Environment, Prehistory & Archaeology of Mount Rainier National Park, Washington
Greg C. Burtchard

Chapter 4:
THE 1995 MOUNT RAINIER ARCHEOLOGICAL RECONNAISSANCE
(continued)

LITHIC ASSEMBLAGES AND SITE VARIABILITY ON MOUNT RAINIER
by Stephen Hamilton and Greg Burtchard

Lithic materials dominate prehistoric archaeological remains at Mount Rainier National Park. In most cases, lithic remains are our only means, not only to establish the presence of a prehistoric site, but to infer function and temporal range as well. Since site testing and data recovery excavations are rare, [28] preliminary inferences typically must be made on the basis of surface evident remains alone. Because of the importance of these artifacts, and because surface survey is often the only look we can reasonably hope to have for many sites, we believe that field analyses of lithic remains during reconnaissance is an important, if limited, site recording procedure. Accordingly, basic analysis of surface visible lithic remains was a part of the field inspection procedure for sites and isolates investigated in 1995 (see Appendix B for observation categories). Individual site forms (see Burtchard and Hamilton 1998) present specific analytical and inferential results. Here, we focus on general assemblage characteristics and broader implications of the present Mount Rainier sample. General characteristics of the Park's artifact classes, technology and raw materials are considered first. Intersite variability in raw materials and debitage count is employed to generate a preliminary five-part site classification scheme. The combined field identified and curated projectile point sample is then described and related to temporal implications for human use of the Park. Finally, patterned variation in artifact assemblages is partitioned into a second typology of site and isolate groups. Throughout the discussion, please bare in mind that observations are based on small samples from surface context only. Results should be considered tentative in lieu of more extensive data recovery procedures and analyses of larger data sets.

General Characteristics of the Mount Rainier Assemblages

Even though the present sample of prehistoric materials is small, combining curated and field observed artifacts allows a somewhat more robust view of overall tendencies. Not surprisingly, combined data show a general tendency toward portable hunting-related tools, late stage manufacture and repair, and use of locally available materials. Projectile points include both dart and arrow-sized implements, suggesting a wide temporal range for the Park and/or exploitation of large body sized animals. Perhaps somewhat more surprising is the presence of a small fraction of heavier ground stone tools, suggesting at least limited, multiple task residence within Park boundaries.

Artifact Classes

At the time of writing, Mount Rainier artifacts available for analysis included those from current surface observation, previous collections curated at the Park's collection facility in Longmire and the test excavated sample from Berkeley Rockshelter (FS 86-02^d). Tables 4.3 and 4.4 list general artifact classes and total item counts for sites and isolated finds respectively. Site and isolate locations can be seen most easily on fold out map Figure 4.2, or on Park quadrant color Figures 2.10 through 2.13.

Table 4.3 Lithic Assemblage Summary, Mount Rainier Prehistoric Sites

Table Notes:
^aTotals include 1995 inventory and museum collection.
^bTotals include 1995 inventory and private collection drawings.
^cMuseum collection surface finds.
^dBerkeley Rockshelter test excavation sample (see Bergland 1988). Absence of bifaces and projectile points in the surface sample (FS 86-02c) probably reflects collector theft.
^eTotals include 1995 surface inventory, a biface fragment from 1990 site documentation and 19 debitage fragments from a sediment sample collected in 1995.

Table 4.4 Lithic Summary, Mount Rainier Isolated Finds

Table Note:
^aQuestionable cultural/Native American origin

Technology

In the non-debitage fraction of the Mount Rainier assemblages, bifaces appear to have played a primary role as cores for imported materials. Such a strategy reduces weight by minimizing nonfunctional stone. Replacement tools can be manufactured as needed on site, using only final retouch techniques. Flake debitage could be used for expedient cutting or scraping activities (not analyzed in this project). Examples of bifaces functioning as potential core/preforms come from Sites FS 90-02 (Sarvant Glaciers), FS 86-01 (Frozen Lake), FS 90-01 (Sunrise Ridge Borrow Pit), and FS 88-01 (Tipsoo Lakes). These larger, percussion bifaces are considered part of flexible biface technology, useful in mobile situations; particularly if raw materials are scarce (see Kelly [1988] for a detailed discussion of flexible biface technology).

The presence of raw material sources on the slopes and in river gravels around Mount Rainier relieves some of the need to carry an extensive tool kit. Presence of expedient cryptocrystalline (CCS) cores, and larger early stage CCS, fine-grained basalt and metasediment flake tools at sites such as FS 86-01 (Frozen Lake), FS 90-03 and 95-07 (Windy Gap One and Two), and FS 95-06 (Mother Mountain) suggests that local materials functioned as an integral part of the Park's technical organization strategy.

Overall, debitage surface assemblages from Mount Rainier reflect this technological organization pattern as well (refer to Table 4.5 below). A significant proportion of flakes are larger cortical flakes (CF = 5% of the sample) and interior flakes (IF = 21%) and shatter (S = 9%). These artifacts were part of a core reduction strategy consistent with use of local materials. Tertiary interior flakes (TIF = 26%) and biface flakes (BF = 26%) typically are associated with prepared core and biface reduction. Tool maintenance and finishing is shown by retouch flakes (RF = 14%).

Table 4.5 Summary of Lithic Debitage from Mount Rainier Surface Assemblages

Table Notes:
^atotals from 1995 inventory and museum collection bNot analyzed, (155) = Total analyzed

A broad reduction trajectory is represented in the Mount Rainier sample. It is not characterized, as one might expect in a high montane environment, by solely late stage reduction of prepared cores and bifaces or tool maintenance. Rather, it appears to be a combined, material efficient strategy using imported materials with expedient use of locally procured CCS, basalt and metasediments. Local raw materials were used to make both expedient flake tools as well as more completely formed bifacial tools.

Clarification of chronological changes (or absence of marked change) in technological organization at Mount Rainier requires more detailed studies with larger samples and temporal controls not possible here. Comparison between sites such as FS 86-01 (Frozen Lake) containing larger bifaces and dart-sized points only with site FS 86-02 (Berkeley Rockshelter) containing only late prehistoric arrow points would be of substantial value in such research.

Raw Materials

Patterned use of raw materials is an aspect of technological organization that reflects such issues as mobility patterns and the distribution of knapable stone. A number of sites in the Mount Rainier sample provide basic information regarding raw material use. Tool stone found in the Park fall into one of three major categories: obsidian, CCS (subsuming cherts and jaspers), and fine-grained basalt or metasediment. CCS emanating from several sources dominate the sample. These materials appear to be affiliated both with imported and curated tools made from sources located outside and inside the Park. The latter were used for the manufacture of expedient flake tools and some curated tools. Undoubtedly, the fact that raw materials were available on the mountain relieved some of the pressure to carry in all the materials and tools needed while in residence. General observations of materials documented with Mount Rainier sites follow.

Obsidian
Obsidian is the most obvious example of an imported material. Obsidian is found primarily in the form of finished projectile points. Only one piece of obsidian debitage, a tertiary interior flake, was found during the reconnaissance. The dearth of obsidian artifacts suggests limited access to this raw material by the people using the mountain.

Cryptocrystalline Silicates
CCS source materials were much more frequently used than obsidian and comprised the bulk of the lithic assemblages in the Park. A number of sites show the use of CCS raw material sources located on the mountain itself. Site FS 90-04 on Tum Tum Peak is a procurement locality in which flake scars on chert boulders are associated with flakes of the same material. Although a formal analysis of flakes was not conducted, it is all but certain that the site represents initial reduction of chert for tool blanks and/or cores that were taken for final reduction elsewhere. Bergland (1988) notes that red microcrystalline silicate (red jasper) artifacts from the FS 86-02 (Berkeley Rockshelter) assemblage are probably derived from a primary geologic source observed near the rockshelter. My (Hamilton) observations of the Berkeley Rockshelter artifacts are consistent with that conclusion and indicate a complete reduction of the local material into tools. It is interesting to note than at FS 86-01 (Frozen Lake), located relatively close to this same lithic source, most of the debitage is also a very similar appearing red jasper.

Site FS 95-06 (Mother Mountain Lake 5554 Site) also appears to be a single primary reduction event. Represented at the site is a single source scatter of light gray chert with an expedient, amorphous core of the same CCS material. In this case, the raw material source appears to be an exposed chert vein in Mist Park below the site and about 500 to 750 m to the east.

It is likely that Mount Rainier and its associated landforms have numerous CCS sources that could have been exploited prehistorically. During the present reconnaissance, several potential source localities were noted (without associated artifacts). A more detailed study of Mount Rainier CCS sources, their distribution and characteristics (chemical and visual attributes) would be useful in learning about raw material usage in the Park. The present data show that CCS was collected on the mountain as well as imported–the latter probably in the form of tools and cores.

As noted above, not all CCS raw materials associated with Mount Rainier's prehistoric sites are derived from local sources. Frozen Lake and Berkeley Rockshelter, for example, contain various kinds of small CCS cores in addition to the local reddish jasper noted above. At least one of the cores from Berkeley Rockshelter was derived from an alluvial nodule parent, most likely collected from a river bed. Unfortunately, because sources of CCS have not been mapped on Mount Rainier, it is difficult to unambiguously differentiate exotic from local cherts at the present time.

Various stages of CCS manufacture occurred at different sites in the Park. Berkeley Rockshelter shows a complete reduction trajectory from source to tool. At the site, much of the tool manufacture appears to be related to hunting gear maintenance, primarily arrow point production and replacement. Frozen Lake has a high frequency of large CCS bifaces but a relatively low count of debitage suggesting that bifaces were imported as blanks for tools and possibly cores from which flakes were struck as they were needed for butchering.

Single events of late stage CCS reduction is also evident at two sites. Debitage at site FS 95-07 (Windy Gap Two) suggests a single mid to late stage percussion biface reduction event of red jasper– probably locally derived. Nearby is another single chert reduction event (FS 90-03, Windy Gap One). The latter debitage cluster is primarily pressure retouch debitage suggesting either late stage manufacture, or more likely bifacial tool maintenance/rejuvenation. An associated but spatially separated dart point is probably an unrelated isolate.

Basalt and Metasediment
Basalt and metasediment are similar in appearance and appear to have played similar roles on Mount Rainier. Metasediment is slightly more brittle and fine-grained, thus, at times appears to have been used for manufacture of smaller tools. Both materials were found locally and used to supplement finer-grained CCS and obsidian tools with larger, more durable cutting tools. FS 86-01 (Frozen Lake) has the largest assemblage of large basalt flake tools presently known in the Park. Here, all of the basalt tools appear to be made from different (based on visual variation), but presumably local, sources. Visually similar fine-grained basalt, metasediment flakes and at least one flake tool are found at a number of the Mount Rainier sites. The Tipsoo Lake Two Site (FS 95-01) is a small cluster of metasediment flakes that appear to be associated with a single reduction event. The flakes are large, but appear to be derived from a prepared core. Generally, basalt and metasediment are rare in the form of debitage, but probably played a significant role as flake tools.

Coarser grained basalt was found as well, but not for tools requiring a precise cutting edge. Basalt as well as other locally derived volcanic stone cobbles were used for heavier tasks such as grinding, pounding, and cooking (FS 90-01, Sunrise Ridge Borrow Pit Site).

Other Materials
Lithic materials beyond those noted above are relatively rare in the Mount Rainier sample. Miscellaneous imported artifacts of exotic materials consisted of a pumice abrader at FS 86-02 (Berkeley Rockshelter), a pumice pipe bowl at FS 63-01 (Fryingpan Rockshelter), a quartzite cobble manuport at FS 95-03 (White River Trail Top), and various curated ground stone objects in the Park's museum collection. The latter were found at unknown localities within the Park.

Material Variability and Site Function

In a very general sense, a relationship can be expected to hold between the density and variety of materials at a site and the nature of site activities; use intensity, and/or the number of times a given location was reoccupied. In general, sites that have been repeatedly used, occupied by multitask or residential groups, and/or occupied by large groups tend to accumulate the largest variety and density of lithic remains. Single or limited use sites, or limited use task-specific places tend to have low artifact accumulations with low material type variability. Lithic procurement (quarry) sites typically have very high material counts and very low material diversity. It is important to be aware, however, that many variables intervene to create exceptions to the above pattern; that observation difficulties exist because of the focus on surface remains only; and that the pattern does not directly consider other indicators of residence such as fire modified stone or heavy stone tools. Even so, material density/diversity patterns provide a preliminary means to discriminate quantitatively between sites in a manner that is both heuristically useful and subject to rejection or refinement at a later date.

Table 4.6 shows the frequency of raw materials presently identified in the Park. Debitage to material and material to debitage ratios are included, though the authors found that with Mount Rainier's small samples, simple count comparisons illustrated variation between sites more dramatically. Note that only data from the 1995 reconnaissance are included in the table with the exception of site FS 90-04 (Tum Tum Quarry). This site is included to illustrate the extreme debitage to material type ratio characteristic of a known lithic procurement site. Figure 4.3 graphically displays site debitage count to raw material variety, showing clear cluster patterns in the analyzed Mount Rainier sample.

Table 4.6 Debitage Raw Material Variability at Mount Rainier

Figure 4.3 Plot of Debitage Counts by Material Variety (1995 Reconnaissance Data)

Four, perhaps five patterns are apparent in the debitage count/variety plot above. The known quarry site on the flank of Tum Tum Peak in the Park's southwest quadrant (FS 90-04) stands out clearly in the upper left lithic procurement corner of the graph. In the far left lower portion of the graph are sites characterized by the reduction of a single raw material type. This assemblage attribute, combined with low artifact counts and low technological diversity (not a plotted variable) suggests that these sites reflect single reduction events. In comparing sites, however, raw material types vary, as does the reduction stage and/or trajectory represented. Indeed, the four plotted single reduction event sites all appear to be somewhat different. They include: a) expedient core reduction of local chert on the east flank of Mother Mountain (FS 95-06); b) prepared core reduction of local metasediment at Tipsoo Lake (FS 95-01); c) mid through late stage biface manufacture of local jasper at Windy Gap (FS 95-07); and d) tool finishing or rejuvenation of probably exotic chert from an imported tool also located at Windy Gap.

Based on assemblage characteristics alone, it appears that the single reduction event sites cannot be placed unambiguously into a single functional category. As a group, they are technologically diverse and appear to crosscut a variety of environmental contexts. It also is important to recognize that these debitage assemblages may be directly related to an on-site processing task (e.g., expedient tool manufacture) or, alternatively, stone reduction during down-time and therefore secondary to the primary site function. Regardless of apparent inferential ambiguity, it is clear that the relatively small, technologically restricted site assemblages represent very short-term, low-intensity uses of the site locality. The fact that raw material from most of these sites (3 of 4) are from locally derived materials is probably the result of minimizing weight by incorporating locally derived materials into a functional, transportable (or disposable) tool kit. Thus, while not lithic procurement sites per se, the single reduction event sites are likely to occur close to the source locality and incorporate some of the reduction stages common to procurement localities.

With the possible exception of the two single source sites noted above and site FS 95-10 at Forgotten Creek in the southwest quadrant, sites clustered at the lower left of the plot (low count/low variety) all display characteristics consistent with low intensity, task specific use. In addition to increased material variety, debitage tends toward later stage reduction and repair expected at a hunting rather than a manufacturing stop. Exceptions include Mt. Pleasant (FS 72-02) and Vernal Park (FS 74-01) rockshelters, both of which have early stage, core reduction as well as secondary and tertiary stages represented. As mentioned above, core reduction may have occurred during down time. Thus, core reduction is expected in these small, high elevation rockshelters use of which may have involved forced down-time as hunters waited for storms to lift. The Forgotten Creek Site (FS 95-10), an upper forest level surface site in the Park's NW quadrant also fits these general parameters. However, presence of a basalt grinding stone at this site coupled with its forest/subalpine ecotonal setting suggest a broader range of activities and perhaps residence.

Several sites provide evidence of multiple task use and/or mixed group residence. These include site FS 95-10 noted above: the central cluster of fours sites in Figure 4.3–an extensive site at Tipsoo Lake (FS 88-01), Frozen Lake (FS 86-01), Sunrise Ridge Borrow Pit Site (FS 90-01), and a site on the northern flank of that ridge (Sunrise Lake Site, FS 95-11)–and the lone high density/high variety site at Buck Lake (FS 71-01) in the Park's extreme NE quadrant. The Frozen Lake Site appears to be a primary reduction and butchering location as noted above. Its exposed alpine setting makes all but very short term residence improbable. The remaining sites are situated in more protected upper forest/lower subalpine ecotonal settings. The mix of materials, artifacts and setting is consistent with the residential inference.

In sum, the variety of artifacts, technological attributes and material classes presently documented at Mount Rainier National Park suggests a limited variety of basic land-use functions. These include (but are not limited to) lithic procurement, early stage lithic reduction and manufacturing, limited-task hunting and hunting/butchering localities, multi-task localities with vegetal processing, and probable mixed group residence. Ecological as well as material contrasts appears to exist between multi-task/residential sites biased toward the upper forest/lower subalpine ecotone, versus limited task sites in subalpine and alpine settings. We urge that more refined test and analytical procedures be directed to a subset of the multi-task/residential sites and to the task-specific localities to better establish functional patterns among Mount Rainier prehistoric site localities. Until such time, patterns discussed here provide a preliminary means to characterize the Mount Rainier prehistoric archaeological record on functional grounds.

Radiocarbon, Stratigraphy, Projectile Points and Temporal Range at Mount Rainier

Functional classes suggested in the sections above and below do not discriminate potential changing use through time, nor do they inform us at all as to the potential temporal depth for human use of the mountain. In Chapter 2, ecological arguments were developed suggesting that human use of Mount Rainier landscapes could have begun as early as about 8,500 years ago and continued throughout the prehistoric period. Unfortunately, Mount Rainier chronometric data unambiguously linked to cultural remains are limited. The only radiocarbon dates currently considered culturally meaningful are two dates from Berkeley Rockshelter (FS 86-02)–1970 ± 80 years: 20 B.C. (Beta 44528) and 1070 ± 90 years: A.D. 880 (WSU 3666). [29] Additional temporal indicators must rely on less precise stratigraphic position and projectile point stylistic attributes. While these data are not sufficient to meaningfully address possible forager to collector land-use changes through time, they nonetheless suggest a long period of use of Mount Rainier landscapes.

Stratigraphic position of chipped stone debitage between Mount Rainier C and Mt. St. Helens Y_n tephra at the Sunrise Ridge Borrow Pit Site (FS 90-01) push the temporal range back to circa 3,750 to 2,300 years for use of that site. Because of the distinct stratigraphic signature left by Mount Rainier's repeated volcanism, it is likely that additional testing at identified sites will expand our stratigraphic/temporal database. At present however, FS 90-01 and FS 86-02 remain our only firmly (if incompletely) dated sites in the Park.

Variation in artifact type, particularly in hafting width, suggests a still longer temporal range for Mount Rainier's prehistoric record. A general, though imperfect, time-related shift from lance to dart to arrow-sized projectile points is now well recognized across North America. At a broad level, large-sized piercing (or multi-task piercing-cutting) points dominated North American hunter-gatherer durable tool kits from earliest entry to about 6,500 to 5,000 years ago. Clovis and Folsom styles– named after type sites in New Mexico but widely distributed across the west–are among the best known of these early tool types. Perhaps responding to the need to exploit medium to large sized fauna at greater distance, the arm extending throwing stick (atl atl) became a common hunting tool during the mid-Holocene. Projectile points attached to dart foreshafts thrown by these implements were smaller than early Holocene lance points, but still large enough to fill multiple piercing and cutting functions. Windust and the locally common, willow leaf shaped Cascade point are examples of dart points typically assigned a mid-Holocene date. Plausibly responding to increasing pressure to exploit still smaller, faster moving game, bow and arrow technology became common between about 2,000 and 1,500 years ago (though the atl atl and dart-sized projectile points continued to be used locally). While morphologically varied, arrow points typically are small and intended solely for piercing.

Because they were designed to be mounted on progressively smaller shafts, lance, dart and arrow points can be discriminated objectively (if approximately) by the hafting width of the base (see Corliss 1972; Thomas 1978). Here, hafting width of greater than 16 mm is assumed to be associated with lance technology. A large assemblage dominated by such large points would be of early Holocene origin; certainly dating earlier than 5,000 years ago. A limited assemblage or, more commonly, an isolate can be very tentatively assigned to this same period. Points are considered to be dart-sized in the haft width range of 7 to 16 mm. An assemblage dominated by dart-sized points typically is assigned a mid-Holocene age range, circa 5,000 to 2,000 years ago. Because darts continued to be used into the late Holocene in the Cascades (cf., Burtchard 1990:147), isolates, particularly Cascade style isolates, should be evaluated with caution. Arrows are assumed to have a haft width of less than 7 mm and are assumed to appear after 2,000 years ago. Accordingly, presence of arrow-sized points, in an assemblage or as an isolate, provides inferential grounds for late Holocene use. [30]

Table 4.7 contains the results of a basic analysis of projectile points observed in the field or curated at Park headquarters in Longmire. Period and age ranges are based on the assumptions outlined above, and should be considered tentative in the absence of larger sample sizes. Such limitations notwithstanding, it should be noted that the wide range of types and haft widths are consistent with a long temporal range for human use of Mount Rainier. These data suggest use from early to late Holocene. Firmer clarification of the extent of that use awaits further archeological research.

Table 4.7 Mount Rainier Projectile Point Types and Inferred Age

Table Notes:
^aMeasurement from an artifact sketch.
^bSee descriptions and measurements in Bergland (1988:50) for a slightly different breakdown.

Artifact Variability, and Site and Isolate Groups on Mount Rainier

In Figure 4.3 and associated text, Mount Rainier sites were grouped in a very preliminary fashion into four or five site groups based on a ratio of lithic debitage count and raw material variety (with interpretive complexity added by context and general assemblage considerations). Here, sites and isolated finds are grouped by reference to their broader assemblage composition irrespective of material type. As before, our intent is to provide a descriptive partition of Mount Rainier prehistoric remains that carries implications for site function. Multiple artifact site groups are considered first and isolates second. A summary consideration based on combined criteria concludes the section.

Assemblage-based Site Groups

Sites containing multiple artifacts are grouped by assemblage structure as an index of site types in Table 4.8 below. The table groups sites into three major categories based solely on surface evident remains. Group 1 is defined by the presence of a broad lithic assemblage; including debitage, heavy tools (hammer stones, grinding stones and other large, course grained lithic tools) and/or fire cracked rock. Lightweight tools (cores, bifaces, flake tools and projectile points) may or may not be present. Group 2 includes sites with only debitage and lightweight lithic tools. Group 3 includes sites with debitage only (a quartzite manuport and bird gastroliths are not counted as tools for purposes here). In essence, this grouping includes sites most likely to be associated with longer term, multi-task use (Group 1), a variety of more limited and/or shorter term mixed task uses (Group 2), and a set of sites more likely to have been task-specific or very short-term use localities (Group 3).

Table 4.8 Sites Grouped by Lithic Assemblage Similarities as an Index of Site Types.

Table Notes:
^a(test sample) results are from Bergland's (1988) work at Berkeley Rockshelter
^bsmall quartzite cobble manuport
^cbird gastroliths (gizzard stones)
Light tools: fine-grained lithic tools including cores, flake tools, bifaces, and projectile points.
Heavy tools: course grained lithic tools, primarily ground stone.

As noted above, Group 1 sites have assemblages that include debitage, heavy tools (particularly ground stone) and/or fire cracked rock indicating cooking activities. We expect lightweight tools to be present as well, though in one case none were found. The artifact assemblage is diverse with evidence that plant food processing is part of a broader range of activities. Combined presence of such remains is consistent with a mixed sex and age, multi-task residential camps. Furthermore, the three Group 1 sites– FS 90-01 (Sunrise Ridge Borrow Pit), FS 95-11 (Little Sunrise Lake) and FS 95-10 (Forgotten Creek) all occupy the upper forest to lower subalpine setting expected to be associated with such use. Interestingly, two of the sites–FS 90-01 and FS 95-11–also appear in the central (multi-task/residential) cluster generated by lithic raw material/debitage count results displayed earlier in Figure 4.3.

Group 2 sites (surface evident lithic debitage and light tools only) can be divided into three subgroups by markedly different tool to debitage ratios. Group 2a sites–FS 86-01 (Frozen Lake) and FS 90-02 (Sarvant Glaciers)–have a very high fraction of formed tools and cores to debitage (above 75%). Indeed in the latter case, tools actually exceed the debitage count. We believe that the high fraction of formed bifaces at these locations is consistent with use for short-term hunting and butchering activities. Both sites are situated in exposed alpine tundra settings, reducing the probability of sustained residence. Combined environmental context and assemblage characteristics suggest mountain goat processing. Because of its high material variety, FS 86-01 grouped with the central moderate intensity/residential cluster in Figure 4.3. It is more likely, however that the relatively large assemblage and material variety is a product of repeated short-term use. Unfortunately, the Sarvant Glaciers site remained under snowpack in 1995 and could not be reevaluated in the field (and hence was not included in Figure 4.3).

Group 2b includes a variety of sites with a tool to debitage ratio falling above 10% and below 50%. As with other Group 2 sites, absence of ground stone and fire cracked rock suggests very shortterm use by a segment of a broader social group. The moderate tool to debitage ratio and absence of heavy tools is expected at base camp sites in which people finished and maintained tools, and perhaps conducted limited processing tasks. Such sites may be short-term locations such as resting stops.

Two rockshelters fall within Group 2b–FS 72-02 (Mt. Pleasant) and FS 86-02 (Berkeley Rockshelter). In the latter case, it is interesting to note similarity between surface and test excavated remains. Despite the larger test sample, the only "heavy tool" was an single stone abrader/shaft smoother, suggesting hunting tool maintenance anticipated for such sites. Small surface sites at Windy Gap (FS 90-03), Mother Mountain (FS 95-06) and Middle Spunkwush Lake near Windy Gap (FS 95-08) are all in similar subalpine context and could plausibly have supported short-term hunting camp functions. The larger site at Tipsoo Lake (FS 88-01) is the most problematic of the group. Its higher raw material variety grouped it with the more residential/intensive /repeated use group in Figure 4.3. Ongoing test excavations at the site (Sullivan pers. com. 1996) indicate that a higher fraction of heavy tools may be present in subsurface context, suggesting that the site is better placed with the multi-task, residential group.

Group 2c sites are distinguished from those above by very low tool to debitage ratios (below 10%). Though lumped together by the assemblage criteria applied here, the sites differ substantially in site context, material variety and probable function. Despite its lack of surface evident heavy tools, site FS 71-01 at Buck Lake has both the highest surface count and greatest raw material variety of any site currently documented in the Park. This site's dispersed scatter above and around the northeastern margin of the lake suggests repeated use unmatched by other sites in the sample. Further testing is recommended to determine the extent to which the surface pattern holds (suggesting repeated shortterm hunting camp use) or adds a greater array of heavy tools to the assemblage (suggesting multi-task, mixed residential use). Because the subalpine setting near the forest margin, and material density and variety, more closely matches Group 1 residential sites than other Group 2 sites, we consider it likely that Buck Lake ultimately will be shown to be a repeated use residential site. Current data, however, suggest repeated, shorter-term, task-specific use.

Site FS 95-07, a single material site at Windy Gap, qualifies as a Group 2 site solely on the presence of a single use flake scraper. As with nearby Windy Gap Site FS 90-03, task-specific camp function is probable, though the dominant activity here (at least as indicated by the lithic remains) appears to have been a single lithic reduction event using material collected at a primary source located somewhere in the vicinity.

Group 3 sites are those that contain only debitage, or debitage with a small apparently unused quartzite cobble (FS 95-03), or bird gastroliths (FS 95-05). The single Group 3a site is the chert quarry on Tum Tum Peak (FS 90-04), which accounts for its high debitage count of single source material (see Figure 4.3). Remaining Group 3b sites vary in the specific character of the debitage assemblage, suggesting functional, cultural, or environmental contextual differences (limiting the utility of group classification). Vernal Park Rockshelter (FS 74-01) almost certainly served short-term camp functions similar to other rockshelters included in Group 2b. The site has been used repeatedly as an historic period camping place, and it is likely that surface evident formed tools simply have been collected and removed. Single source assemblage characteristics of the small Tipsoo Two Site (FS 95-01) suggest a single event, perhaps related to more general use of the lakeside environment evident at the larger, more complex Tipsoo One Site (FS 88-01). Two adjacent small sites on Sunrise Ridge (FS 95-03 and FS 95-04) and site FS 95-05 on the upper cirque rim overlooking Mist Park appear to be single event stops or very limited use areas. Again, such inferences must be considered suspect in the absence of more robust test data.

The three to six-part site grouping suggested here is based on sites documented during the present reconnaissance effort. We emphasize again, however, the limitations of the database upon which the classification scheme relies. Small sample size and data skewing from surface observation in a forested, volcanically active depositional environment undoubtedly color what otherwise seem to be quantitatively sound arguments. As more sites are found and more test data become available, the groupings will need to be changed to fit increasing site diversity. Nevertheless, the site categories defined below provide preliminary organization of Mount Rainier sites on the basis of observed assemblage structure.

Mount Rainier Isolated Find Groups

By definition, isolate assemblages (1 to 2 artifacts) are not amenable to refined quantitative taxonomy. Generally, we expect isolates to represent specific, non-repeated short-term events or tasks–most often spent shots, or lost or broken and exhausted tools. Table 4.9 lists currently documented isolates in Mount Rainier National Park. Be aware, however, that several, perhaps many, of these isolates may be more than they appear. Because Mount Rainier contains Holocene paleosols buried under varying depths of volcanic tephra, it is likely that some localities that appear to be isolates are surface manifestations of substantially larger, more complex buried sites. Both site and isolate locations are shown on Figure 4.2, and on the color quadrant maps in Chapter 2.

Table 4.9 Isolated Finds Grouped by Similarity of Lithic Remains

Table 4.9 is intended more as a descriptive listing of isolated remains than an attempt to arrive at functional divisions. Even so, a few comments are in order about the contents of each category. Group A isolates are solely projectile points. Such finds are among the most likely tools to be lost during a hunting event. Accordingly, Group A are among the most likely artifacts to represent truly isolated, idiosyncratic use of the landscape. Group B isolates are non-projectile point tools. They may represent truly lost tools or a very limited (or buried) specific task locality such as an expedient butchering site. Group C includes isolates with both debitage and tools. Because more than one item is involved, the probability is increased that the finds represent specific task localities in which tools were maintained, or a flake tool made and used (as opposed to simple tool loss). Group D isolates are debitage only. If flake tools were present, they have been collected and removed or buried. Like Group C, debitage is not likely to reflect simple loss. Accordingly, despite their unassuming character, Groups C and D, and perhaps Group B isolates should be considered likely candidates for site classification when subsurface test procedures become possible. Group E isolates are manuports. Manuport functions are unknown, but could plausibly represent raw material transport. Since they cannot be linked directly to predominantly prehistoric technology, isolated manuports cannot be linked unambiguously to prehistoric use. However, manuports included in Table 4.9 are not associated with historic period rubbish and, therefore, are considered most probably prehistoric in origin.

In sum, lithic analyses pursued here have been devoted to developing interpretive discussion of different aspects of lithic assemblages that make up Mount Rainier's prehistoric archaeological record. The foregoing series of tables and accompanying text are intended to provide: 1) a descriptive summary of the currently analyzed lithic sample; and 2) preliminary temporal and functional interpretations based on available assemblage characteristics. These data, while limited, suggest that: a) Mount Rainier landscapes have been used for a period of time exceeding 5,000 years in the prehistoric past; b) such use continued throughout the Holocene; and c) quantitatively discernable, functionally relevant variation exists among sampled sites and isolates in the Park. In Chapter 5, combined assemblage and environmental data are used to develop a working model of Mount Rainier site types and distribution. The section below concludes this chapter's review of reconnaissance results by evaluating the relationship between Mount Rainier's current archaeological database and a set of environmental variables commonly assumed to be linked to presence of prehistoric remains.