(1):38-51.

Barnett, H. G.

1937 Culture Element Distributions: VII Oregon Coast. Anthropological Records 1(3). University of California Press, Berkeley.

Bergland, Eric O.

1988 Archaeological Test Excavations at the Berkeley Rockshelter Site (45-PI-303), Mt. Rainier National Park, Washington. SRD Report of Investigations No. 1. Eugene, Oregon.

Binford, Lewis R.

1973 Interassemblage Variability-The Mousterian and the "Functional" Argument. In The Explanation of Culture Change, edited by Colin Renfrew, pp. 227-254. Duckworth, London.

1977 Forty-seven Trips. In Stone Tools as Cultural Markers, edited by R. V. S. Wright, pp. 24-36. Australian Institute of Aboriginal Studies, Canberra.

1979 Organization and Formation Processes: Looking at Curated Technologies. Journal of Anthropological Research 35:255-273.

Binford, L. R., and G. I. Quimby

1972 Indian Sites and Chipped Stone Materials in the Northern Lake Michigan Area. In An Archaeological Perspective, edited by L. R. Binford, pp. 346-372. Seminar Press, New York.

Bordes, F., and D. Crabtree

1969 The Corbiac Blade Technique and Other Experiments. Tebiwa 12:1-21

Caldwell, J. R.

1958 Trend and Tradition in the Prehistory of the Eastern United States. American Anthropologist, Memoir No. 88.

Cleveland, G. C., J. Jeffrey Flenniken, D. R. Huelsbeck, R. Mierendorf, S. Samuals, and F. Hassan

1976 Preliminary Archaeological Investigations at the Miller Site, Strawberry Island, 1976, A Late Prehistoric Village Near Burbank, Franklin County, Washington. Washington Archaeological Research Center Project Report No. 46, Pullman.

Crabtree, Don E.

1966 A Stone-Worker's Approach to Analyzing and Replicating the Lindenmeier Folsom. Tebiwa 9:3-39.

1967a Notes on Experiments in Flintknapping: 3, The Flintknappers Raw Materials. Tebiwa 10:8-24.

1967b Notes on Experiments in Flintknapping: 4, Tools Used for Making Flaked Stone Artifacts. Tebiwa 10:60-73.

1968 Mesoamerican Polyhedral Cores and Prismatic Blades. American Antiquity 33:446-478.

1972a The Cone Fracture Principle and the Manufacture of Lithic Materials. Tebiwa 15:29-42.

1972b An Introduction to Flintworking. Occasional Papers of the Idaho State University Museum No. 28, Pocatello.

Crabtree, D. E. and B. R. Butler

1964 Notes on Experiments in Flint Knapping: 1 Heat Treatment of Silica Materials. Tebiwa 7(1):1-6.

Collins, Michael B.

1975 Lithic Technology as a means of Processual Inference. In Lithic Technology: Making and Using Stone Tools, edited by E. Swanson, pp. 15-34. Mouton Publishers, The Hague.

Cole, D. L. and F. Leonhardy

1964 Interim Reports for 1963-1964, John Day Dam Reservoir Project, Part II: Report on Survey and Excavation on Blalock Island. Ms. on file, Museum of Natural History, University of Oregon, Eugene.

Crandell, D. R.

1969 Surficial Geology of Mount Rainier National Park. Geological Survey Bulletin 1288. U. S. Government Printing Office, Washington, D. C.

Curtis, E. S.

1913 The North American Indian, vol. 9. Cambridge, Massachusetts.

Dancey, William S.

1968 Archaeology of Mossyrock Reservoir, Washington. Unpublished Master's thesis, Department of Anthropology, University of Washington, Seattle.

Daugherty, Richard D., J. Jeffrey Flenniken, and Jeane M. Welch

1987a A Data Recovery Study of Layser Cave (45-LE-223) in Lewis County, Washington. Studies in Cultural Resource Management No. 7. USDA Forest Service, Pacific Northwest Region, Portland.

1987b A Data Recovery Study of Judd Peak Rockshelters (45-LE-222) in Lewis County, Washington. Studies in Cultural Resource Management No. 8. USDA Forest Service, Pacific Northwest Region, Portland.

Douglas, William O.

1950 Of Men and Mountains. Harper and Brothers, Inc. New York.

Draper, J. A. and J. J. Flenniken

1984 The Use of the Scanning Electron Microscope for the Detection of Heat Treated Lithic Artifacts. Northwest Anthropological Research Notes 18(1):117-123.

Ellingson, J. A.

1968 Late Cenozoic Volcanic Geology of the White Pass-Goat Rocks Area, Cascade Mountains, Washington. Unpublished Ph.D. dissertation, Washington State University, Pullman.

1972 The Rocks and Structure of the White Pass Area, Washington. Northwest Science 46:9-24.

Ellis, H. Holmes

1965 Flint-Working Techniques of the American Indians: An Experimental Study. Ohio Historical Society, Columbus.

Ertec Northwest, Inc.

1981 Test Excavations at the Cowlitz Falls Archaeological Sites. Submitted to R. W. Beck and Associates, Seattle.

Flenniken, J. Jeffrey

1979 Comments on Heat Treatment. Flintknappers' Exchange 26(1):8-9.

1981 Replicative Systems Analysis: A Model Applied to the Vein Quartz Artifacts from the Hoko River Site. Washington State University Laboratory of Anthropology Reports of Investigations No. 59, Pullman.

1984 The Past, Present, and Future of Flintknapping: An Anthropological Perspective. Annual Review of Anthropology 13:187-203.

1985 Reduction Techniques as Cultural Markers. In Stone Tool Analysis: Essays in Honor of Don E. Crabtree, edited by M. G. Plew, J. C. Woods, and M. G. Pavesic, pp. 265-276. University of New Mexico Press, Albuquerque.

1987 The Lithic Technology of the East Lake Site, Newberry Crater, Oregon. Lithic Analysts Research Report No. 1, Pullman.

Flenniken, J. J. and E. G. Garrison

1975 Thermally Altered Novaculite and Stone Tool Manufacturing Techniques. Journal of Field Archaeology 2:125-131.

Flenniken, J. J. and J. C. Haggarty

1979 Trampling as an Agency in the Formation of Edge Damage. Northwest Anthropological Research Notes 13:208-214.

Flenniken, J. J. and T. L. Ozbun

1988 Archaeological Investigations in Newberry Crater, Deschutes National Forest, Central Oregon. Lithic Analysts Research Report No. 4. Pullman.

Flenniken, J. J., T. L. Ozbun, and A. C. Fulkerson

1989 Archaeological Test Excavations at the Warehouse Site, 35LA822, Blue River Ranger District, Willamette National Forest, Oregon. Lithic Analysts Research Report No. 10. Pullman.

Flenniken, J. Jeffrey, Terry L. Ozbun, A. C. Fulkerson, and Carol Winkler

1990 The Diamond Lil Deer Kill Site: A Data Recovery Project in the Western Oregon Cascade Mountains. Lithic Analysts Research Report No. 11. Pullman.

Flenniken, J. J. and A. W. Raymond

1986 Morphological Projectile Point Typology: Replicative Experimentation and Technological Analysis. American Antiquity 51:603-614.

Flenniken, J. J. and A. L. Stanfill

1980 A Preliminary Technological Examination of 20 Archaeological Sites Located During the Cultural Resource Survey of the Whitehouse Ranch Public Land Exchange. Contract Abstracts 1:23-30.

Flenniken, J. J. and J. P. White

1983 Heat Treatment of Siliceous Rocks and Its Implications for Australian Prehistory. Australian Aboriginal Studies 1:43-48.

Flenniken, J. J. and P. J. Wilke

1989 Typology, Chronology, and Technology of Great Basin Dart Points: An Anthropological Perspective. American Anthropologist 19:149-158.

Franklin J. F. and C. T. Dyrness

1988 Natural Vegetation of Washington and Oregon. Reprinted. Oregon State University Press, Corvallis. Originally Published 1973, USDA Forest Service, Pacific Northwest Forest and Range Experiment Station, Portland.

Frison, G. C.

1974 The Casper Site: A Hell Gap Bison Kill on the High Plains. Academic Press, New York.

1978 Prehistoric Hunters of the High Plains. Academic Press, New York.

Fugro Northwest, Inc.

1979 Archaeological Consulting Services for the Cowlitz Falls Hydroelectric Project. Submitted to R. W. Beck and Associates, Seattle, W. W.-1115-HG1-CA3007.

1980 Cultural Resources Assessment for the Cowlitz Falls Hydroelectric Project. Submitted to R. W. Beck and Associates, Seattle.

Gibbs, George

1855 Report of Mr. George Gibbs to Captain McClellan on the Indian Tribes of the Territory of Washington. In Reports of Exploration and Surveys, to Ascertain the Most Practicable and Economical Route for a Railroad from the Mississippi River to the Pacific Ocean, 1:402-434, Executive Document No. 78, 2nd Session, 33rd Congress, 1853-1854.

Gilreath, A. J.

1983 Bifacial Debitage and Sampling at a Small Lithic Scatter: An Experimental Study. Unpublished Master's thesis, Department of Anthropology, Washington State University, Pullman.

Goldschmidt, W.

1951 Nomloki Ethnography. Publications in American Archaeology and Ethnology 42. University of California, Berkeley.

Grinnell, G. C.

1898 The Story of the Indian. Appleton, New York.

Harrison, C. A. and W. O. Powell

1977 Packwood Lake Geology and Shoal Stability. Ms. on file, Materials Engineering and Geologic Services, Gifford Pinchot National Forest.

Holmes, W. H.

1892 Modern Quarry Refuse and the Palaeolithic Theory. Science 20:295-297.

1893 Distribution of Stone Implements in the Tidewater Country. American Anthropologist 6:1-14

1897 Stone Implements of the Potomac-Chesapeake Tidewater Province. Bulletin of American Ethnology Annual Report 15:13-

1919 Handbook of Aboriginal American Antiquities, Part I: Introductory to the Lithic Industries. Bureau of American Ethnology Bulletin 60.

Honcharsky, Susan

n.d. Packwood Lake Scenic Area, Draft Management Plan. Ms. on file, Packwood Ranger District, Gifford Pinchot National Forest. Packwood, Washington.

Irwin-Williams C. and H. T. Irwin

1966 Excavation at Magic Mountain. Proceedings of the Denver Museum of Natural History 12.

Jacobs, M. M.

1934 Northwest Sahaptin Texts, Part 1. Columbia University Contributions to Anthropology 19. Columbia University Press, New York.

1937 Historic Perspectives in Indian Languages of Oregon and Washington. Pacific Northwest Quarterly 28:55-74.

Jermann, J. V. and R. D. Mason

1976 A Cultural Resource Overview of the Gifford Pinchot National Forest, South-Central Washington. Office of Public Archaeology, Institute for Environmental Studies, University of Washington, Seattle.

Jermann, J. V., J. R. Benson, and D. E. Lewarch

1988 Archaeological Data Recovery at the McClellan Meadows Site (45-SA-116). Ms. on file, USDA Forest Service, Gifford Pinchot National Forest, Vancouver, Washington.

Johnson, L. L.

1978 A History of Flintknapping Experimentation, 1838-1976. Current Anthropology 9:337-372.

Johnson, G. M.

1985 The Use of the Scanning Electron Microscope in Studying the Heat Treatment of Prehistoric Lithic Artifacts from the North Florida Weeden Island Period McKeithen Site. Scanning Electron Microscopy 2:651-658.

Jones, Shawn

1987 Identification of Lithic Samples. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle, Washington.

Kalin, J. K.

1981 Stem Point Manufacture and Debitage Recovery. Archaeology of Eastern North America 9:134-175.

Keeley, L. H.

1980 Experimental Determination of Stone Tool Uses: A Microwear Analysis. Prehistoric Archaeology and Ecology Series, K. W. Butzer and L. G. Freeman, editors. University of Chicago Press.

Kehoe, John

1890 Communicated (Letter to the Editor postmaked Vance, Washington, September 8, 1890). Bee-Nugget, October 3, 1890. Chelalis, Washington.

Kiona, Mary

1964 Taped interview in Sahaptin with English Translation, conducted 9/31/64, by Martha Hardy, Packwood, Washington. Translated by Minnie Placid. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle, Washington.

1965 Taped interview in Sahaptin with English translation, conducted 8/17/65, by Martha Hardy, Packwood, Washington. Translated by Joyce Eyley. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle, Washington.

Klein, C. and C. S. Hurlburt

1985 Manual of Mineralogy (after James D. Dana). 20th ed. James Wiley and Sons, New York.

Knowles, F. H. S.

1953 Stone-Workers's Progress: A Study of Stone Implements in the Pitt-Rivers Museum. Occasional Papers on Technology 6. Pitt-Rivers Museum, University of Oxford

Landis, and Lothson

1982 Phase II: Intensive Survey, Test Excavation and Evaluation of Fallen Arches Cave Site, 45SA41, Gifford Pinchot National Forest, Skamania County. Eastern Washington University Reports in Archaeology and History 100-27. Cheney, Washington.

Lewarch, D. E. and J. R. Benson

1989 Paradise Hills Saddle Site T6N R7E-30/01, Archaeological Testing, Drought Timber Sale Road Construction. Ms. on file, Gifford Pinchot National Forest, Vancouver, Washington.

Liddle, Janet A. and Jeffrey A. Markos

1987 The Ware Divide Archaeological Site, 45-LE-288: Significance Evaluation. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle, Washington.

Lucas, Bob

1989 Southwestern Washington High Lakes Surveys. Fishery Management Report 89-1. Washington Department of Wildlife.

Mack, C. A.

1989 Prehistoric Upland Occupations in the Southern Washington Cascades. Archaeology in Washington 1:49-58.

Mandeville, M. and J. J. Flenniken

1974 A Comparison of the Flaking Qualities of Newhaka Chert Before and after Thermal Pretreatment. Plains Anthropologist 19:146-148.

Marden, Guy

1988 Archaeological Investigations at the Monument Headquarters Site (45CL319), Clark County, Washington. Ms. on file, Mount St. Helens National Volcanic Monument Headquarters, Gifford Pinchot National Forest.

Marshall, Lorna

1976 The !Kung of Nyae Nyae. Harvard University Press, Cambridge.

Mathis, Mark A.

1976 A Research Design for the Investigation of Prehistoric Lithic Utilization in the Western Lowlands of Northeast Arkansas. Ms. on file, Department of Anthropology, University of Arkansas, Fayetteville.

McClure, R. H., Jr.

1985 Archaeological Testing and Significance Evaluation, Site 45LE251, Cispus River, Washington. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle, Washington.

1987 Significance Evaluation of Archaeological Site 45-LE-285 Packwood Lake, Lewis County, Washington. Ms. on file, Packwood Ranger District, Gifford Pinchot National Forest. Packwood, Washington.

1989 Alpine Procurement in the Southern Washington Cascades: Preliminary Research. Archaeology in Washington 1:59-70.

McClure, R. H., Jr. and J. A. Markos

1990 Test Excavations at Squatter's Knob Rockshelter (45LE295), Lewis County, Washington. Ms. on file, Randle Ranger District, Gifford Pinchot National Forest. Randle Washington.

Miss, C. J.

1989 Evaluation of the Burton Creek Rockshelter, Site 45-LE-266, Lewis County, Washington. Ms. on file, Packwood Ranger District, Gifford Pinchot National Forest. Packwood, Washington.

Mullineaux, D. R. and D. R. Crandell

1981 The Eruptive History of Mount St. Helens. In The 1980 Eruptions of Mount St, Helens, Washington, edited by P. W. Lipman and D. R. Mullinaeux, pp 3-15. Geological Survey Professional Paper 1250. U. S. Government Printing Office, Washington, D. C.

Mullinaeux, D. R., J. H. Hyde, and M. Rubin

1975 Widespread Late Glacial and Postglacial Tephra Deposits from Mount St. Helens Volcano, Washington. Journal Research
U. S. Geological Survey 3:329-335.

Muto, G. R.

1971 A Stage Analysis of the Manufacture of Stone Tools. In Great Basin Anthropological Conference 1970, Selected Papers, edited by C. Melvin Aikens, pp. 109-118. University of Oregon Anthropological Papers No. 1.

Ozbun, Terry Lee

1987 A Technological Analysis of the Lithic Assemblage from the Buttonhole Rockshelter/Quarry Site, Northeastern New Mexico. Unpublished Master's thesis, Department of Anthropology, Washington State University, Pullman.

PCSP (Packwood Community Study Proceedings)

1954 Packwood on the March, Part 4, History Committee Report. University of Washington Community Development Bureau, Seattle. Ms. on file Packwood Ranger District, Gifford Pinchot National Forest. Packwood, Washington.

Pond, A. W.

1930 Primitive Methods of Working Stone, Based on the Experiments of Halvor L. Skavlem. Logan Museum, Beloit College. Beloit.

Powell, J. W.

1875 Exploration of the Colorado River of the West and its Tributaries in 1869-1871 under Direction of the Secretary of the Smithsonian Institute. U. S. Government Printing Office, Washington, D. C.

Purdy, B. A.

1974 Investigations Concerning the Thermal Alteration of Silica Minerals: An Archaeological Approach. Tebiwa 17:37-66.

Ray, V. F.

1942 Culture Element Distribution: XXII Plateau. Anthropological Records 8, No. 2. University of California Press, Berkeley.

1974 Handbook of the Cowlitz Indians. In Coast Salish and Western Washington Indians vol. III, edited by D. Horr, pp. 245-315. Garland Publishing Company, New York.

Raymond, A. W.

1986 Prehistoric Reduction and Curation of Topaz Mountain, Utah, Obsidian: A Technological Analysis of Two Lithic Scatters. Unpublished Master's thesis, Department of Anthropology, Washington State University, Pullman.

Rice, D. G.

1964a Indian Utilization of the Cascade Mountain Range in South Central Washington. The Washington Archaeologist 8(1):5-20.

1964b Test Excavation at Wild Rose Rockshelter, A Site on the Eastern Slopes of the Cascades. The Washington Archaeologist 8(4):2-23.

1965 Archaeological Test Excavations in Fryingpan Rockshelter, Mount Rainier National Park. Washington State University, Laboratory of Anthropology, Report of Investigation 33. Pullman.

1969 Archaeological Reconnaissance, South-Central Cascades. Washington Archaeological Society, Occasional Paper No. 2.

Rick, J. W.

1978 Heat-Altered Cherts of the Lower Illinois Valley: An Experimental Study in Prehistoric Technology. Northwestern University Archaeological Program Prehistoric Records 2. Evanston.

Ruby, R. H. and J. A. Brown

1985 A Guide to the Indian Tribes of the Pacific Northwest. University of Oklahoma Press, Norman.

Semenov, S. A.

1964 Prehistoric Technology. Translated by M. W. Thompson. Barnes and Noble, New York.

Sethe, William

1938 Memorandum for Forest Supervisor, Columbia National Forest, dated December 15, 1938, Historical Information Files, Columbia National Forest. Old Timers Contacted and Old Diaries Examined, p. 3. Ms. on file, Gifford Pinchot National Forest. Vancouver, Washington.

Schamberger, M. L.

n.d. Mount Rainier's Mammals. The Anundsen Publishing Company, Decorah, Iowa. Copies available from Mount Rainier Natural History Association, Longmire, Washington.

1970 Mammals of Mount Rainier National Park. Unpublished Ph.D. dissertation, Oregon State University, Corvallis.

Scott, S. A., C. M. Davis, J. J. Flenniken

1986 The Pahoehoe Site: A Lanceolate Biface Cache in Central Oregon. Journal of California and Great Basin Anthropology 8(1):7-23.

Schuster, Robert L.

1990 Letter to Bill Johnson, Packwood Hydroelectric Project, Packwood Washington, dated February 27, 1990 from United States Department of the Interior, Geological Survey, Denver, Colorado. Ms. on file Packwood Ranger District, Gifford Pinchot National Forest. Packwood, Washington.

Scott, S. A.

1985 An Analysis of Archaeological Materials Recovered During Test Excavations of Six Prehistoric Sites on the Deschutes National Forest in Central Oregon. Cultural Resource Report No. 2. USDA Forest Service, Pacific Northwest Region, Deschutes National Forest.

Shippee, J. M.

1963 Was Flint Annealed Before Flaking? Plains Anthropologist 8(22):271-272.

Simmons, G. C.

1950 The Russell Ranch Formation. Unpublished Master's thesis, Washington State College, Pullman.

Skinner, C.

1983 Obsidian Studies in Oregon: An Introduction to Obsidian and An Investigation of Selected Methods of Obsidian Characterization Utilizing Obsidian Collected at Prehistoric Quarry Sites in Oregon. Unpublished Master's thesis, Department of Interdisciplinary Studies, University of Oregon, Eugene.

Smith, Marion W.

1940 The Puyallup-Nisqually. Columbia University Contributions to Anthropology, vol. 32. Columbia University Press, New York.

Smith, Allan H.

1953 The Indians of Washington. Research Studies of the State College of Washington 21(2):85-113.

1964 Ethnographic Guide to the Archaeology of Mt. Rainier National Park. Washington State University. Submitted to the National Park Service. Contract No. 14-10-0434-1422.

Spencer, Lee

1987 Testing and Evaluation at 45-SA-117, The Falls Creek Site: A Culturally Enriched Deposit in the Interior Upland Region of the Columbia Gorge. Lee Spencer Archaeology Paper 3.

Spier, L.

1936 Tribal Distributions in Washington. General Series in Anthropology No. 3. George Banta Publishing Company, Menasha.

Stanfill, Alan L.

1976 Stages of Cobble Reduction and the Base Settlement/Specialized Activity Site Dichotomy: A Test Case with the 1976 Village Creek Collections. Ms. on file, Department of Anthropology, University of Arkansas, Fayetteville.

Steward, J. H.

1943 Culture Element Distributions: XXIII Northern and Gosuite Shoshoni. Anthropological Records 8(3). University of California Press, Berkeley.

1970 Basin-Plateau Aboriginal Sociopolitical Groups. Reprinted. University of Utah Press, Salt Lake City. Originally published 1938, Bureau of American Ethnology Bulletin 120, Smithsonian Institution.

Stewart, O. C.

1941 Culture Element Distributions: XIV Northern Paiute. Anthropological Records 6. University of California Press, Berkeley.

1942 Culture Element Distributions: XVIII Ute-Southern Paiute. Anthropological Records 6. University of California Press, Berkeley.

Stilson, M. L. and G. Thompson

1988 Cowlitz Falls Project, Report on Archeological Testing, Terrace Area (45LE209W). Dames and Moore. Submitted to Bechtel Civil, Inc., Contract No. 15846-001-020.

Streuver, S.

1970 Pioneer Archaeology in an Illinois Middle Woodland Mound. Wisconsin Archaeologist 51(2).

Swanson, D. A. and G. A. Clayton

1983 Generalized Geologic Map of the Goat Rocks Wilderness and Roadless Areas (6036, Parts A, C, and D), Lewis and Yakima Counties, Washington. United States Department of the Interior, Geological Survey Open-File Report 83-357.

Taylor, W. and W. T. Shaw

1927 Mammals and Birds of Mount Rainier National Park. U. S. Government Printing Office, Washington, D. C.

Tiet, J. H.

1928 The Middle Columbia Salish. University of Washington Publications in Anthropology 2:83-128. Seattle.

Titmus, G. L. and J. C. Woods

1986 An Experimental Study of Projectile Point Fracture Patterns. Journal of California and Great Basin Anthropology 8(1):37-49.

Topik, C. T., N. M. Halverson, and D. G. Brockway

1986 Plant Association and Management Guide for the Western Hemlock Zone, Gifford Pinchot National Forest. USDA Forest Service, Pacific Northwest Region, Portland.

Towner, R. H. and M. Wharburton

1985 Projectile Point Rejuvenation: A Technological Analysis. Paper presented at the 50th Society for American Archaeology Conference, Denver.

Tringham, R., G. Cooper, G. Odell, B. Voytek, and A. Whitman

1974 Experimentation in the Formation of Edge Damage: A New Approach to Lithic Analysis. Journal of Field Archaeology 1:171-196.

Voegelin, E. W.

1942 Culture Element Distributions: XX Northeast California. Anthropological Records 7(2). University of California Press, Berkeley.

Watson, W.

1950 Flint Implements, An Account of the Stone Age Techniques and Cultures. Trustees of the British Museum, London.

Webster, C. L.

1889 Ancient Mounds in Iowa and Wisconsin. Smithsonian Report for 1887, Part 1. Washington, D. C.

Wharburton, M

1980 The Stone Tools from Tell Halif, Israel: A Technological Perspective. Unpublished Master's thesis, Department of Anthropology, Washington State University, Pullman.

Wilke, P. J.

1986 Aboriginal Game Drive Complexes at the Near Whisky Flat, Mineral County, Nevada. Paper presented at the 1986 Great Basin Conference, Las Vegas.

Woods, J. C.

1988 Projectile Point Fracture Pattern and Inferences about Tool Function. Idaho Archaeologist 11(1):3-7.

Yamaguchii, D. K.

1983 New Tree-Ring Dates for Recent Eruptions of Mount St. Helens. Quaternary Research 20:246-250.

1985 Tree-Ring Evidence for a Two-Year Interval Between Recent Explosive Eruptions of Mount St. Helens. Geology 13:554-557.

Young, D. E. and R. Bonnichson

1984 Understanding Stone Tools: A Cognative Approach. Peopling of the Americas Process Series, vol. 1. Center for the Study of Early Man, University of Maine at Orono.

1985 Cognition, Behavior, and Material Culture. In Stone Tool Analysis: Essays in Honor of Don E. Crabtree, edited by M. G. Plew, J. C. Woods, and M. G. Pavesic, pp. 91-131. University of New Mexico Press, Albuquerque.

Previous Investigations

Lithic Technology: Background

The Problem

Environment

Summary

Introduction

The Warehouse Site: 35LA822