OBSIDIAN ACQUISITION AND USE IN PREHISTORIC CENTRAL EASTERN CALIFORNIA: A PRELIMINARY ASSESSMENT Mark E. Basgall NASMUCH AS THE DOMINANT CULTURAL REMAINS encountered at archaeological sites within most of Cali- fornia and the Desert West typically constitute flaked stone artifacts, successful decipherment of prehistoric behavior depends largely on evaluation of such debris. Especially critical is an ability to track the spatio- temporal dimensions of stone tool use across large regions. Although many raw materials employed by past populations continue to thwart efforts at reliable source characterization, techniques developed over the past two decades have provided a means of assigning obsidian to parent formations with some assurance. This has enabled archaeologists in California, where volcanic glass is both abundant and saw widespread native use, to explore a wide range of culture- historical and processual issues (Bouey and Basgall 1984; Ericson 1981; Hughes 1984; 1986; Jack 1976; T. Jackson 1974; 1986). It is the case, of course, that obsidian distributions reflect directly only geographic displacement of material from its place of origin, yet many archaeologists persist in viewing regional source profiles as relatively straightforward signatures of trade, territoriality, and other behaviors that operate within a strong sociological matrix. Numerous treatments of obsidian in Califomia exemplify this perspective. In their study of stone- working activities at the Bodie Hills quarry, for example, Singer and Ericson (1977) suggest that "analysis of production at quarry sources offers a relatively inexpensive and comprehensive means to investigate prehistoric trade," employing the results of their analysis to "estimate the quantities of items produced for export" as a function of time (1977: 171). The clear presumption here is that most or all debris found at the quarry locality relates to activities carried out by specialized task groups producing items intended for long-distance, intergroup exchange. Patterns of obsidian dispersion do in some in- stances reflect fairly elaborate processes of intra- and intergroup interaction, particularly with regard to complex social formations (cf. Torrence 1986), but among many hunter-gatherer populations lithic procure- ment is a fundamental component of subsistence- settlement organization and occurs primarily or wholly within that context. Seen in this light, source profiles often relate more directly to aspects of residential stability and group provisioning than to specialized collection forays or formalized socioeconomic ex- change relationships. This is, of course, what Binford (1977, 1979) has referred to as the "embeddedness" of raw material acquisition. Preliminary assessment of obsidian source profiles in cental-eastern California suggests that acquisition/ use patterns shifted dramatically over time, and that the behavioral processes responsible for source composi- tion during different periods relates, variously, to both incidental raw material procurement and regularized access through exchange. ConCributions of the Archaeological Research Facility Number 48, December 1989 Data from three archaeological localities are compared in an effort to identify both localized use trajectories and regional trends in obsidian procure- menL The results have implications for long-term change in subsistence-settlement organization, as well as for the structure and magnitude of social interaction across the region. THE GEOLOGICAL CONTEXT Eight major obsidian sources have been identified in central-eastem California that were exploited by prehistoric populations (Figure 1). From north to south these are Bodie Hills, Mt. Hicks (in extreme western Nevada), Mono Craters, Truman/Queen, Mono Glass Mountain, Casa Diablo, Fish Springs, and the Coso volcanic field (cf. Ericson 1977; Ericson, Hagan, and Chesterman 1976). There are further indications that several of these occur in secondary depositional contexts (though in proximity to primary outcrops), that more limited float sources occur in some areas (usually constituting small nodules of limited utility), and that as yet undiscovered, chemically distinct sources (with high quality glass) may exist (see below). Viewed from a synchronic vantage, previous archaeo- logical analyses suggest that three of the known quarry areas, Bodie Hills, Casa Diablo, and Coso, supplied obsidian to much of west-central and/or southwestern California (Bouey and Basgall 1984; Ericson 1977, 1982; Jack 1976). Truman/Queen and Mt. Hicks, in contrast, appear to have provided volcanic glass primarily to populations in west-central and central Nevada (Hughes 1983, 1985; Hughes and Bennyhoff 1986). Finally, with few exceptions, materials from Fish Springs, Mono Glass Mountain, and Mono Craters were used only in restricted zones surrounding the quafries (cf. Bettinger 1982; Hughes and Bettinger 1984; R. Jackson 1985; Hughes [this volume]). Such a synchronic perspective has obvious limitations in a region with 9000-12,000 years of time depth, and it is diachronic variability that provides insight into chang- ing patterns of subsistence-settlement organization and intergroup interaction. THE ARCHAEOLOGICAL CONTEXT Obsidian source characterization in central-eastern Califomia, like elsewhere in the West, has seen limited application in all but a few of the archaeological investigations conducted. Further, many studies have focused on so-called trans-Sierran exchange, the large- scale movement of volcanic glass from the western Great Basin into the California heartland (Basgall 1983; 1984; Bouey and Basgall 1984; Ericson 1981; Hall 1983; 1984; R. Jackson 1984; T. Jackson 1984), rather than the implications of intra-regional source profiles. The present assessment concentrates on implications of obsidian dispersion within the area, looking at data from three localites that essentially span the north- south axis of the greater central-eastern Califomia (Owens Valley to Mono Lake) region. Data are examined from: (1) CA-Iny-30, an extensive, multi- component occupation site in the extreme south of Owens Valley; (2) a series of sites within Long Valley caldera, located between Owens Valley and the Mono Basin; and (3) CA-Mno-446, situated on the southwest- ern edge of the Mono Lake basin (Figure 1). CA-INY-30 Major excavations at Iny-30, on Lubkin Creek immediately northwest of Owens Lake, resulted in identification of several spatio-temporally discrete occupations that span the last 7000-9000 years (Basgall and McGuire 1988). These components were assigned to each of the temporal periods recognized in central- eastern Califomia (e.g., Bettinger and Taylor 1974) using radiometric assay, obsidian hydration, and time- sensitive artifact forms: the Lake Mohave/Little Lake interval (pre-3200 B.P.); the Newberry period (ca. 3200-1350 B.P.); the Haiwee period (ca. 1350-650 B?P.); and the Marana period (ca. 650-100 B.P.). Lake MohavelLittle Lake Component The earliest component at Iny-30 is represented by a mixed deposit containing material attributable to both the Lake Mohave and Little Lake periods. Associated projectile points include two edge-ground Lake Mohave stem fragments, as well as one definite and two provisionally classified Little Lake series specimens. Obsidian hydration values from this portion of the site are consistently old: 19 measurements on artifacts sourced to the Coso volcanic field average 11.2 microns (sd = 1.3) and six characterized as Casa Diablo obsidian average 12.4 microns (sd = 1.2). The deposit is generally sparse and poorly devel- oped, containing few organic residues and a function- ally restricted artifactual inventory. In addition to the points, materials assigned to the component include bifaces in different stages of manufacture, both formal and casual flake tools, flake cores, a number of core- cobble implements, considerable quantities of debitage, and a handful of animal bone fragments. Flaked stone debitage (of obsidian, cryptocrystalline, and fine- grained igneous material) is dominated by tertiary percussion flakes produced during biface reduction; pressure retouch debris is conspicuous by its near 112 Obsidian in Prehistoric Central Eastern California FIGURE 1 FIGURE 1. Map of Central-Eastern California 113 Contributions of the Archaeological Research Facilty Number 48, December 1989 absence. Faunal detritus from the Lake Mohave/Litde Lake component includes artiodactyl, rodent, water- fowl, and reptile remains; large mammal forms com- prise slightly over one-third of the assemblage, small mammal/bird bone the bulk of the remainder. Newberry Component Apart from some scattered debris on the site surface, Newberry occupation at Iny-30 is represented by four buried houses, one refuse pit, and associated cultunal materials. Projectile points from these contexts include ten Elko series, 20 Humboldt Basal-notched, and a single Humboldt Concave-base form. Six radiocarbon determinations from the structures range from 1860 ? 70 B.P. (Beta-20518) to 1220 ? 70 B.P. (Beta-12660). Obsidian hydration measurements available for three house features average 5.3-5.4 microns (55 specimens; sd = 0.5-1.0) for glass attrib- uted to Coso and 4.9-5.0 microns (13 specimens; sd = 0.3-0.8) for artifacts sourced to the Casa Diablo quarry. Diagnostic beads recovered from this deposit include Saucer Olivella (Type G2 by the Bennyhoff and Hughes [1987] typology), common in the Great Basin in assemblages dating from ca. 2150-1850 B.P. Taken together, these chronological data suggest an occupa- tional peak within the late Newberry period (Basgall and McGuire 1988). The four houses are consistently large (ca. 4-5 m in diameter), with well-developed floors and collapsed structural supports, and contain a large quantity of tools, debitage, and organic remains. Caches of still serviceable artifacts, including projectile points, bifacial preforms or blades, milling equipment, and bone awl (deer metapodial) blanks were present in three of the structures. The inclusive artifact inventory encom- passes a wide variety of finished and unfinished biface forms, casual and formal flake tools, cores, core-cobble implements, milling stones and handstones, drills, beads, and bone tools. Co-dominant amounts of biface thinning and pressure retouch debris reflect emphasis on both tool production and finishing/resharpening. The substantial milling assemblage from Newberry contexts (47 items) is characterized by tools that are well-shaped, heavily used, and appear to have been extensively curated. Fauna recovered from the four structures and trash pit include a range of artiodactyls, lagomorphs, rodents, birds, reptiles, and fish. The principal taxa, however, comprise mountain sheep (78% of identifiable artiodactyl remains), black-tailed jackrabbit (78% of lagomorphs), and grebes (84% of avifauna). A rich inventory of flotation-derived paleobotanical remains shows an emphasis on resources from dryland habitats. Abundant Poaceae include SPQw1xlous (dropseed) and Q[ywpsj (ricegrass), while non-grasses are dominated by ChbnQooium (goosefoot), Ld ium (box-thon), and Mentzelia (blazing-star). Not insignificant quantities of pinyon were recovered as well. Haiwee Component Unfortunately, little discrete Haiwee period deposit was encountered at Iny-30. The most intact context dating to this interval consisted of a deep pit, underly- ing a more extensive Marana component, which contained well-developed midden and fire-affected rock. Dating is based on five Rose Spring projectile points, radiocarbon assays of 960 + 100 B.P. (Beta- 12663) and 760 ? 100 B.P. (Beta-12664), and obsidian hydration measurements on Coso artifacts that average 4.4 microns (13 specimens; sd = 0.6). Apart from projectile points, artifacts associated with the limited Haiwee deposit include percussion and pressure flaked bifaces, casual flake tools, several milling implements, and a couple pieces of modified bone. Debitage from the component shows increasing emphasis on pressure retouch and flake-based (as opposed to biface) technologies. Faunal debris reflects use of both dryland and wetland habitats, focused on the former. Although artiodactyl, lagomorph, rodent, waterfowl, reptile, and fish bone are present, rabbits appear to comprise the principal prey taxa. Paleobo- tanical remains reflect exploitation of diverse microenvironments, Sporobolous. Scizs (bulrush), Juncus (rush), and Mentzelia constituting the most abundant charred seed forms. Given the likelihood that the pit feature producing Haiwee period remains represents a rather specialized, temporally restricted event, inferences regarding general pattems of behavior during this period are, of course, highly suspect. Marana Component By all measures, aboriginal occupation of Iny-30 appears to have reached its zenith during the Marana period. lTree clusters of housepit depressions were identified at the site, together with a large bedrock milling area and several extensive midden accumula- tions. The magnitude of late prehistoric/protohistoric use is exemplified in the recovery of over 4500 Owens Valley Brown Ware sherds frno the surface of the deposit (1250 more from subsurface situations). Chronometric data from Marana contexts include a suite of 11 radiocarbon dates from seven structures that range from 710 ? 70 B.P. (Beta-22294) to 180 ? 60 BP. (Beta-20521) (a statistically "modem" assay probably relates to protohistoric occupation judging by 114 Obsidian in Prehistoric Central Eastern California associated glass trade beads, but could also represent a recent root burn episode). Some 17 Desert Side- notched and 21 Cottonwood series points were recov- ered from intact Marana deposit, together with 102 late period shell beads (mostly Thin Lipped [Ela and Elb], Ground [Hla] and Semi-Ground Disk [Hlb], and Cupped [Kl] and Bushing [K2] Olivella [Bennyhoff and Hughes 1987]) and 23 Steatite Disk beads. Seven hundred ninety-three of the 800 potsherds recovered from intact stratigraphic contexts were attributable to the Marana period. Finally, obsidian hydration values on Coso artifacts from late prehistoric deposit average 2.3-3.5 microns (44 specimens; sd = 0.1-0.8). Artifactual remains from Marana deposits at Iny-30 include a range of biface forms, drills, various kinds of flake tools, cores, core-cobble implements, an extensive milling inventory, bone tools, ceramics, as well and numerous omament types (of shell, bone, and stone). Flaked stone debitage suggests a flake-based orienta- tion, and evidence of more standard biface reduction is limited. Unlike the Newberry assemblage, milling tools from the Maran component tend to be little modified and expedient. Subsistence residues, both faunal and floral, show a decided wetland orientation. Although the same broad suite of resources are present, waterfowl (ducks and grebes) constitute nearly 50% of the assemblage; freshwater mussel (Anod nta) shell occurs in abundance during this period for the first time. The paleobotanical inventory is less diverse than that from Newberry features, with few grasses and most seeds deriving from various shrubs and herbs (especially Descurainia [tansy mustard], Ruia [ditch-grass], Sciwus [bulrush], Juncus [rush], and I jha [cattail]). Pinyon nutshell was both moderately abundant and ubiquitous across sampled Marana contexts. Long Valley Caldera Data from Long Valley derive fron a series of excavation and survey projects that together encompass much of the caldera. Providing a monitor of obsidian use during the earliest period of human occupation is the Komodo site (Mno-679), a Paleoindian encamp- ment in the central part of the valley (Basgall 1987; n.d.). Containing a seres of basally-thinned, edge- and basally-ground concave-base projectile points, the site has been provisionally dated to the early Holocene on the basis of artifact tpology and source-specific obsidian hydration measurements. A sample of 20 points chemically attributed to the Casa Diablo source yielded hydration values between 7.5-12.2 microns, averaging 9.6. These readings are considerably larger than the maximum measurements obtained on Little Lake series points in the area, and the mean cofre- sponds to an age of 8100 BP. using the curvilinear rate proposed by Hal (1984; Hall and Jackson, this vol- ume). Other mateials from the site include biface preforms and a variety of formal and casual flake tools; neither milling equipment nor organic remains were recovered (Basgall 1987, n.d.). The other excavated sites incorporated into the Long Valley sample, Mno-529 (Basgall 1983), Mno- 561 (Hall 1983), Mno-714 (Basgall 1984; R. Jackson 1986), and Mno- 1529 (Basgall 1984) are all situated on the western edge of the caldera, within the modern town of Mammoth Lakes. Although each location has produced some evidence for the pursuit of typical subsistence and maintenane activities, the assemblages are dominated by debris that relates to extensive stoneworking and, apparently, exchange-related biface production (e.g., Bouey and Basgall 1984; Hall 1984). These localities date primarily to the interval between 3150 and 1350 B.P. as indicated by artifact typology (Elko and Humboldt series points) and obsidian hydration. Finally, some of the source data used to construct projectile point and debitage profiles are drawn from an extensive survey conducted in the central and eastern portions of the caldera (R. Jackson 1985). The sites, being spatially removed from major trans-Sierran exchange corridors, may provide a more reliable reflection of intra-regional obsidian acquisition/use patterns. Both point fonns and micron values on debitage from these locations span the last 7000 years. CA-MNO-446 The Lee Vining Creek site, Mno-446, is situated on the lower reaches of the aforementioned drainage, about three kilometers southwest of Mono Lake. With an areal expanse of about 2700 sq m, excavations at the site yielded a relatively diverse inventory of flaked, ground and battered stone tools, and copious quantities of unmodified chipping debris. Organic preservation was poor, but a small assemblage of faunal remains was also recovered. Following E. L. Davis (1964:261), Bettinger (1981) characterized Mno446 as a seasonally occupied base camp, from which plant and animal procurement activities were staged and exchange- related obsidian production was implemented. The site has been subjected to considerable depositional churning, as reflected in both the vertical distribution of time-sensitive artifacts and obsidian hydration measurements (Bettinger 1981:20-21, 57); however, two gross temporal components can be identified. Hydraion readings for Casa Diablo obsidian from the upper 50 cm of deposit average 3.1 microns, corresponding to an age of roughly 1000 B.P. as 115 Contibudons of the Archaeological Research Facility Number 48, December 1989 TABLE 1 DISTANCE FROM ARCHAEOLOGICAL LOCALITIES TO KNOWN OBSIDIAN SOURCES (IN KM) CA-Iny-30 Bodie Hills Mt. Hicks Mono Craters Tnmnanjueen Mono Glass Mtn. Casa Diablo Fish Springs Coso 210 200 170 160 145 140 60 55 computed from Hall's (1984; Hall and Jackson, this volume) rate. Values from below 50 cm in depth average 4.5 microns, in accordance to ca. 2000 B.P. by the same fonnulation. Finally, temporal parameters of all diagnostic projectile points from the site (i.e., Elko, Humboldt, Little Lake, and wide-stem) fall prior to 1350 BP., and appear related to the earlier, deeper component (Bettinger 1981; Hughes and Bettinger 1984). THE SOURCE PROFILES Obsidian source profiles have been constructed for each of the thre archaeological localities using the results of seveal X-ray fluorescence (XRF) tace element analyses. Sample sizes for each site are as follows: Iny-30, 420 specimens; Long Valley, 718 artifacts (34 from Mno-679, 684 other); and Mno-446, 70 items. Distances from the archaeological localities to known obsidian sources represented in the samples are presented in Table 1. CA-INY-30 Obsidian samples, both debitage and tools, were obtained from each of the components described previously. The nearest source of abundant, high- quality glass is the Coso volcanic field, located some 55 km to the south (see Hughes 1988). The Fish Springs quarry lies about the same distance (60 kIn) away; however, obsidian from that source occurs in more limited quantities and is often of poorer quality (i.e., nodules are smaller and inclusions more prevalent). Long Valley 70 65 25 40 5 0 75 200 CA-Mno-446 40 45 10 55 40 40 125 245 Other known sources present in the Iny-30 collection are at least 140 km distant. Lake MohavelLittle Lake Component The obsidian assemblage from the earliest cultural component at Iny-30 exhibits several distinctive patterns (Table 2). The assemblage has the highest source diversity observed at the site, containing ten kinds of glass with a wide geographic dispersion. This diversity cross-cuts both the tool and debitage classes, even in groups represented by only a few specimens. Finally, chemically "unknown" obsidian types are common in the assemblage, all five groups occurring and comprising fully 23% of the analyzed sample. If obsidian from the Coso volcanic field can be considered the "local" material of choice, being close at hand, abundant, and of high quality, then it is significant that 51% of the overall assemblage derives from elsewhere (Table 3). This relationship remains vitually constant in both the tool and debitage classes. Newberry Component The source profie reconstructed for the Newberry occupation provides a marked contrast to the previous pattern (Table 2). In the first place, there is increased use of Coso obsidian and lesser diversity in the assem- blage with respect to both the number of source groups present (eight) and their geography. Non-local glasses originate primarily from the Long Valley area, some 150 km to the north, with the intermediate Fish Springs quarry contributing far less material than the Truman/ 116 Obsidian In Prehistoric Central Eastern Caltfornia TABLE 2 OBSIDIAN SOURCE COMPOSITION BY COMPONENT, CA.INY-30 LM-LL NEW HAI MAR Total FT CT DB FT CT DB FT DB FT DB FT CT DB CosojCoso? 8 8 20 41 21 95 18 11 44 35 111 29 161 Fish Springs 2 3 3 2 - 3 2 5 3 12 3 11 Tnunan/Queen 1 - - 7 - 1 - 1 1 - 9 - 2 Casa Diablo 1 1 8 12 - 16 - - 1 - 14 1 24 Mono GlassMt - - - 1 - 1 - - - 1 1 - 2 Bodie Hils - - 2 - - - - - - - - - 2 Unknown #1 3 4 3 8 - - 2 1 1 - 14 4 4 Unknown #2 - - 1 - - 5 - - - 1 - - 7 Unknown #3 - 1 1 - - 1 - - - - - 1 1 Unknown #4 1 - 2 - - - - - 1 - 2 - 2 Unknown #5 - - 1 - - - 1 - - - 1 - 1 Total 16 17 41 71 21 122 24 15 53 40 164 38 218 Note: LM-LL, Lake Mohave-Little Lake component; NEW, Newberry; HAI, Haiwee; MAR, Maana; FT, formal tool; CT, casual tool; DB,debitage; Coso?, denotes specimens with PPM values similar to,but not identical with, source standards established for theCoso volcanic field. Data derived from Basgall and McGuire(1988). TABLE 3 PROPORTIONAL REPRESENTATION OF EXOTIC (NON-COSO) OBSIDIAN BY COMPONENT, CA-INY-30 (IN PERCENT) Combined Sample Formal Tool Casual Tool Debitage Lake Mohave-Little Lake 51.3 50.0 52.9 51.2 Newberry 36.3 42.3 0.0 22.1 Haiwee 25.6 25.0 26.7 Marana 15.0 17.0 12.5 All Components 28.3 32.3 23.7 26.1 117 Contributions of the Archaeological Research Facilty Number 48, December 1989 Queen or Casa Diablo localities. Secondly, there is significant imbalance in the profiles between tools and debitage, 42% of the former deriving from non-Coso sources but only 22% of the chipping debris (Table 3). Lastly, "unknown" chemical types are conspicuous by their near absence in the sample; the group falls from 23% to 6% representation in the assemblage. Haiwee and Marana Components Given their similarities, source profiles for the Haiwee and Marana components are examined to- gether. Both exhibit further shifts in obsidian acquisi- tion/use patterns (Table 2). The overall representation of glass attributed to the Coso volcanic field increases, especially in the tool category. Source diversity remains essentially constant; however, more northern localities (i.e., in the Long Valley region) all but disappear from the assemblage. There is likewise better balance between the source profies of tools and debitage, the proportion of "exotic" glasses varying by no more than 4.5% (Table 3). The "unknown" chemi- cal types continue to occur in minimal quantities, constituting 10% of the Haiwee sample and only 3% of the Marana artifacts analyzed. Non-Coso obsidians comprise between 15% and 23% of the respective samples, but it may be significant that approximately one-tenth of this material can be traced to the Fish Springs locality (Tables 2 and 3). Implications The shifts apparent in the above source profiles have a number of implications for obsidian procure- ment and use pattens in southern Owens Valley. Perhaps the most striking aspect of the Lake Mohave/ Little Lake profie is the proportion of "unknowns" represented - 23% of the sample total. With the exception of the Komodo site (Mno-679) in Long Valley (see below), occurrence of geologically intrac- table obsidian appears to be relatively rare in the central-eastem California region; one of these glass types (Unknown #1) is present at Mno-679, the others remain apparently unreported. While it is, of course, risky to speculate on the disposition of such sources, the fact that the glasses have not been regularly encoun- tered in the region suggests the possibility that some or all of them originate to the east (i.e., Nevada), areas more poorly documented geochemically. 'Te large amount of Casa Diablo and Fish Springs obsidian in the early component (23%), together witi traces of TrumanQJueen and Bodie Hills material (4%), appears to reflect regularized interaction with areas to the north; the representadon of Coso (49%) indicates contacts with the south. All in all, the Lake Mohave/Little Lake profile suggests lithic procurement over an extremely broad area, extending north-south for a distance of at least 265 mn. The fact that obsidian groups present in the Lake Mohave/ Little Lake assemblage generally occur in the form of both formal tools and debitage may indicate one of several things: first, that early populations were highly mobile, visiting numerous quarry areas fre- quently, but acquliring raw material witiin the matrix of an extensive subsistence-settlement system (i.e., an "embedded" strategy in Binford's sense); second, that these people made forays to a diverse set of quarry locations for the express purpose of procuring lithic raw materials (i.e., direct access); or third, that these groups were involved in formalized exchange relationships that regularly provided glass from many areas in an indirect fashion (i.e., formal social exchange). In light of current data available concerning cultural adaptation during this temporal period, the first possibility seems the more likely. Lacking a developed midden and having only a sparse, functionally restricted artifactual inventory, the Lake Mohave/Little Lake component hardly appears consistent with the kind of centralization expectable were either regular forays made to procurement areas (with return) or materials being imported through exchange on a regular basis. Indeed, given the wide temporal range attributed to the early deposit at Iny-30, it almost certainly represents a palimpsest of residue discarded over seveal thousand years of time. Materials discarded during one episode reflect foraging near one quarry area (e.g., to the north), whereas those left during a later occupation relate to exploitation of entirely different resource areas (e.g., to the south or east). Implications of the Newberry period obsidian assemblage are rather different. If it is assumed that some or all of the "unknown" glass types originate in the east, then their absence from the sample would seem to imply much reduced interaction with such areas. Instead, regular acquisition appears to have been along a north-south axis - a bipolar system, if you will. Further, the high proportion of "northern" tools (28%), but reduced amounts of debitage attributable to those sources, suggests that interaction with the Long Valley area was micro-temporally disjuncL This is to say that the debitage pattern (as that of casual flake tools), indicative of on-site production activities, is consistent with more immediate, perhaps regulaized, access to glass from the Coso volcanic field. Given that the overall Newberry period artifactual assemblage was characterized by a wider range of tool forms (including curated milling equipment), abundant ecofactual debris, and, most significantly, formal l18 Obsidian in Prehistoric Central Eastern California TABLE 4 SOURCE COMPOSITION OF ARTIFACT SAMPLE FROM THE KOMODO SITE, CA-MNO-679 Projectile Point Biface Uniface Total Casa Diablo 21 3 2 26 Trunan/Queen - - 1 1 Mono Glass Mtn 1 - - 1 Fish Springs 8 - - 8 UnknownA 1 2 - 3 Unknown B 2 -2 Unknown C - 2 2 Total 31 7 5 4 Note: Unknown A at Mno-679 provides a chemical match with Unknown #1 at Iny-30; other "unknowns" from the two sites appear to be geochemically distinct. Data unpublished. structures with evidence of caching behavior, it is tempting to view the north-south interaction in terms of redundant residential shifts that probably occurred on a seasonal basis; both climatic parameters in the two areas and diagnostic subsistence remains within the Newberry deposit (Basgall and McGuire 1988) suggest southern occupation during the warmer part of the year. The Haiwee and Marana period source profiles exhibit the greatest degree of regularity. The reliance is on Coso, the nearest, high-quality obsidian, but the general balance between tools and debitage of all glass types indicates increased formalization of flaked lithic acquisition pattens. Were obsidian being procured primarily during the course of residential movements, tool and flake profiles would be expected to show greater divergence. This relationship is best illustrated in the case of Fish Springs material, which occurs in near equal pr rions within both classes. The decrease in "unknown" glasses in the Marana sample likely reflects further reduction in the zone of regular interaction. It is becoming increasingly evident that the resource intensification and settlement centralization characteristic of the historic period in Owens Valley (e.g., Steward 1933, 1938) emerged sometime after 1300 BIP. (Basgall and McGuire 1988; Basgall, McGuire, and Gilreath 1986; Bettinger 1977; 1982; 1989). The territorial control of Fish Springs obsidian reported by Bettinger (1982) in the Big Pine area begins about this time, and shell bead frequencies increase dramatically after 650 B P. (86% of r occur- rences date to the Marana period; see, Basgall, McGuire, and Gilreath 1986). In all, it seems likely that the Haiwee and Marana source profiles follow from formalized exchange relationships, and reflect regular- ized social interaction with other Paiute populations to the north and Shoshone groups to the south (and, perhaps, east). Long Valley Caldera Attempts to develop reliable source profiles for the Long Valley locality are complicated by two factors: first, many of the sites that have been investigated in detail are aligned with major exchange-related obsidian production, and hence offer a poor reflection of more typical, intra-regional acquisition/use pattems; and second, with the exception of these stoneworking camps closely related to trans-Sierran exchange, few sites have been excavated which contain contextually sound, well-dated assemblages. Nonetheless, available data do provide a basis for characterizing some facets 119 Contributions of the Archaeological Research Facility Number 48, December 1989 TABLE 5 SOURCE COMPOSITION FOR PROJECTILE POINTS BY TEMPORAL PERIOD, LONG VALLEY CALDERA Lake Mohave Little Lake Newberry Haiwee Marna Total Casa Diablo 3 8 57 9 8 85 Truman/Queen 1 2 17 6 - 26 Mono Glass Mtn - 2 8 - 2 12 Fish Springs 2 - 3 2 1 8 Mt.Hicks 1 - 1 - 1 3 Bodie Hills - - 3 - - 3 Total 7 12 89 17 12 137 Note: Point forms are attributed to periods as follows: Lake Mohave, Great Basin Stemmed series; Little Lake, Little Lake series; Newberry, Elko and Humboldt series; Haiwee, Eastgate and Rose Spring ypes (Rosegate series); Marana, Desert Side-notched and Cottonwood types (Desert series). Data derived from Basgall (1983, 1984) and R. Jackson (1985). occur in the associated chipping debris (given limita- tions in visual characterization of small flakes, further sources could easily be present in the assemblage). CA-Mno-679 Geochemical and "megascopic" (Bettinger, Delacorte, and Jackson 1984) analyses have been performed on obsidian from the Komodo site. The x- ray fluorescence results display a number of parallels with the early, Lake Mohave/Little Lake component at Iny-30 (Table 4). In the first place, a large proportion (32%-60%) of the specimens within each of the artifact categories derive froinm non-local sources. Secondly, the source composition suggests considerable interaction with a number of areas (though the focus of stonework- ing is clearly on retooling with local, Casa Diablo glass). And third, there is a sigficant representation of "unknown" obsidians in the assemblage (at 16%). Unfortunately, no XRF data are available for debitage from the site, hence precise comparisons between tools and flakes are impossible. It is worth noting, however, that visual characterization of a sample from Mno-679 (following criteria developed by Bettinger, Delacorte, and Jackson [1984]) suggests that the representation on non-Casa Diablo obsidian is dramatically higher than at most, chronologically later sites studied in the caldera. It appears that a Fish Springs, Truman/Queen, and one of the "unknowns" Other Datafrom Long Valley A population of 137 projectile points (excluding those from Mno-679) from sites in Long Valley shows some basic shifts in source representation dtrough time (Table 5). Although the sample of diagnostic points is limited for most temporal periods, available data document the most extensive use (57%) of non-local glass during the Lake Mohave interval (ca. 7000-5000 BP.). Such obsidian originates at quarries located some 65 km to the north and 75 km to the south. If not a function of the increased sample relative to earlier periods, diversity in the Newberry sample may reflect an enlarged annual range and more frequent interaction with areas to the north (i.e., Bodie Hills and ML Hicks). Still, the fact that local glases (i.e., Casa Diablo and Mono Glass Mountain) together comprise 73% of the total indicates regular tool replacement with raw materials from the caldera Only during the Haiwee period is there a high proportion of "exotic" obsidian represented; 35% of the glass derives from Truman/ Queen, 40 hn to the northeast, and 12% from Fish Springs, some 75 kn to the south. 120 of source structure. Obsidian us Prehistori Central Eastern California Source data available for debitage assemblages across Long Valley consistently indicate that Casa Diablo, the local glass of choice, is the principal material being worked at virtually all sites during every time period (Table 6). It does appear to be the case that a larger number of "early" (Newberry and pre- New- berry) collections contain higher proporions of non- local obsidian, but the pattern is far from uniform. Samples from sites A61 and A60, for example, both with early components, contain no non-Casa Diablo glass. Since the analyzed flaking debris more often relates to tool production/replacement than refurbish- ment, heavy representation of Long Valley material is not unexpected. Implications Unevenness in the quality of source data for all periods and artifact categories in Long Valley preclude in-depth evaluation of acquisition ad use pattens. Nonetheless, several trends are worth highlighting. Tools from pre-5000 BP. components (at Komodo and points from the Iake Mohave interval) show a diversity of source types, many showing consistent wpresenta- tion. As with the Lake Mohave/Little Lake component at Iny-30, this profile probably relates to an extensive settlement strategy that regularly brought populations in proximity to a variety of raw material sources. hat tools from distant quarries remained in the tool-kits suggests that they had been obtained relatively recently. TABLE 6 SOURCE COMPOSITION OF DEBITAGE FROM SELECTED SITES IN LONG VALLEY CALDERA BY TEMPORAL PERIOD. Period CD QN MGM BH UNK Total F160 LM 23 2 11 - - 36 A60 LM-LL 22 - - - - 22 A61 LL 15 - - - - 15 Mno-529 N 98 - 1 - 1 100 Mno-561 N 63 - 2 - - 65 Mno-1529 N 60 - - - 60 A33 N 46 3 1 - 53 53 A40 N-H 22 - 1 - - 23 F149 N-H 16 - 10 - - 26 Mno-714 H 15 - - - - 15 A48 H 23 - 1 - - 24 A43 H 28 - - - - 28 A45 H 37 - - - - 37 F156 H-M 23 2 3 - - 28 A42 M 21 - 3 - - 24 Total 512 7 35 1 1 556 Note: Age ascriptions are based on obsidian hydration data for specimens attributed to the Casa Diablo source; LM, Lake Mohave; LL, Litde Lake; N, Newberry, H, Haiwee; M, Marana (two notations denote a temporaly mixed assemblage); CD, Casa Diablo; QN, TrunmianQueen; MGM, Mono Glass Mountain; BH, Bodie Hills; UNK, unkmown. Data derived from Basgall (1983, 1984), Hall (1983), and R. Jackson (1985). 121 Contributions of the Archaeological Research Facility Number 48, December 1989 TABLE 7 SOURCE COMPOSITION OF OBSIDIAN SAMPLES FROM MNO-446 Debitage Projectile 0-50 cm 50+ cm Point Total Bodie Hills 4 2 4 10 ML Hicks 1 - - 1 Truman/Queen - 1 2 3 Casa Diablo 14 27 4 45 MGM/MC 7 1 2 10 Unknown - - 1 1 Total 26 31 13 70 Note: Mono Glass Mountain and Mono Craters obsidian were not chemically distinguished in this study. Data from Bettinger (1981) and Hughes (1981). Data for the Little Lake period in Long Valley (ca. 5000-3000 B?.) are rather at odds with expectations: both tools and debitage reflect near exclusive use of local glass which, at face value, would imply marked residential stability. This, however, may largely follow from the small sample available for examination. The pattern evident in points attributable to the Newberry period is, again, somewhat similar to that found at Iny-30. Glass was being obtained from a series of quarries some distance from the caldera, and the source composition is suggestive of an extensive subsistence-settlement system. Evidently, populations were interacting with areas 65-75 km to the north and south. The large proportion of Truman/Queen speci- mens present in the Haiwee period point sample, though not matched in debitage profiles, seems to indicate regular contact or movement between Long Valley and upland zones 40 km to the northeast. Given the rather ambiguous position of Long Valley vis permanent occupation during the historic period (e.g., Basgall 1983; Hall 1983), it is possible that these materials were deposited by groups that typically resided in areas outside the caldera. Finally, Marana period profiles are consistent with primary reliance on local obsidian. CA-Mno-446 Source proflles from Mno-446 exhibit some dramatic differences. Looking first at the debitage, the assemblage recovered from depths of greater than 50- cm (most of which dates to the Newberry period [3150- 1350 B?.], but some likely earlier) is dominated by glass from sources located to the south. Within this group, 87% of the obsidian derives from Casa Diablo, with twace amounts (3% each) originating at Truman/ Queen and either Mono Glass Mountain or Mono Craters (see Hughes [this volume]). Only 7% of the obsidian debitage was attributable to northern localities, all from the Bodie Hills quarry (Table 7). Although southeem sources continue to dominate flaking debris from the uppermost 50 cm of site deposit (dated post-1350 B.P.), a significant proportion of the glass is traced to the north (Table 7). The representa- tion of Casa Diablo material drops to 54%, and the Mono Glass Mountain/Mono Craters category com- prises 27% of the assemblage. However, nearly 20% of the late prehistoric debitage is attrbutable to either Bodie Hills (15%) or Mt. Hicks (4%). The projectile point profile is considerably more varied than that of the chipping debris from either temporal component. While southem source localities are also predominant, there is more even representation of the different quarries, and Bodie Hills provides its 122 Obsidian in Prehistoric Central Eastern California peak contribution of 319% of the total (Table 7). As well, the only "unknown" from the site occurs within the projectile point category (unfortunately, no data are available that might permit relating this chemical type to "unknown" glasses from previously discussed archaeological localities). Implications Bettinger (1981; Hughes and Bettinger 1984) has explored the implications of these data. With regard to shifts in the debitage profiles, he proposes that they reflect changing boundary conditions that caused a disruption in direct access to the Casa Diablo obsidian source. Two specific processes could account for these patterns: either some adjustment to the limits of a well- demarcated territory held by groups inhabiting Mono Basin proper, or, alternatively, some alteration to the annual foraging range of those populations (Hughes and Bettinger 1984:165). The later possibility is given greater credence insofar as there is no evidence to suggest that late prehistoric/historic populations in the Mono Basin were formally land-holding. Chronologically later materials, showing an increased reliance on both more local glass (Mono Craters?) and northem quarries (Bodie Hills and ML Hicks), do appear to indicate less interaction with the Long Valley area, perhaps as a result of both increased volcanic activity and territorial control by resident groups (cf. Bouey and Basgall 1984; Hall 1983; 1984). It remains unclear, however, whether the subsistence- settlement system was drastically altered or direct access to Casa Diablo obsidian became increasingly restricted. In accounting for divergence in the source profiles of projectile points and debitage during the pre-1350 B.P. period, it has been argued that the two artifact categories operated within separate "sociocultural" subsystems or matrices. The debitage is thought to represent "refuse of a utilitarian tool production system" expected to "conform to the principle of least effort" (Hughes and Bettinger 1984:168), and hence should be dominated by glass from the nearest source. By contrast, it is proposed that "individual hunters may have fashioned projectile points from obsidian obtained at sources near their places of birth, either because they were more familiar with (them) or because it was an act of socioreligious significance" (Hughes and Bettinger 1984: 168). In this case, source variability follows from non-virilocal post-marital residence patterns. Viewed from the perspective underpinning the present examination, it seems far more likely that the greater diversity of glass types in the Mno-446 projec- tile point assemblage relative to associated debitage is a function of disjunctions in raw material procurement episodes. That is, materials represented in the tool class (92% of which are small, spent, proximal fragments; see Bettinger [1981:27, Figure 6]) were acquired during an earlier portion of the foraging cycle, being discarded during retooling activities and replaced with points produced from more recently acquired glass. The large proportion of items made from obsidian originating to the south (67%) suggests that access to the Long Valley area was temporally proximal, which may also account for the high percentage of Casa Diablo glass in the flake assemblage. In sum, it is unnecessary to posit the existence of elaborate sociological processes to account for the disparity in point and debitage source profiles. Indeed, both artifact categories represent utilitrian products and would, ceters Daribus, be expected to move within the same "sociocultural" context. CONCLUDING OBSERVATIONS This preliminary characterization of diachronic variability in obsidian acquisition and use within central-eastem California has shown that source diversity for earlier temporal periods appears to be, in general, greater than later in the occupational sequence. It has been suggested that this reflects the existence of relatively more mobile, extensive subsistence-settle- ment systems that brought groups in proximity to a wider range of lithic resources during their annual foraging cycle. Within later time periods, there is more regular use of obsidian from nearby sources. Such a pattern is probably tied to the concomitant emergence of intensive land-use strategies (including pronounced tefritorial control) and formalized intergroup exchange relationships. Distant lithic resources were increasingly brought to consumers rather than the reverse. Further evaluation of prehistoric obsidian dispersion within central-eastem California, using enlarged samples from a greater number of archaeological contexts, will no doubt amplify and refine the patterns identified in the present study. Two fmal observations, methodological in scope, are in order. The behavioral reconstructions offered in this paper are heavily dependent on source analysis of multiple artifact classes, minimally involving both formalized tools and unmodified chipping debris. A number of previous studies have demonstrated that different artifact types often have dramatically diver- gent source profiles. In northwest California and southwest Oregon, for example, Hughes (1978; Hughes and Bettinger 1984) has shown that large blades inferred (on the basis of ethnographic counterparts) to have operated within "socioceremonial" spheres were consistently manufactured from obsidian originating at 123 Contributions of the Archaeological Research Facility Number 48, December 1989 great distance from their fina context, whereas utilitar- ian tools (i.e., typical bifaces and projectile points) derive from quarries more proximal to the subject sites. The use of "exotic" glass in the former group is thought to reflect strictures on blade coloration, as well as a greater "worth" attributed to more costly, difficult to obtain material. In this case, a sociological argument (bolstered by a specific ethnographic analog) provides a seemingly parsimonious explanation for observed discontinuities in source representation. However, in other cases where similar proposals have been made such as that outlined by Bettinger (1981; Hughes and Bettinger 1984) to account for diversity in projectile point and debitage profiles in central-eastern Califomia - recourse to such arguments is unnecessary and almost certainly misleading. Rather, source variation is more effectively viewed as a consequence of settlement and technological organization (e.g., Bamforth 1986; Basgall and McGuire 1988; Basgall, Hall, and Hilde- brandt 1988; Gramly 1980; White 1984). It is empha- sized that in neither case would it be possible infer the specific processes involved without examining multiple kinds of artifactual residue. A second observation relates to the importance of identifying the mode of lithic acquisition operative within a given archaeological context An extremely mobile, wide-ranging subsistence-settlement adaption is capable of producing relatively great geographic displacement in lithic raw materials in the absence of trade or formalized, intergroup exchange. As one means of approaching this problem, it has been argued here that exchange-related acquisition will be marked by the regularized occurrence of source ypes, and by their equitable representation in both tool and waste classes. Alternatively, incidental procurement carried out in the context of settlement relocation will be reflected by disjunction in the source profiles of tools and debitage. Distant material sources (exploited at an earlier stage of the foraging cycle) will occur primarily in the form of broken, reworked, and expended tools, whereas more proximal sources (presumably visited more recently) will be represented by more serviceable tools and manufacturing residues. REFERENCES CITED Bamforth, D.B. 1986. Technological efficiency and tool curation. American Antiquity 51:38-50. Basgall, ME. 1983. Archaeology of the Forest Service Forty site (CA-Mno-529), Mono County, California. Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, California. 1984. The archaeology of Mno-1529: A secon- dary reduction site in Mammoth Lakes, Mono County, Califoniia. Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, California. 1987. Paleoindian occupation in central-eastern Califomia: the Komodo site. Current Research in the Pleistocene 4: 50-54. n.d. Archaeology of the Komodo site, an early Holocene occupation in cental-eastem California. IN: Early Human Occupation in Western North America, edited by J.A. Willig, C.M. Aikens, and J. Fagan. Nevada State Museum Anthropological Papers. In press. Basgall, ME., M.C. Hall, and W.R. Hildebrandt 1988. The late Holocene archaeology of Drinkwater Basin, Fort Irwin, San Bernardino County, Califomia. Report on file with the U.S. Army Corps of Engineers, Los Angeles District, Los Angeles, Califomia. Basgall, ME. and K.R. McGuire 1988. The archaeol- ogy of CA-Iny-30: Prehistoric culture change in the southern Owens Valley, Califomia. Report on file with the California Department of Transportation, Sacramento. Basgall, ME., KR. McGuire, and AJ. Gilmath 1986. Archaeological test excavations at CA-INY-30: a multi-component prehistoric site near Lone Pine, Inyo County, California. Report on file with the California Department of Transportation, Sacramento. Bennyhoff, J.A. and RE. Hughes 1987. Shell bead and ornament exchange networks between California and the western Great Basin. Anthropological Papers of the American Museum of Natural History. 64 (2). Bettinger, Ri. 1977. Aboriginal human ecology in Owens Valley: prehistoric change in the Great Basin. American Antiquity 42:3-17. 1981. Archaeology of the Lee Vining site, FS- 05-05-51-219 (CA-Mno-446), Mono County, Califomia. Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, Califomia. 124 Obsidian in PrehIstorIc Cecral Rau,*Cdalfomla 1982. Aboriginal exchange and territoriality in Owens Valley, California. IN: Contexts for prehistoric exchange, edited by JE. Ericson and T.K. Earle, pp. 103-128. Academic Press, New York. 1989. The archaeology of Pinyon House, Two Eagles, and Crater Middens: three residentil sites in Owens Valley, Inyo County, California. Anthropological Papers of the American Museum of Natural History, No. 67. Bettinger, R1., M.G. Delacorte, and RJ. Jackson 1984. Visual sourcing of central eastern Califomia obsidians. IN: Obsidian studies in the Great Basin, edited by RE. Hughes. Contributions of the University of California Archaeological Research Facility 45:63-78. Bettinger, RI. and RE. Taylor 1974. Suggested revisions in archaeological sequences of the Great Basin in interior southern California. Nevada Archeological Survey Research Paper 5:1-26. Binford, L.R.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 An- thropological Research 35:255-273. Bouey, P.D. and ME. Basgall 1984. Trans-sierran exchange in prehistoric Califomia: the concept of economic articulation. IN: Obsidian studies in the Great Basin, edited by R.E. Hughes. Contri- butions of the University of California Ar- chaeological Research Facility 45:135-172. Davis, E.L. 1964. An archaeological survey of the Mono Basin and excavations at two rockshelters, Mono County, Califomia. University of California Archaeological Survey, Los Ange- les, Annual Report 6: 251-392. Los Angeles. Ericson, J.E. 1977. EgaliLarian exchange systems in California: a preliminary view. IN: Exchange Systems in Prehistory, edited by T.K. Earle and J.E. Ericson, pp. 109-126. Academic Press, New York. 1981. Exchange and production systems in Cali- fornian prehistory: the results of hydration dating and chemical characterization of obsidian sources. British Archaeological Reports, International Series 110. 1982. Production for obsidian exchange in California. IN: Contexts for Prehistoric Exchange, edited by JE. Ericson and T.K. Earle, pp. 129-148. Academic Press, New York. Ericson, J.E., TA. Hagan, and C.W. Chesterman 1976. Prehistoric obidian in Califonia, II: geologic and geogphica s. IN: Advances In Obsidian Glass Stdies: Archaeological and Geochemical Perspectives, edited by RE. Taylor, pp. 218-239. Noyes Press, Park Ridge, New Jersey. Gramly, R.M. 1980. Raw material source areas and 'curatd' tool assemblages. American Antiquity 45:823-833. Hall, M.C. 1983. Late Holocene hunter-gatherers and volcanism in the Long Valley-Mono Basin region: prehistoric change in the eastern Sierra Nevada. Unpublished Ph.D. dissertation, Department of An logy, University of California, Riverside. 1984. Obsidian, paleoeconomy, and volcanism in the eastern Sierra Nevada. Paper presented at the 19th Biennial Meeting of the Great Basin Anthropological Conference, Boise, Idaho. Hughes, RE. 1978. Aspects of prehistoric Wiyot exchange and social ranldng. Journal of California Anthropology 5: 53-66. 1981. Appendix I: Sources of obsidian at Lee Vining (CA-Mno-446), Mono County, Califomia. IN: Archaeology of the Lee Vining site, FS-05-05-51-219 (CA-Mno-446), Mono County, California, by R.L. Bettinger, pp. 78-91. Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, California. 1983. X-ray fluorescence characterization of obsidian. IN: The archaeology of Monitor Valley: 2. Gatecliff shelter, by D.H. Thomas. Anthropological Papers of the American Museum of Natural History 59: 401-408. 1984. (Editor) Obsidian studies in the Great Basin. Contributions of the University of California Archaeological Research Facility 45. 1985. Obsidian source use at Hidden Cave. IN: The archaeology of Hidden Cave, Nevada, by D.H. Thomas. Anthropological Papers of the American Museum of Natural History 61:332- 353. 1986. Diachronic variability in obsidian procure- ment patterns in northeastern California and southcentral Oregon. University of California Publications in Anthropology 17. 1988. The Coso volcanic field reexamined: im- plications for obsidian sourcing and hydration dating research. Geoarchaeology: An Interna- tional Journal 3: 253-265. 125 Contributons of the Archaeological Research Faciity Number 48, December 1989 Hughes, RE. and J.A. Bennyhoff 1986. Early trade. IN: Handbook of North American Indians, volume 11: Great Basin, edited by W.L. d'Azevedo, pp. 238-255. Smithsonian Institu- tion, Washington, D.C. Hughes, RE. and RL. Bettinger 1984. Obsidian and prhistoric sociocultural systems in California. IN: Exploring the limits: frontiers and bounda- ries in prehistory, edited by SP. De Atley and FJ. Findlow. British Archaeologkal Reports, International Series 223:153-172. Jack, R.N. 1976. Prehistoric obsidian in California, I: geochemical aspects. IN: Advances in Obsid- ian Glass Studies: Archaeological and Geo- chemical Perspectives, edited by RE. Taylor, pp. 183-217. Noyes Press, Park Ridge, New Jersey. Jackson, RJ. 1984. Obsidian hydration: Applications in the westen Great Basin. IN: Obsidian Studies in the Great Basin, edited by R.E. Hughes. Contributions of the University of California Archaeological Research Facility No. 45:173-192. 1985. An archaeological survey of the Wet, An- telope, Railroad, and Ford timber sale compart- ments in the Inyo National Forest. Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, California. 1986. Archaeological investigations at the Triple R site (CA-Mno-714). Report on file with the U.S. Forest Service, Inyo National Forest, Bishop, California. Jackson, T.L. 1974. The economics of obsidian in centl California prehistory: application of x-ray fluorescence spectrography in archaeology. Unpublished M.A. thesis, Department of Anthro- pology, San Francisco State University. 1984. A reassessment of obsidian production analyses for the Bodie Hills and Casa Diablo quarry areas. IN: Obsidian Studies in the Great Basin, edited by RE. Hughes. Contributions of the University of California Archaeological Research Facility No. 45: 117- 134. 1986. Late prehistoric obsidian exchange in centrl California. Unpublished Ph.D. disserta- tion, Depament of Anthropology, Stanford Uni- versity. Singer, C. and JE. Ericson 1977. Quarry analysis at Bodie Hills, Mono County, Califomia: a case study. IN: Exchange Systems in Prehistory, edited by T.K. Earle and J.E. Ericson, pp. 171- 190. Academic Press, New York. Steward, J.H. 1933. Ethography of the Owens Valley Paiute. University of California Publications in American Archaeology and Ethnology 33:233-350. 1938. Basin-Plateau aboriginal sociopolitial groups. Bureau of American Ethnology Bulletin 120. Smithsonian Instittion, Washing- ton, D.C. Torrence, R. 1986. Production and exchange: the prehistoric obsidian trade in the Aegean. Cambridge University Press, New York. White, G. 1984. Th archaeology of LAK-510, near Lower Lake, Lake County, California. Report on file with the California Department of Transpor- tation, Sacramento. 126