UC Merced Journal of California and Great Basin Anthropology

Title Boulders to Bifaces: Initial Reduction of Obsidian at Newberry Crater, Oregon

Permalink https://escholarship.org/uc/item/2dp718q0

Journal Journal of California and Great Basin Anthropology, 13(2)

ISSN 0191-3557

Author Ozbun, Terry L.

Publication Date 1991-07-01

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California Journal of California and Great Basin Anthropology Vol. 13, No. 2, pp. 147-159(1991).

Boulders to Bifaces: Initial Reduction of Obsidian at Newberry Crater, Oregon

TERRY L. OZBUN, Lithic Analysts, P.O. Box 684, Pullman, WA 99164.

JL HE archaeological remains in Newberry 700,000 years ago, basalt flow eruptions from Crater, Oregon are associated with a large the Newberry vent began, creating the volcano prehistoric obsidian quarry complex. The main (MacLeod et al. 1981:87; MacLeod 1982:125). attraction to Newberry Crater obsidian through­ Between about 500,000 and 100,000 years ago, out prehistory appears to have been the avail­ the caldera formed as the result of a series of ability of boulder-size blocks (greater than 256 collapses following enormous tephra eruptions, mm. diameter, Wentworth scale). These large- some more than 40 km.' in volume (Higgins sized obsidian blocks were used to make ex­ 1973; MacLeod et al. 1981; MacLeod 1982: tremely large cores, bifaces, and projectile 125). Mount Mazama (Crater Lake), located points. Although manufacture of large tools some 100 km. to the southwest (Fig. 1), erupted may have been the norm at quarry areas approximately 6,800 years ago (Bacon 1983) throughout western North America, it is infre­ depositing a blanket of tephra nearly one meter quently reflected in lithic assemblages from sites thick over the entire volcano. Approximately across the landscape because of use-life reduc­ 1,550 years ago, a pyroclastic eruption from a tion prior to discard. vent at the southern edge of the caldera (Fig. 2) Experimental and archaeological data are deposited a narrow plume of pumice several used here to determine aspects of prehistoric meters thick in the southeastern quadrant of the lithic technology represented at three sites crater and east across the High Lava Plain. Ash within the obsidian quarry complex in Newberry from the same vent flowed toward Paulina Lake Crater. These data indicate that several dif­ approximately 1,350 years ago followed imme­ ferent sets of technological activities occurred at diately by extrusion of the Big Obsidian Flow sites within the crater; however, all involved (Macleod et al. 1981:91; MacLeod 1982:125). reduction of obsidian into large-sized bifacial Of the nine exposed obsidian flows (Fig. 2), artifacts. five stratigraphically overlie Mazama tephra and, therefore, must be less than 6,800 years SETTING old. The other four are older than 6,800 years. Newberry Volcano, a quaternary shield Archaeological quarry/workshops are associated volcano, covers an area of approximately twelve with several of these obsidian sources and other hundred square kilometers on the western High prehistoric sites are located in many areas of the Lava Plain (Fig. 1). The caldera is 6-8 km. in crater. Three of these sites have been test ex­ diameter and contains two lakes (Paulina Lake cavated and analyzed (Flenniken and Ozbun and East Lake), numerous cinder cones, and at 1988). Data obtained from investigation of least nine obsidian flows (including two separate these three sites are used to examine one aspect East Lake flows) (Fig. 2). of prehistoric lithic technology: large initial The quaternary geology of Newberry Vol­ artifact size as a planned design concept to max­ cano is extremely complex. Approximately imize the use-life of obsidian tools across the 148 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

COWLITZ FALLS SITE

WASHINGTON

OREGON

BEAR SADDLE SITE

GRAYLING SPRINGS SITE ',LAVA ISLANO ROCKSHELTER DUSTY MINK SITE \ |^ ^-- ^_

DIAMOND LIL SITE f O NEUBERRY VOLCANO PEPPER ROCKSHELTER '

HIGH LAVA PLAIN

O Crater Lake

100 km

Fig. 1. Location of Newberry Volcano and archaeological sites with Newberry obsidian artifacts. landscape away from lithic resources. BIG OBSIDIAN FLOW SITE Results of test excavations at three sites within Newberry Crater and the lithic analyses The Big Obsidian Flow site (35DS212) is of the recovered artifacts form the basis for this one of several quarry/workshop sites located study. Raw data, concerning topography, strati­ within the crater. The site consists of a con­ graphy, and lithic assemblages, and interpreta­ tinuous scatter of artifacts around the entire tions of those data are presented in detail else­ periphery of the Big Obsidian Flow (9 km. in where (see Flenniken and Ozbun 1988) and, circumference) and extends from the edge of the therefore, will not be reproduced here. Further flow only about 15 m. (Fig. 2). Artifacts were test excavations within the crater have recently not found on top of the flow, probably because been completed (Connolly 1991). of the treacherous and jagged topography. Pre- BOULDERS TO BIFACES 149

OBSIDIAN FLOW

Fig. 2. Newberry Crater interior and investigated archaeological sites. historic use of the Big Obsidian Flow site must movable obsidian blocks occurred. Analysis of be less than 1,350 years old since the material debitage and formed artifacts from the Big Ob­ quarried did not exist before that time. sidian Flow site have allowed reconstruction of Large blocks (i.e., boulder-sized and much the obsidian reduction sequence employed by greater) of flakeable obsidian are easily acces­ prehistoric knappers at the site (Fig. 3). The sible at edge of the flow. Archaeological evi­ analysis indicates that bifacial cores and bifacial dence that large flakes were removed directly blanks were manufactured at the source and ex­ from the blocks is indicated by negative flake ported from the site. Bifacial cores, bifacial scars evident on many of the obsidian blocks. core fragments, and debitage from their manu­ Also, the huge (greater than 40 cm. diameter) facture, recovered from the excavations, suggest hammerstones associated with the site and the that the cores frequently exceeded 20 cm. in presence of large detachment scars on large length (Fig. 4). The size of the bifacial blanks discarded bifacial cores and bifacial core produced was approximately 12 cm. in length, fragments suggest that direct flaking of im­ estimated both from blanks broken (Figs. 5 and 150 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Flake Blank

Bifacial Blank

Preform

Point

Bifacial Core 7 cm. Fig. 3. Reconstructed obsidian reduction sequence for Newberry Crater archaeological sites.

6) and rejected (Fig. 7) in manufacture, and Artifacts from the lower component are from debitage corresponding to that stage of concentrated at the surface of a paleosol directly reduction (Flenniken and Ozbun 1988). beneath Mazama ash. Charcoal associated with the artifacts in the paleosol has been dated to PAULINA LAKE SITE 8,210 ± 60 B.P. (Flenniken and Ozbun 1988) The Paulina Lake site (35DS34) is located and 7,080 + 80 B.P. (Connolly 1991). The on the southern shore of Paulina Lake (Fig. 2) lithic assemblage of the lower component exhib­ and exhibits two distinct components. The its emphasis on production of dart points (Fig. upper component (above Mazama ash) contains 8). Measurements from stage-diagnostic formed evidence for an emphasis on tool finishing at the artifact fragments and late-stage bifacial thinning site. The reduction sequence is represented by flakes were used to estimate the size of the dart arrow point blanks, arrow point preforms, and points produced at the site. Estimates from the broken arrow points. debitage are based on measurement of the BOULDERS TO BIFACES 151

Fig. 4. Bifacial core from the Big Obsidian Flow site (length is 24 cm.). lengths of late-stage percussion bifacial thinning SOUTHEAST CORRIDOR SITE flakes only. This debitage category most ac­ curately reflects intended formed artifact size The Southeast Corridor site is situated near because complete flakes of this category extend the rim of the crater along a natural access route from the margin to the midline of the biface in (Fig. 2). Materials recovered from the site most cases and, therefore, their lengths are overlie a thick deposit of 1,550 year-old pumice representative of approximately one-half the derived from a Newberry Volcano eruption. width of the biface from which they are re­ Two small concentrations of debitage and moved. Unbroken late-stage percussion bifacial formed artifacts comprised the entire site thinning flakes recovered from the lower com­ assemblage. These concentrations were deter­ ponent of the Paulina Lake site average 29 mm. mined to be segregated reduction locations in length, suggesting a width of 58 mm. for the (SRLs) representing single individual flint- finished blanks. A proportionate length calcu­ knapping events (Flenniken and Stanfill 1980; lated on the basis of recovered formed artifacts Flenniken and Ozbun 1988). SRLs are rela­ from the site and experimental data indicates a tively rare in the archaeological record since length of 10-12 cm. This estimate corresponds most lithic scatter sites contain accumulated with the estimated size of blanks leaving the Big debitage from repeated flintknapping events at Obsidian Flow site (Fig. 3), although the two the same location over a period of time. Analy­ sites are temporally unrelated. sis of debitage from an SRL results in a more 152 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 5. Bifacial blank fragment from the Big Obsidian Flow site (length is 5.4 cm.). precise definition of reduction technology since tioned as dart point preforms, lanceolate points residues from different events are not mixed (cf. Connolly and Baxter 1986; Daugherty et al. together. 1987), or trade bifaces (cf. Scott et al. 1986). The obsidian reduced at the Southeast COMPARISON OF REDUCTION Corridor site was likely obtained from the East TRAJECTORIES AND SITE FUNCTION Lake Flows quarry located less than one kilo­ meter from the site. One SRL was produced as All three of the sites described above are the result of reduction of flake blanks to bifacial located in close proximity to expansive obsidian blanks by percussion. The other SRL involved sources and were likely situated wholly or partly reduction of a flake blank to a pressure-flaked to take advantage of the abundant lithic re­ biface. The length (greater than 2 cm.) of the sources in the crater. Although each analyzed pressure flakes recovered suggest that a large assemblage represents different stages of reduc­ biface was produced and exported from the site. tion and potentially different end products, all Although the site dates to a period traditionally are related to the same or similar reduction associated with bow and arrow technology, the technologies (Fig. 3). The lithic technological assemblage does not appear to be related to activities that occurred indicate a pattern of arrow point production. The bifaces exported locationally staged manufacture of bifacial from the Southeast Corridor site may have func­ artifacts within, and outside of, the crater. BOULDERS TO BIFACES 153

Fig. 6. Bifacial blank fragment from the Big Obsidian Flow site (length is 6.3 cm.).

Quarry sites at source locations in the such as the Southeast Corridor site and sites in ;rater, such as the Big Obsidian Flow site, were the crater's western access corridor along jsed for obsidian procurement and reduction to Paulina Creek (Ritchie 1987), were used as Difacial cores and blanks. At places such as the temporary stopping places for intermediate or Paulina Lake site and other lake shore sites (cf. final reduction stages while in transit from Flenniken 1987), flake blanks and bifacial quarries to locations outside the crater. Dlanks manufactured at the quarries were All sites investigated were remarkable in the :hinned and pressure flaked into projectile large scale of items manufactured. Cores and points. Sites in the crater located on lake shores blanks produced at the quarries are large and appear to have been repeatedly occupied because correspond proportionally with the sizes of pre­ 3f the available permanent water supply and forms and points produced at other sites located possibly other lacustrine resources. The dur­ within the source area. Assemblages analyzed ation of these occupations may have been longer range in age from at least 7,000 years to less ;han at other sites without water. Duration of than 1,350 years old. Although smaller arrow 3ccupation probably correlates well with com­ points were manufactured at some of the youn­ pletion of reduction trajectories and accumula­ ger sites, large bifaces were also produced. tion of the corresponding debitage and broken Large-sized artifacts were exported from New­ 3T rejected formed artifacts. Other locations. berry Crater for at least 6,000 years. 154 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 7. Bifacial blank (length is 12.5 cm.) from the Big Obsidian Flow site. This blank was rejected at the quarry because of a flintknapping error.

SIZES OF ARTIFACTS FROM River approximately 30 km. west of the New­ NON-QUARRY RELATED SITES berry Caldera (Fig. 1). Sixty and eighty-seven percent, respectively, of the materials analyzed Materials exported from Newberry Crater were characterized as Newberry Volcano obsid­ were not generally transported long distances ian, although the sample sizes were small (n= (R. Hughes, personal communication 1990) for 15 each). Most of the points recovered from use or trade although a small number of artifacts the Dusty Mink site were related to arrow point from the western Cascade Range (cf. Flenniken technology; however, the other lanceolate and et al. 1990) and as far away as Cowlitz Falls, dart points and preforms (n = 3) average 55 mm. Washington (Fig. 1) have been identified as in length. Three complete, or nearly complete, Newberry Volcano obsidian by x-ray fluores­ dart points recovered from the Grayling Springs cence (XRF) analysis (Ellis et al. 1991). site average 32 mm. in length. Intensive prehistoric use of Newberry obsidians More than half (52%) of the artifacts was largely restricted to the flanks of the sourced by XRF (n=103) from Lava Island volcano and the Upper Deschutes River basin Rockshelter (35DS86), 30 km. northwest of the (C. Skinner, personal communication 1990). caldera (Fig. 1), have been attributed to New­ Artifacts from two sites in the Upper berry sources (Minor and Toepel 1982). Lan­ Deschutes River basin were found to have New­ ceolate bifaces found in a cache at this site berry Volcano obsidian as the primary raw ma­ (n = 28) were all complete and averaged 76 mm. terial source (XRF analysis; McFarland 1989). in length. The technological context of these The Dusty Mink (35DS502) and Grayling cached artifacts has been debated in the liter­ Springs (35DS381) sites are located on Fall ature (cf. Scott et al. 1986); however, all agree BOULDERS TO BIFACES 155

Fig. 8. Stage diagnostic formed artifacts from the Paulina Lake site: preform fragment (upper left); bifacial blank fragment (upper right); dart point fragment (lower left); exhausted dart point (lower right). Length of exhausted dart point is 4 cm. the artifacts are not projectile points discarded itself, lanceolate (n = 5) and dart (n = 7) points as a result of use-life exhaustion. The cache found outside the cache are considerably smaller context probably accounts for the somewhat (46 mm.) in average length. larger average size of these artifacts in com­ Lithic assemblages from the western Cas­ parison with other artifacts found at more typ­ cade Range sometimes contain small proportions ical sites. Even at the Lava Island Rockshelter of Newberry obsidian, usually reworked formed 156 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY artifacts (Nilsson 1989). Two complete ex­ pects of curated technologies, such as tool main­ hausted dart points recovered at the Diamond tenance through rejuvenation and lateral cycling Lil site (35LA807), for example, were traced to (Schiffer 1972), are also evident at these sites. Newberry Volcano (Flenniken et al. 1990). Curated technologies, however, cannot be de­ These points measured 19 and 26 mm. in scribed on the basis of subsistence-settlement length. Only small, unidentifiable biface frag­ strategies without regard for local and regional ments were traced to Newberry in assemblages lithic raw material distributions or tool pro­ recovered at the Bear Saddle site (35LIN301; duction and maintenance costs. This contextual Nilsson 1989) and Pepper Rockshelter (35 aspect of raw material procurement and use has LA801; Churchill and Jenkins 1989). These been referred to as the "lithic landscape" small artifacts are probably characteristic of (Gould and Saggers 1985; Bamforth 1986). Newberry materials that have been transported The region that was occupied by the prehistoric over 100 km. across the crest of the Cascade people who used Newberry obsidian (mostly to Range. the north and west) is an area where obsidian tool stone is relatively scarce compared to areas LARGE INITIAL SIZE AS A PLANNED south and east of the crater. While other DESIGN CONCEPT sources were used, evidence from the sites dis­ Comparisons of lithic assemblages between cussed here indicate that tool-kit curation was sites located in a source area such as Newberry practiced on Newberry materials. As the New­ berry obsidian was transported from the quarry, Crater and sites located at various distances use of that material decreased and was replaced away from the source indicate that use of that by a variety of other materials. material is conditioned by tool curation prac­ tices. Binford (1980) described anticipatory Among the sites in the Newberry use-region technology as "curated," meaning that elements discussed above, the proportion of the assem­ of it are designed for easy transportation and blages containing Newberry materials decreases long-term use. Certain aspects of lithic assem­ from nearly 100 percent at the quarry, to one blages have been distinguished as diagnostic of half or more at sites within the intensive use curation. Binford (1979:268) noted that area, to only small percentages at distances "staging behavior" or the stepped manufacture where other sources dominate the region. This of tools at various locations on the landscape is fall-off phenomena has been documented and characteristic of curation. Bamforth (1986:38) discussed elsewhere with reference to exchange systems as well as curated technologies (Gould described curated tools as and Saggers 1985). effective for a variety of tasks, . . . manu­ Individual artifacts within the system also factured in anticipation of use, maintained through a number of uses, transported from change as they are used and maintained. As a locality to locality for these uses, and recycled result of attrition, breakage, and rejuvenation, to Other tasks when no longer useful for their artifacts become smaller through their use-lives primary purposes. until they are laterally cycled to other uses or Analyses of the Newberry source area lithic discarded. Evidence for this process has been assemblages and assemblages from other sites demonstrated through analysis of artifact sizes at where Newberry obsidian is found (cf. McFar­ the various sites. Dart points produced at the land 1989; Nilsson 1989; Flenniken et al, 1990) Newberry source area were estimated to be ap­ indicate that the staged manufacture was prac­ proximately 10-12 cm. in length. Similar arti­ ticed by users of Newberry obsidian. Other as­ facts found at sites located 30 km. from the BOULDERS TO BIFACES 157 source were found to be approximately half of additional transportation costs. that size, on average. Points found at sites 100 Large initial artifact size is proposed as a km. away or more were less than one third of planned design concept intended to compensate the probable original size. This reduction in the for expected use-life reductions in size through size of individual artifacts at various distances attrition, breakage, and rejuvenation processes. from the source area reflects the expected result Since long use-life is a desired attribute of the of artifact use-life reduction. The reduction technology, and since the users expect attrition appears to be very systematic and controlled. and breakage to occur, it would not be surpris­ Systematic processing of stone materials ing that they also would insure a longer func­ through stages of manufacture, use, mainte­ tional utility through manufacture of larger nance, and eventual discard is a well-document­ artifacts. Of course, the upper limits of artifact ed aspect of prehistoric lithic technologies. size are conditioned by functional requirements These technologies are organized to take advan­ of the systems in which they are used, and the tage of the mobility patterns practiced by size of the raw materials available. Since the hunter-gatherers so that tool stone is periodically available size of raw materials at Newberry replenished, and so that the tool kits in use are greatly exceeds the estimated size of artifacts maintained at levels where replacement and produced there, artifacts produced at Newberry repair of worn or broken parts can be accom­ may evidence the largest optimal size. plished when necessary (Raymond 1990). This ACKNOWLEDGEMENTS is especially important for subsistence strategies that involve intensive procurement and proces­ J. Jeffrey Flenniken (Lithic Analysts) directed sing of seasonally available food resources at fieldwork and was senior author of the original report. Jeff also drafted the figures, contributed locations where stone is not available, such as ideas for this paper, and developed the general the Diamond Lil site (Flenniken et al. 1990). research orientation utilized. Jill Osbom, Jan The cognitive design aspects of lithic tool McFarland, Linda Clark, and Steve Matz of the Deschutes National Forest graciously provided infor­ manufacture (cf. Hassan 1987) that account for mation and use of the graphics from the original their maintainability (Bleed 1986) are manifest report. Craig Skiimer (University of Oregon) and in many attributes of the artifacts themselves, Richard E. Hughes (Geochemical Research Labora­ and the assemblages as a whole. These attri­ tory) furnished general information about the archaeological distribution of Newberry obsidian. butes include the use of portable bifacial cores Jeff Markos (Lithic Analysts) made helpful com­ and blanks (Flermiken and Ozbun 1988; Wilke ments on an early draft of this paper. and Flenniken 1988), use of a "specialized repair kit that includes ready-to-use extra com­ REFERENCES ponents" (Bleed 1986:739), and point notching Bacon, Charles R. to facilitate repairable, rather than deleterious, 1983 Eruptive History of Mount Mazama and Crater Lake Caldera, Cascade Range, breakage patterns (Flenniken and Wilke 1989). U.S.A. Journal of Volcanology and Geo- Bleed (1986:739) also noted that maintainable thermal Research 18:57-115. systems are "generally light" to allow for easy Bamforth, Douglas B. transportation. This aspect of the technology is 1986 Technological Efficiency and Tool Cura­ contrary to the apparent emphasis on production tion. American Antiquity 51(l):38-50. of large-sized artifacts at Newberry Crater as Binford, Lewis R. described above. It is proposed here that the 1979 Organization and Formation Processes: Looking at Curated Technologies. Jour­ manufacture of large-sized artifacts is beneficial nal of Anthropological Research 35(3): to use-life longevity in a manner that outweighs 255-272. 158 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

1980 Willow Smoke and Dogs' Tails: Hunter- Flenniken, J. Jeffrey, Terry L. Ozbun, A. Catherine Gatherer Settlement Systems and Ar­ Fulkerson, and Carol J. Winkler chaeological Site Formation. American 1990 The Diamond Lil Deer Kill Site: A Data Antiquity 45(l):4-20. Recovery Project in the Western Oregon Bleed, Peter Cascade Mountains. Report on file at the 1986 The Optimal Design of Hunting Weapons: State Historic Preservation Office, Salem, Maintainability or Reliability. American Oregon. Antiquity 51(4):737-747. Flenniken, J. Jeffrey, and Alan L. Stanfill Churchill, Thomas E., and Paul Christy Jenkins 1980 A Preliminary Technological Examina­ 1989 Archaeological Investigations of Pepper tion of 20 Archaeological Sites Located Rockshelter (35LA801) and Katz Rock­ During the Cultural Resource Survey of the Whitehorse Ranch Public Land Ex­ shelter (35LA802). Report on file at the change. Contract Abstracts and CRM State Historic Preservation Office, Salem, Archaeology l(l):23-30. Oregon. Fleimiken, J. Jeffrey, and Philip J. Wilke Cormolly, Thomas J. 1989 Typology, Technology, and Chronology 1991 Archaeological Investigations Along the of Great Basin Dart Points. American Paulina-East Lake Highway within New­ Anthropologist 91(1): 149-158. berry Crater, Central Oregon. Oregon State Museum of Anthropology, Univer­ Gould, Richard A., and Sherry Saggers sity of Oregon, OSMA Report 91-6. 1985 Lithic Procurement in Central Australia: A Closer Look at Binford's Idea of Em- Connolly, Thomas J., and Paul W. Baxter beddedness in Archaeology. American 1986 New Evidence on a "Traditional" Topic Antiquity 50(1):117-136. in Pacific Northwest Prehistory. In: Contributions to the Archaeology of Hassan, Fekri A. Oregon 1983-1986, K. M. Ames, ed., pp. 1987 Prolegomena to a Grammatical Theory of 129-146. Eugene: Association of Oregon Lithic Artifacts. World Archaeology Archaeologists, Occasional Papers No. 3. 19(3):281-295. Daugherty, Richard D., J. Jeffrey Flenniken, and Higgins, Michael W. Jeanne M. Welch 1973 Petrology of Newberry Volcano, Central 1987 A Data Recovery Study of Judd Peak Oregon. Geological Society of America Rockshelters (45-LE-222) in Lewis Bulletin 84(2):455-488. County, Washington. Portland: USD A MacLeod, Norman S. Forest Service, Pacific Northwest Region, 1982 Newberry Volcano, Oregon: A Cascade Studies in Cultural Resource Management Range Geothermal Prospect. Oregon No. 8. Geology 44(11):123-131. Ellis, D. v., J. S. King, D. E. Putnam, and G. MacLeod, Norman S., David R. Sherrod, Lawrence Thompson A. Chitwood, and Edwin H. McKee 1991 Archaeological Excavations at Cowlitz 1981 Newberry Volcano, Oregon. In: Guides Falls, Lewis County, Washington. Report to Some Volcanic Terranes in Washing­ on file at the Office of Archaeology and ton, Idaho, Oregon, and Northern Cali­ Historic Preservation, Olympia, Washing­ fornia, D. A. Johnson and J. Donnelly- ton. Nolan, eds., pp. 85-91. Alexandria, VA: Flermiken, J. Jeffrey United States Geological Survey Circular 1987 The Lithic Technology of the East Lake 838. Site, Newberry Crater, Oregon. Report McFarland, Janine Ruth on file at the State Historic Preservation 1989 An Analysis of Two Post-Mazama Pre­ Office, Salem, Oregon. historic Flaked Stone Scatters in the Flenniken, J. Jeffrey, and Terry L. Ozbun Upper Deschutes River Basin of Central 1988 Archaeological Investigations in New­ Oregon. Master's thesis, Oregon State berry Crater, Deschutes National Forest, University, Corvallis. Central Oregon. Report on file at the Minor, Rick, and Katherine A. Toepel State Historic Preservation Office, Salem, 1982 Lava Island Rockshelter: An Early Hunt­ Oregon. ing Camp in Central Oregon. Report on BOULDERS TO BIFACES 159

file at the State Historic Preservation at the State Historic Preservation Office, Office, Salem, Oregon. Salem, Oregon. Nilsson, Elena Schiffer, Michael B. 1989 Archaeological Data Recovery Invest­ 1972 Archaeological Context and Systemic igations at the Bear Saddle Site, 35 Context. American Antiquity 37(2): 156- LIN301, Willamette National Forest, 165. Oregon. Report on file at the State Historic Preservation Office, Salem, Scott, Sara A., Carl M. Davis, and J. Jeffrey Oregon. Flenniken Raymond, Anan W. 1986 The Pahoehoe Site: A Lanceolate Biface 1990 Preliminary Analysis of the Eastgate Cache in Central Oregon. Journal of Points and Preforms from the Nicholarsen California and Great Basin Anthropology Cache, Winnemucca Lake, Nevada. Pa­ 8(l):7-23. per presented at the aimual Northwest Wilke, Philip J., and J. Jeffrey Flenniken Anthropological Conference, Eugene, 1988 Bifacial Flake Core Reduction in Western Oregon. North America: Implications for Pre­ Ritchie, Ian McKay history. Paper presented at the Great 1987 Paulina Highway Project Cultural Re­ Basin Anthropological Conference, Park source Inventory Report. Report on file City, Utah.