16- O O 14- 0 C 12 10 e 0 Os 0 6 a O 4 2 0 V- . ' ' ' . .. . .. .. .. .. ... 0 1, 2,000 300 4,0 7?,000 8,0 00 10, YEARS B.P. RADIOCARBON DAMES A INY-30 (n-10) 0 FORT IRWIN (n-12) 0 INY-372 (n-4) O INY-306/H AND INY-3812 (n-2) Figure 1. Pairings of Coso Hydration Values and Uncorrected Radiocarbon Dates from the Western Mojave Desert. 14 Table 4. Pairings of Coso Hydration Values and Uncorrected Radiocarbon Dates from the Western Mojave Desert. _ Hydration / Site INY-30 INY-30 SBR4504, L. 5 SBR-5365 INY-30 INY-30 INY-30 INY-30 INY-30 SBR-2659 SBR4483, L. 9 SBR-4449 SBR-4786 INY-3806/H INY-30 INY-3812 INY-30 INY-30 SBR-4786 INY-372 SBR-5367 INY-372 INY-372 INY-372 SBR4966 SBR-5250 SBR-5250 SBR4562 mean n (A) Uncorrected Radiocarbon Mean in Years B.P. 9 4 14 26 12 27 34 34 3 6 10 10 2 9 96 22 19 40 Uncorrected Radiocarbon Values in Years B.P. 2.3 2.4 4.6 4.3 2.7 3.2 3.2 2.9 4.4 6.4 5.4 5.4 5.3 4.4 5.2 5.0 5.3 5.4 7.2 7.6 7.8 7.8 8.4 8.2 12.4 15.4 14.2 12.8 NOTE: All hydration specimens were geochemically identified as 'Coso" glass using XRF analysis. Correlations frmINY-30 presented in Basgall and McGuire 1988:Table 17 (hydration sample size and stnaddeviations not presented there); SBR4562 from Basgall and Hall 1992; SBR-5250 and SBR-5367 from Hall 1992; all other SBR (San Bernardino County) pairngs from Gilreath et al. 1987; INY-372 correlations from Drews and Elston (1983), Clewlow et al. (1970), and Ericson (1978); INY-3806/H and -3812 from Delacorte and McGuire 1993. the hydration band on a piece of Co obsidian found there is thicker than the band on a specimen of the same age found in Rose Valley. This trend nts for the Fort Irwin smples shown in Figure 1 to plot higher than pairings from other areas. In an atte to bring order to chaos, Basgall (1990) examined the radiocarbon/Coso hydration rim values reported from a variety of depositional contexts. He derived a Coso hydration rate from the 10 radiocarbon/hydration value pairings obtained at the Lubkin Creek Site, and using Lee's (1969) temperature integration equation, he calculated EHT for a variety of contexts throughout the southern third of California, thus allowing different depositional settings with differing EHTs to factor that influence. The EHT correction factor for Haiwee/Coso was detemined to be 19.30 C. He concludes that much, if not most, of the variability in Coso glass hydration values documented by many archaeologists- the variability that has spawned more than 15 hydration rates (see Basgall and True 1985; Cleland 1988, 1989, 1990; Drews and Elston 1983; Ericson 1977; Friedman and 15 s.d. 0.4 1.1 1.0 0.5 0.4 0.8 1.0 0.8 0.3 0.7 2.7 1.1 0.0 0.2 3.1 1.5 1.6 2.5 180?60 270?70 275 290?i60 360 390 410?80 480?60 860 1180 1190?110 1290 1310?70 1380 1410 1470 1695 1745 2010?70 2240:?145 2240?50 2900?t80 3520?80 3580?80 7230 8240 8410?140 9440 260?60 and 290?60 330?60. and 390?90 310?70 and 470?70 760?100 and 960?100 1150?200 and 1210?70 1040?150 and 1540?70 two identical values 1160?90 and 1600?100 1220?70 and 1600?70 1340?50 and 1600?i60 1530?80 and 1860?70 1650? 100 and 1840?80 7140?290,7150?290, and7400?280 8180? 150 and 8300? 110 9410? 115, and 9470?115 Obradovich 1981; Garfinkel and McGuire 1981; Koerper et al. 1986; McGuire et al. 1982; Meighan 1978, 1981; Michels 1983; Stevenson and Scheetz 1989) - can largely be explained by the effective temperature of the environment in which the glass has been deposited. As an outcome of his review, he offers a single hydration rate for Coso glass which factors in the mean annual temperature of the area from which the archaeological sites occur. This rate and the conversion factors for temperature account for the radiocarbon/hydration pairings found at Malibu, Orange County, Fort Irwin, Owens Valley, and the Coso area. The rate he has developed is: LOG Y = (2.32 [LOG (X x a)]) + 1.50 where X is the rim measurement in microns, Y is the date in years before present, and a is the EHT correction factor. The correction factor for Haiwee/Coso is 0.8723, it being slightly warmer than the Lubkin Creek site in southern Owens Valley. In the subsequent chapters (see Chapters 5 and 8), this single rate is applied to the Coso obsidian hydration data from our investigations, and to comparative data obtained from outlying areas. LITHIC PRODUCTION AND EXCHANGE As Ericson has so aptly pointed out, "The quarry remains the logical site to begin the study of a stone-tool using culture" (1984:1). This study approaches the Volcanic Field from two points of view. The first objective is to document toolstone manufacturing patterns evident in the procurement zone. Our concern is with answering fairly basic questions: where did production occur; when did exploitation occur, i.e., what is the shape of the prtion/exloitation curve for Coso obsidian; and what kinds of items in what quantities were produced at Coso? Attempts to place the exploitation of Coso obsidian in a regional context lead us to explore answers as to why Coso obsidian was exploited (for the producer's consumption or for trade); and who was exploiting the quarry (resident populations or visitors). WHAT QUARRY DEPOSITS WERE EXPLOITED Geological processes have determined the distribution of obsidian deposits, but many factors interacted to determine the degree to which a deposit was used. Chief among them are the quality and abundance of workable stone at a deposit, the deposit's location relative to other desirable resources (e.g., water and food), the ease or difficulty of getting to the deposit (e.g., does it necessitate climbing to the top of a steep rhyolite dome, or excavating a mining pit), and the acuteness of need for stone tools. Both lag and major primary deposits occur within the Coso obsidian procurement zone. Virtually all exposures of workable quality stone were used, even small patches of secondary deposits with comparatively few cobbles. By no means, however, was exploitation so extensive that it approached exhausting the overall supply. When given ready access to both, what factors precipitated the use of lag deposits instead of the seemingly superior primary deposits? We will begin to address this issue by plotting the spatial distribution of lag and primary deposits in the Coso Volcanic Field through tme. How the exploitation of the Coso quarries articulated with a group's exploitation of other quarries in the region is also of interest. Basgall (1989) presents compelling data that early populations had direct access to many obsidian quarries, e.g., Coso, Fish Springs, Casa Diablo, Queen/Truman/Hicks, and Bodie/Pine Mountain (Map 6), among others, as a consequence of their expansive mobility pattern. Changes in mobility range or pattern would, of course, affect which quarries were visited, as well as how frequently a group would have been able to exploit a particular quarry. Factors such as territoriality and resource control, in combination with population growth, may have significantly determined what quarry locations a group could have exploited, as well as which specific deposits they actually did exploit. WHEN DID EXPLOITATION OCCUR As the first step in exploring variables that affect the structure of a lithic production system, one must necessarily determine when the resource was exploited, and assess the changing magnitude of quarrying activities throughout the past. Production curves constructed for Casa Diablo (Hall and Basgall 1994), Bodie Hills (Singer 16 & MT. HICKS OWENS LAKE ACOSO O 0 OBSIDIAN BuTTE 0 100 MILES 0 200 KILOMETERS Map 6. Major Obsidian Sources in Eastern California. 17 m V-6 . : "IPP Cl vo^ vl VO^ C* and Ericson 1977), and the Sugarloaf Quarry (Elston and Zeier 1984) are remarkably similar (Figure 2). Each shows one major period of -peak production, now commonly placed at roughly 3000-1000 B.P. corresponding with the Newberry and early Haiwee periods, and a major production decline after ca. 1000 B.P. Given the similarities of these three major obsidian quarries with respect to their geographical location (just off-set from the east front of the Sierra Nevada), their shared cultural context (being at the western edge of the Great Basin culture region and across the Sierra divide from foothill and central valley groups), the gross manufacturing activities that occurred there (see below), and their spatial relationship to one another (Map 6), it would be remarkable had their production histories been widely dissimilar. Considering how different the methods were that derived these production curves, their similarity is even more remarkable. The production curve for Casa Diablo is based on about 2500 hydration specimens mostly from secondary reduction sites located along the west edge of the procurement zone, some of which were used at various periods in the past also as habitation sites. The Bodie Hills hydration profile is built from approximately 100 hydration samples collected from across the quarry zone. The production curve for the Sugarloaf Quarry is based on about 135 hydration rim readings on specimens from four sites in or imiatly adjacent to Sugarloaf Mountain. Thus, one can rightfully be concerned with how representative each sample is of the quarrying and production history of these large quarry exploitation zones. We are not in a position to independently judge the veracity of the production curves for Casa Diablo and Bodie Hills, but we are in a position to re-evaluate the production curve for the Coso Volcanic Field, and certainly such an action is appropriate. Our interest in the spatial and temporal distribution of Coso obsidian, and the relationship between the various aspects of the full lithic exploitation pattern, prompts us to develop separate production curves for the primary quarries, the lag quarries, the numerous off-quarry obsidian reduction sites that exist within the Volcanic Field, as well as major obsidian reduction sites located well outside of the procurement zone. WHAT WAS BEING PRODUCED AND IN WHAT QUANTITIES Quarry studies have shown, ti and again, that the predominant items manufactured throughout the western United States are bifaces and cores, with the reduction techniques varying in response to time, place, and material constraints. The obsidian quarries in eastern California are no exception. Studies at and near the Casa Diablo obsidian procurement zone (Adams 1986; Basgall 1983, 1984, 1987; Bettinger 1977; Bouscaren and Wilke 1987; Goldberg et al. 1990; Hall 1983; Jackson 1985; Michels 1965; Mone 1986) consistently show biface production to have been the overwhelmingly predominant activity (Table 5). Unifacially-thinned "bifaces" have recently been identified as a common variant in Casa Diablo biface reduction strategies (Snner and Ainsworth 1990). Most Casa Diablo studies, however, have been of secondary eduction locations, not locations where workable nodules of Casa Diablo obsidian occur natally. It is worth noting that only MNO-577 listed in Table 5, is a quarry site and represents the only site in the group where cores outnumber bifaces. Singer and Ericson conclude from their study at Bodie Hills that the quarry deposit "was a single activity site devoted exclusively to the production of bifaces and blades for export" (1977:181). They recognized that "two distinct artifact forms were produced for export: (1) prismatic blades, and (2) partially finished bifaces..." (Singer and Ericson 1977:177-178). The "prsmatic blades" might well be functional equivalents to Casa Diablo acially tned bifaces and "Coso" cores as descnrbed below. They also note that biface forms shift through time from large bipointed items to medium and small straight or concave-based forms. In addition to identifying the major items of manufacture, they offer some necessarily rough calculations that help us appreciate the manitde of stone working at Bodie, estmating that 470 million pieces of debitage exist, and figung "somewhere between 4.79 and 8.62 million bifaces of all shapes and sizes were produced" (Singer and Ericson 1977:185). These figures should be viewed, of course, very cautiously. In an early overview study of quarrying activities at Sugarloaf Mountain, Elston and Zeier found that biface reduction had been the prevalent activity. Referring to all "reduction pieces" as cores, they classify 39% (n=93; Elston and Zeier 1984:104) as bifaces, 34% as irregular polyhedral or cobble/boulder/rhomboidal cores, and a remaining 23% as plano-convex cores, alternatively referred to as 'Cosow" cores and large flake blanks. "Coso" cores are primarily large flakes that 18 coso - SUGARLOAF LL n 1 n n l in i1 10 12 1'4 HYDRATION VALUE (MICRONS) HYDRATION VALUE (MICRONS) BODIE HILLS HYDRATION VALUE (MICRONS) Figure 2. Hydration Profiles for Coso (Elston and Zeier 1984; Basgall 1990), Casa Diablo (Hall and Basgall 1994; Hall 1984), and Bodie Hills (Ericson 1982) Source Areas. 19 10- 5- z FK2 -- L- 0 4 C; 1O O F] w w z 18- 12- 6 u I -s ? - - - m ? r. F LI Table 5. Flaked Stone Assemblages from Selected Casa Diablo Sites. CA-MNO -529 -561 -574 -577a -578 -833 -1529 Bifaces 35 539 49 25 100 16 176 Cores 56b 32 1 33 23 - 5 Other Obsidian Artifacts 524c 322d 14 7 67 4 97 Debitage (weight) 25,338 32,402 40,521 88,500 121,025 -3,000 (35.2 kg) NOTE: a - only site of this group where obsidian occurs naturally, qualifying it as a quarry. MNO-529 from Basgall 1983; b - Basgall offers the following regarding the number of cores: "core fragments...may largely represent shatter resulting from the reduction of large, tabular blanks, and not reflect core reduction at all" (Basgall 1983:64); c - includes 508 edge modified flakes of which 336 are from the surface. MNO-561 from Hall 1983; d - unifaces (n=224) were not differentiated from edge modified flakes. MNO-574, -577, -578, and -833 from Goldberg et al. 1990; MNO-1529 from Basgall 1984. have been rudimentarily percussion flaked from one side, in a fashion that would tend to thin the item in cross- section. They consider such items as a variant form affiliated with a biface reduction sequence, and the illustrations they provide certainly indicate that subsequent work would generate regularized, thinned bifaces. From these prior studies, one can reasonably expect that biface production was the prevalent reduction activity applied to Coso obsidian; that cores might be more prevalent than bifaces at quarry loci, with the former items virtually non-existent at off-quarry reduction locations; and that a unifacially worked variant figures into the early and middle portion of the biface reduction sequence with some frequency. The degrees to which these trends are supported by data from lag quarry deposits, primary quarry deposits, and at off-quarry sites both within and outside of the Coso obsidian procurement zone remains to be documented. Lithic production analysis has become increasingly concerned with reduction sequences used, the changing techniques applied from one manufacturing stage to the next, and the various forms an artifact assumes during its use-life. With this increasing desire to understand more fully the range of factors that account for the flaked stone tool assemblages found in sites, artifacts as well as debitage have come under greater scrutiny. The earliest studies of North American quarries have emphasized that items found at quarries are largely rejected items and unfinished pieces. Consequently, those assemblages present a quite biased impression of the quantity and kinds of items successfully produced and transported from an area. For that reason, in determining the magnitude and range of stone-tool working that occurred at a location, debitage resulting from both failed and successful reduction events is superior to the artifact assemblage as a basis for study. Flaked stone studies have shown that different stages of the manufacturing process frequently occurred in disjunct locations. It is not uncommon to discover that the kinds of items produced at a site differ from the kinds of items discarded there. Thus, in assessing the relative or absolute frequencies of the kinds of items manufactured of Coso obsidian, it will be necessary to consider how much production occurred at varying locations based on debitage types and quantities, and compare that to the number and kinds of items discarded there. An additional issue confounding attempts to determine the number of items produced or the amount of stone working that occurred, revolves around the technological shift from dart to arrow-sized points. This shift is well documented to have occurred throughout the United States by at least 1200 B.P. Whether this shift resulted in the production of more or less stone-working debris is debated. Smaller arrow-sized items obviously can be generated from smaller stardng forms, thus it is a safe assumption that all things being equal, more arrow-sized points than dart-sized points could be produced from the same amount of toolstone; put another way, the quantity of debris generated from producing 100 arrow-sized points very likely weighs substantially less than the debris generated from producing 100 dart-sized points. Ericson (1982), however, has suggested that a bow hunter needed somewhere on the magnitude of ten times as many arrows as an atlatl hunter. Holding the position that the pre-1000 B.P. production peak at the quarries is a result of eastern Sierran groups producing items for exchange with California groups west of the Sierra, he reasons that the need for more finished tools in later times, compounded by the fact that population levels were sbstantially higher after adoption of the bow-and-arrow, actually resulted in a late period decline in stone-working at eastern California obsidian quarries. 20 ... the primary producers at the sources could not meet the increasing demands for finished items from central California. To meet this increasing demand it appears that the primary producers later changed their technology from biface production to blade-flake production, to outright export of raw materials... Most likely the dispersion of the production systems away from the sources is a direct response to the limited manpower available for production under such conditions [Ericson 1982:145-146]. WHY WAS COSO OBSIDIAN EXPLOITED The sheer quantity of chipping debris and discarded items found at the major obsidian quarries in eastern California has led many to conclude that production far exceeded the needs of resident populations. The western Great Basin, excluding Owens Valley, is thought to have supported moderate to low population densities throughout prehistory. While there is debate as to what those levels were over time, the notion that population densities were higher in much of California west of the Sierran crest throughout most of prehistory meets with little opposition. The market for stone tools west of the Sierran crest is often cited as the impetus for the development of extensive exchange networks. The distribution of Coso glass in prehistoric sites quite far from the source (for example, Ericson 1977; Hughes and True 1985; Jack 1976), and the presence of shell beads originating from the Pacific Ocean or Gulf of Mexico in eastern California prehistoric sites are typically given as evidence that such a system was in operation. Coso glass, for example, has been recovered from prehistoric sites throughout the southern half of California, extending east to the Colorado River, west to the Pacific Ocean, and north to Monterey Bay. Bettinger (1982a), Bettinger and King (1971), Ericson (1977, 1982), and Singer and Ericson (1977) all assume that eastern California obsidians were traded across the mountains, and have explored the influence that regularized exchange had on the socio-economic complexity of hunter-gatherers, considering alternately sedentism, etoriality, and the like. In an attempt to explain the origin of hunter-gatherer exchange systems, Ericson (1984) proposes a model of population growth which leads to sedenism and increased territoriality, which ultimately leads to a growing need for non-local resources. These conditions are conducive to the development of large-scale exchange systems operating to move important products among different groups. The issues here revolve around how much Coso obsidian production occurred in excess of the resident population's own needs; and what mechanisms account for the distribution of Coso obsidian throughout southern California. Conceding that Coso obsidian was traded among groups, what does its distribution suggest about the structure of the exchange network: in what forms was it exchanged, in what volume was it exchanged, in what directions did it move, and was craft specialization involved? Torrence (1986) and Arnold (1987) provide the following list of archaeological indicators of production specialization: (1) very high volume of production materials; (2) identifiable workshops separated from other subsistence/residential areas; (3) distinct pattern in regional distribution, reflecting organized and controlled production and exportation; (4) high tcnological sandardization and high rates of success; (5) control over critical resources; and (6) specialists' tools in certain burials. Further, if specialized production was practiced to produce the exchange commodity, it is argued that the labor investment and the specialists must have been underwritten by others. The degree to which the lithic production patterns documented for Coso support the existence of craft specialization will be examined, after due consideration has been given to the production-for-consumption needs of resident groups. WHO HAD ACCESS TO COSO OBSIDIAN Obviously resident populations had direct access to the resources available in the Volcanic Field. Were they, however, single-handedly responsible for the vast mount of stone-working that occurred, or did non-residents also have direct access? Considering this issue for the Casa Diablo quarry, Bouey and Basgall (1984) argue that when quarry production was at its highest level, the socio-political condition of the resident population was not consistent w si specialized production and exchange. At the same period of time when production appears to have peaked, the organizational complexity and population level of people in central California (Goldberg et al. 1986; 21 King 1976; Moratto 1972; Moratto et al. 1978) are considered sufficiently developed to support a regularized trade/exchange system., Consequently, Bouey and Basgall (1984) conclude that groups from the western Sierra traveled to the quarry, where they manufactured the desired items, then transported them back over the Sierra. In this scenario, residents and non-residents were independently both producers and consumers. Bouscaren and Wilke (1987) cite the absence of western Sierran lithic materals and flaked stone tool artifact types in Long Valley, among other patterns, to argue that producers were residents, leading to the tacit conclusions that residents generated the massive amounts of chipping debris, and that Casa Diablo items obtained by western Sierra people were initially fashioned by eastern Sierran people. Goldberg et al. take a third position, suggesting that perhaps "the group using both the east and west side of the Sierra were one and the same, at least during the Middle Horizon" (1990:182- 183). From the above discussions, four main objectives can be distilled regarding the obsidian production patterns of Coso: (1) documenting the shape of the production curve and evaluating how it correlates with the production curves of other eastern California obsidian sources, as well as consumption curves outside the procurement zone; (2) documenting the pattern of obsidian acquisition, and assessing how it correlates with the acquisition patterns modeled for other eastern California obsidian sources; (3) evaluating what the production pattern for Coso obsidian suggests about population growth, territoriality, and mobility patterns; and (4) documenting the distribution of Coso obsidian, and evaluating what it suggests about mechanisms for exchange. SUBSISTENCE AND SETTLEMENT As clearly documented by the foregoing discussion, useful inferences regarding obsidian production and exchange cannot be made without considering the range of constraints associated with the larger socio-economic system. At Casa Diablo, for example, Bouey and Basgall (1984) note that many of the requirements connected with specialized obsidian production (e.g., sedentary population cotolling the source) were not entirely consistent with the settlement system reconstructed during Newberry period use of the Inyo-Mono region (i.e., high residential mobility). The following discussion suarizes subsistence-settlement pattern models offered for areas surrounding the Coso Volcanic Field, developing the socioeconomic context necessary to evaluate the changing role of Coso obsidian in southeastern California prehistory. By virtue of lying at the boundary between the Mojave Desert, Sierra Nevada, Owens Valley, and arid lands to the east (e.g., Death Valley), and the likelihood that prehistoric peoples from these localities visited Coso to some degree or another, subsistence-settlement pattern models developed for these ars are outlined below. Emphasis is placed on research conducted in the norn Mojave Desert and Owens Valley, as these areas have been the primary focus of large-scale subsistence-settlement pattern studies. Important comparative data are also provided from the western Mojave, southern Sierra Nevada, White Mountains, Death and Deep Springs valleys, and the Coso region. NORTHERN MOJAVE DESERT Until recently, models of early Holocene adaptations in the northern Mojave Desert have been largely derived from the work of Warren and his colleagues (Warren 1967, 1984, 1986; Warren and Crabtree 1986; Warren et al. 1984). Their sequence begins with the Lake Mohave Period (10,000-7000 B.P.), a temporal interval thought to have been characterized by severe climatic change. Between about 10,000-9000 B.P., Warren argues that temperatures increased, but precipitation remained relatively stable. This climatic regime filled the valleys and basins with streams, mares, and lakes, and produced vegetation communities that supported abundant populations of artiodactyls. Lake Mohave settlements, subsumed under the more general Western Pluvial Lakes Tradition (Bedwell 1973; Hester 1973), were thought to have been concentrated along lake shores, and produced artifact assemblages reflecting heavy emphasis on hunting; e.g., Lake Mohave and Silver Lake points, leaf-shaped bifaces, large ovate-domed and elongate keeled scrapers, crescents, engraving tools, and only a minor presence of milling equipment. After 9000 B.P., trends toward increased temperature and aridity intensified, reaching critical proportions between about 800-7000 B.P. Most lowland lakes and streams became dry, and vegetation communities providing food for larger game animals retracted into a few isolated contexts. In response to increasing aridity, human settlements appear to have been more restricted in space, concentrated in higher elevations where few ephemeral 22 streams and lakes still existed. Correlated with this environmental change was the emergence of the Pinto Basin complex. Rather than representing a new cultural group with a different subsistence focus, the complex was thought to have developed out of a single Lake Mohave-Pinto Basin cultural tradition, with continued emphasis on hunting large game. Support for this interpretation was based on: (1) spatio-temporal overlap of Lake Mohave and Pinto points; (2) the continued use of fine-grained igneous materials for bifacial tools, in contrast to cryptocrystalline materials so prevalent in later cultural complexes; (3) continued reliance on percussion techniques of lithic reduction, also distinctly different from later assemblages where pressure flaking is common; and (4) a dominance of morphologically and functionally similar hunting and animal processing tools (e.g., projectile points, biface "knives", and domed and keeled scrapers). By 6500-4000 B.P., during the height of the mid-Holocene climatic optimum, the number of clearly defined occupation areas decreased, and most appear at springs. In addition to this settlement location adjustment, domed and keeled scrapers are replaced by flake scrapers and there is an increase in the frequency of handstone and millingstones, the latter indicating a greater reliance on seed collecting and processing. Faunal assemblages indicate that artiodactyls became a negligible compot of the overall diet, restricted only to high elevation spring locations. Finally, toward the end of this temporal interval, sites of any kind are difficult to identify, further indicating a substantial decline in human population. The apparent continuity of a hunting focus in the face of deteriorating climatic conditions, and a corresponding reduction in artiodactyl availability, indicates to Warren (1986) that Pinto Complex people practiced an adaptive strategy that had outlived its viability. Perhaps due to strong cultural values and traditions, attempts by early inhabitants of the Mojave Desert to preserve an adaptive strategy geared toward hunting may have ultimately failed, forcing the near abandonment of the desert during the arid middle Holocene. According to Warren et al. (1984), productivity of large game increased during the Gypsum Period (4000- 1500 B.P.) due to ameliorating mid-Holocene climatic conditions. Large game hunting resumed importance while use of plant rources con ed to intensify, the latter reflected by increases in the frequency of milling equipment. The combination of increased environmental productivity and intensification of production systems resulted in population expansion, and a hypothesized shift from the family-based organization of the Pinto period, to multi- family bands. By the end of the Gypsum period, the introduction of the bow and arrow (presumably increasing the efficiency of large game procurement) combined with a period of increased aridity, required even greater reliance on plant and small animal resources. Throughout the Saratoga Springs (10-800 B.P.) and Shoshonean (800 B.P. - Historic) periods, these changes in subsistence orientation are thought to have caused a reduction in the number of large residential bases and general dispersal of the population into smaller family groups. More recent work at Fort Irwin (Map 2) by Basgall (1991) and Hall (1992) provides a contrasting view of early Holocene adaptations. Based on an extensive review of paleoenvironmental data, they question long-held assumptions about post-glacial habitats as portrayed by those who support the notion of the Western Pluvial Lakes Tradition. Rather than characterizing the late Pleistocene/early Holocene interval as a time of perennial rivers and lakes, it apes that by ca. 10,000 B.P. surface water in the north-central Mojave Desert was restricted to spring/seep discharge and episodic playa lakes of variable frequency and duration. Vegetation was also somewhat marginal, consisting of desert scrub below 1200-1500 m (39404920 ft) and juniper-shrub woodland in higher elevations. By 7000 B.P. the juniper-shrub woodland had retreated to elevations above 1700-1800 m (5580-5900 ft) while the near-modern creosote scu community was probably in place throughout the area by about 3500 B.P. (Basgall et al. 1988:33-53). Artifact assemblages from Lake Mohave contexts, like those documented by Warren (1986), include an abundance of bifaces made of fine-grained igneous materials (basalt, rhyolite, and felsite), formed flaked tools of various material types, and only minor frequencies of milling equipment. Rather than viewing bifaces and formed flaked tools as specialized implements used to process large game (butchering, skinning, and hide-scraping), Basgall (1991) argues that most of the former were probably used as cores (many were early stage forms exhibiting very little wear), while the latter apparently served a wide range of functions due to the presence of multiple wear configurations. Although artiodactyls appear to have been exploited when encountered, faunal remains obtained from a variety of sites indicate diversified diet centered on lagomorphs, rodents, and reptiles (not including tortoise). Lake Mohave sites were also found to cross-cut a wider range of non-lacustrine settings, providing further evidence for a more generalized adaptation than traditionally attributed to this interval of time. 23 Pinto/Little Lake period assemblages recovered by Basgall (1991) and Hall (1992) are broadly similar to those described by Warre~n. Flaked stone tools and debitage are roughly comparable to the preceding period, showing a continued preference for the use of fine-grained igneous materials. Faunal assemblages also remain the same, emphasizing the exploitation of small game (now including tortoise). In contrast to the Lake Mohave pattern, a sharp increase in the frequency of milling equipment and battered stone implements occurs, suggesting greater reliance on hard seeds. Site location data show continued use of a wide range of habitats, not necessarily adjacent to water. These distributions, coupled with the generalized mix of tools and faunal remains, suggested to Basgall (1991) and Hall (1992) that a relatively broad-based subsistence regime was in place throughout the early and middle Holocene. Furthermore, the presence of highly curated assemblages (i.e., well-worn tools made from exotic materials), indicated a settlement system characterized by a substantial degree of residential mobility. Gypsum/Newberny period (ca. 3300-1350 B.P.) deposits also contain highly curated assemblages, including a full complement of flaked, ground, and battered stone tools, as well as diversified accumulations of faunal remains composed largely of lagomorphs, tortoises, and artiodactyls (Basgall 1991; Basgall et al. 1988; Hall 1992). In contrast to earlier periods, emphasis on percussive production of igneous tools shifted to the almost exclusive use of cryptociystalline materials, often modified by pressure retouch - a change accompanied by major increases in the frequency of bifaces and associated debitage. Whereas earlier settlement systems appear to have reused specific locations on a regular basis, creating accumulations of debris within relatively large site areas, most Gypsum/Newberry sites are relatively small, probably indicating a wide-ranging settlement system geared to recurrent, short-term occupation of generalized areas rather than specific locations. During the Saratoga/Haiwee (ca. 1350-650 B.P.) and Shoshonean/Marana (650 B.P. - Historic) intervals, Fort Irwin data indicate that foraging radii decreased and became more regularized (Basgall 1991; Hall 1992). Evidence of this shift includes a decrease in the frequency of curated tools (local materials dominate all artifact types) and a more regular use of particular locations (sites became slightly bigger and in some cases contained patches of midden). The density of milling equipment continues to increase, taken to reflect intensified use of local plant resources. Faunal remains show a greater diversity of small game, including a higher frequency of tortoise. Production of bifaces and associated debitage decreased substantially, perhaps a response to using bow and arrow technology: because projectile points were no longer the outcome of biface reduction but made from modifying flakes, both flaked stone material needs and the production of debitage were reduced. It was also noted that decreased biface manufacture could have resulted from a reduced need to gear-up for long distance travel (see Kelly 1988). WESTERN MOJAVE DESERT In contrast to Fort Irwin, comprehensive subsistence-settlement studies have not been accomplished for the western Mojave. A recent overview by Sutton (1988a) does identify several settlement trends that have interesting implications when viewed from a larger, inter-regional perspective. Lake Mohave and Pinto/Little Lake period materials have rarely been encountered, consisting of a few sites tentatively cross-dated with projectile points and flaked stone assemblages dominated by rhyolite (Glennan 1971). After 3000 B.P. archaeological visibility increases substantially, allowing several inferences regarding settlement struture. Sutton (1988a) identifies large permanently occupied villages supported by numerous special purpose sites, the later used to exploit specific resources on a part- time basis. Located in the vicinity of Antelope Valley (Map 2), major settlements were often quite large and complex, containing deep middens, cemeteries, structures, and numerous ade items from coastal California (mostly shell beads). This system, fully in place during the Saratoga Springs/Haiwee period, was thought to reflect a transition from an earlier pattern of seasonal tanshumance to one charcterizd by a more sedentary existence made possible, to some degree, by profits obtained from administering the exchange of Coso obsidian for shell beads (Sutton 1988a:77). This general pattern is thought to have persisted until about 300 B.P. when, for reasons unknown, large villages were abandoned. A similar transition was also encountered by Sutton (1991) in Fremont Valley, just north of Antelope Valley. Prior to 1500 B.P., evidence of occupation was sporadic, limited to a few scattered projectile points and a low number of relatively wide hydration readings. Although less substantial than Antelope Valley settlements, excavations at two Saratoga Springs/Haiwee period residential bases yielded diverse assemblages of artifacts and faunal remains, as well as remnants of structures. The one site for which analysis is complete (CA-KER-221 1) 24 contained abundant milling equipment, as well as a faunal assemblage dominated by jack rabbit and, to a lesser degree, tortoise. Similar to his findings in Antelope Valley, occupations in Fremont Valley appear to have extended until around 300 B.P., at which time a shift to an unknown pattern occurred. Evidence that jack rabbit exploitation was emphasized during Saratoga Springs/Haiwee times has also been found in the El Paso Mountains (Map 2; McGuire et al. 1982). Prior to 1300 B.P., they argued, a broad-based subsistence pattern focused on hunting large ungulates was present. In contrast, components dating between 1300- 700 B.P. produced faunal assemblages dominated by jack rabbit. These data, when combined with the abundant presence of milling equipment, appear to reflect a shift to more intensive use of local plant and small game resources in an attempt to mitigate increased human population density - a construct fully consistent with the resource intensification model forwarded by Bettinger and Baumhoff (1982; see below). OWENS VALLEY Owens Valley (Map 2) has been the subject of numerous subsistence-settlement studies, the majority conducted by Bettinger (1975, 1977, 1982a, 1989). Except for a few sites, however, data from the Lake Mohave- Little Lake periods are quite sparse and have contributed little to these studies. One exception is the Lubkin Creek Site (CA-INY-30), a large multi-component site just south of Lone Pine (Map 2; Basgall and McGuire 1988). The early assemblage here includes projectile points, bifaces, a variety of formal and informal flake tools, some heavier core-related implen, but no milling equipment. This narrow range of tools, associated with faunal remains that include a relatively high proporton of artiodactyl, appear to represent short-term use of the area by people focused on hunting. One of the most remarkable aspects of the assemblage was its diversity in flaked stone materials. Obsidian from a variety of sources was identified, as well as various ciyptocrystalline, quartzitic, basaltic, and other fine-grained igneous materials. Similar to findings at Fort Irwin, these data are thought to reflect a foraging area extending hun s of kilometers north, east, and probably other undocumented directions from the site (Basgall 1989; Basgall and McGuire 1988). Although largely consistent with findings from the northern Mojave Desert, the above reconstruction encounters difficulties incorporating results of Harrington's (1957) excavations at the Stahl site (CA-INY-182, Map 2). Excavations here revealed a Little Lake period occupation area consisting of hearths, obsidian caches, graves, and a wide range of flaked and ground stone tools, in addition, perhaps, to "house floors." Most researchers, including Harrington (1957), considered the site a perment village, an assessment that conflicts with a highly mobile adaptive pattern. In a more recent evaluation of the site, however, Bettinger (1989) argues that it probably functioned as a seasonal base camp regularly occupied by people who traveled over a wide area during other times of the year. Well-defined Newbeny period occupation of the Lubkin Creek site did not occur until around 2000 B.P. Marked by burned structural remains associated with hearths, biface and bone tool caches, a variety of ground and flaked stone tools, and diverse floral and faunal remans, these materials are thought to represent residential use, but only on a seasonal basis. The non-permanent nature of these settlements is also reflected by patterns of flaked stone material acquisiton and use. While Coso obsidian is the predominant material, non-loc obsidians from Long Valley and areas further north are represented by finished tools (and comparatively little debitage), indicating prior visits to these localities. Unlike the Lake Mohave-Little Lake periods, when foraging radii seem to have encompassed numerous geographic areas, Newberty period support a more regularized settlement system, where residential bases were occupied seasonally, moving along a north-south trending as. Regional climatic patterns combined with seasonally specific ecofactual rem suggest a transhumance pattern of summer occupations in northern areas, moving to southern areas during cooler periods of the year (Basgall 1989; Basgall and McGuire 1988). Such an interpretation is largely consistent with Bettinger et al. (1984) findings at CA-INY-2146 (south of Bishop) and Bettinger's (1989) recent evaluation of Newberty period subsie -settlement patterns, where relatively mobile populations are thought to have focused on a carefully selected suite of plants, and favored the use of large mammals. For this to have been the case, Bettinger (1989) further argues that population levels must have been sufficiently low to allow groups to freely circulate between high quality resource concentrations, making it more advantageous to shift residential locations when local resources diminished, than to remain in place and exploit lower raking resources. 25 A major restructuring of these patterns appears to have developed during the Haiwee period, and continued forward until ethnographic contact (Bettinger 1989; Bettinger and Baumhoff 1982). Excavations by Bettinger (1989) at three sites near Big Pine identified a Haiwee period emergence of permanent or semi-permanent lowland villages (characterized by residential structures, bedrock milling features, extensive assemblages of flaked and ground stone tools, and a diverse set of floral and faunal remains), probably supported by upland pinyon camps, and centralized seed production stations. The retionship between these sites suggested to Bettinger that seasonal movements had become spatially confined, resulting in more intensive use of resources within progressively smaller foraging areas. Probably responding to increased population pressure, these trends toward settlement centralization and resource intensification are also reflected by data from CA-INY-30 (Basgall and McGuire 1988). Although Haiwee period use of the Lubkin Creek Site was quite marginal, resource intensification was clearly evident in Marana components by increased exploitation of riparian and lacustrine settings (e.g., shellfish, water fowl, tule seeds), and greater emphasis on pine nut exploitation. Reduced residential mobility was also indicated by decreased flaked stone mateal diversity, a more even balance between tool and debitage material types, and greater use of expedient, non- curated milling equipment. According to Bettinger (1989), this adaptive shift culminated in Owens Valley during the ethnographic period where inter-village exchange and inherited political authority (i.e., chieftainship) helped maintain an increasingly rigorous degree of territoriality. SOUTHERN SIERRA NEVADA AND WHITE MOUNTAINS Excavations in the southern Sierra Nevada (Garfinkel and McGuire 1980; Garfinkel et al. 1984; McGuire 1981), west of the Coso Volcanic Field, have produced results consistent with the resource intensification model proposed by Bettinger and others. Prior to 3500 B.P. (Little Lake period), evidence of human occupation in the southern Sierran uplands (above about 1830 m [6000 ft]) is restricted to a sporadic scatter of projectile points and the infrequent wide obsidian hydration reading. During the Newberry period, use increased to a limited degree, as indicated by the presence of biface reduction locations, hunting camps, and short-term occupations that may have been associated withe pinyon exploitation. Late in time (post 1400 B.P.), however, use of the uplands intensified. Temporary occupation camps as well as large base camps, with their associated subsistence and habitation activities, appear during this time... The increased amount of vegetal processing equipment associated with these site types affirms the primacy of plant resources, of which pinyon was undoubtedly the most important [McGuire 1981:57]. High in the White Mountains northeast of Coso (Map 2), Bettinger also encountered late period resource intensification in upland habitats (3050-3810 m [10,000-12,500 ft]). Prior to 1400 B.P., White Mountain uplands are thought to have been used on a relatively short-term basis for hunting (probably bighorn sheep), reflected by sites with hunting blinds, sparse lithic scatters, and debris/tools associated with hunting and butchering animals. Sites after 1400 B.P. contain remains of well-built houses (with circular, multiple-course footings), diverse assemblages of plant and animal procurement and processing tools, as well as debris associated with repair and production of these tools. This dramatic shift in alpine land use appears to have been a response to regional population growth that decreased rates of return for lowland subsistence activities to the point where it became cost-effective to use alpine plants and other costly resources (e.g., pinyon, small seeds) previously used casually or ignored altogether [Bettinger 1991:675]. DEATH AND DEEP SPRINGS VALLEYS Interpretive summaries of Death Valley (Map 2) subsistence-settlement systems by A. Hunt (1960), C. Hunt (1975), and Wallace (1958, 1977) elucidate trends broadly similar to those found elsewhere in eastern California. 26 Death Valley I (Lake Mohave) assemblages, thought to date between 9000-7000 B.P., include Lake Mohave and Silver Lake projectile points, knives, scrapers, and choppers, but no milling equipment. Most of these sites are located in the lowlands on gravel benches next to active as well as extinct springs, and they seem to represent ea. nal, though recurrent, occupancy by mobile people focused on hunting large game (Wallase 1958:11). This was followed by a substantial hiatus - a 2000 year period attributed to the mid-Holocene climatic optimum. The subsequent Death Valley II/Mesquite Flat Complex dates between 5000-2000 B.P. and has two phases. The earliest phase includes Pinto and Gypsum series points found at small sites that appear to have been temporary camps. Other tools include knives, choppers, scraper-planes, and drills. The later phase is represented by several large occupation sites, many on the edge of dry washes in lowland setti ngs. Elko series projectile points predominate, and are accompanied by expanding-base drills, hanmerstones, choppers, scraper-planes, mortars and pestles, and an abundance of large flakes and cores - the latter reflecting an extensive stone working technology. Hunig is considered a pmary component of the economy given the bundance of projectile points, knives, and other implements presumably used to slaughter and prepare large game. The presence of mortars and pestles is considered to reflect increased reliance on plant foods, while settlement size implies a fairly large, semi-sedentary population. Wallace (1958) viewed these sites as seasonal gathering areas, from which smaller groups would split- off during other times of the year to hunt and forage elsewhere. Specialized hunting sites found in the uplands potentially represent hunting parties originating from the lowland residential bases. Similar to Warren's work in the northern Mojave, the expansion of population was largely attributed to increased environmental productivity stemig from neoglacial events that took place during this interval (Hunt 1975; Wallace 1958). Death Valley III (Saratoga Springs) dates between 2000-1000 B.P. and includes Eastgate and Rose Spring points as signature artifacts, a diverse mix of flakred and battered stone tools (e.g., knives, scrapers, drills, hammerstones, choppers), high frequencies of milling equipment (largely manos and metates), and a significant presence of shell beads, bone implements, and pottery sherds of Puebloan origin (Warren 1984). Perhaps in response to icing adity, populations appear to have occupied a wider range of habitats, focusing on the use of plant resources more than during previous intervals (Hunt 1975; Wallace 1:977). Death Valley IV (Desert Shoshone) is characterized by "many small and a few rather large settlements... Their living sites are amongst the mesquite-covered dunes, usually some distance from springs or other water sources. Although these were seasonal camps, many appear to have been inhabited recurrently" (Wallace 1958:14). Mesquite camps include fire-affected rock, charcoal, a variety of flaked stone tools (including Cottonwood and Desert Side-notched projectile points), Owens Valley Brown Ware pottery, twined and coiled basketry, Olivella beads, abnmdant milling stones and handstones, as well as a significant presence of long, cylindrical stone pestles, and wooden mortars from mesquite logs. Upland seaal camps have also been encountered, many including "storage pits, charcoal filled roasting pits, bedrock milling places and gathering crooks -- all connected with pinyon harvesting" (Wallace 1958:15). This general pattern of seasoal movements from the valley floor up to the mountain slopes appears to have continued into the ethnographic period (Warren 1984:371). Delacorte's (1990) work in Deep Springs Valley reveals parallel patterns to those of Death Valley, 125 km southeast (Map 2). Newberry period stement systems i seasonal base camps in the lowlands (characterized by a relatively high density and diversity of ground, battered, and flaked stone tools), as well as short-term seed processing areas (milling equipent co ntrations with a lower frequency of flaked stone implements) and hunting camps largely flaked stone tools and debitage scatters). Similar to Owens Valley reconstructions, the presence of exotic lithic materials are taken to reflect a system with a high degree of residential mobility, with groups using widely spaced base camps, and occupying areas well beyond the bounds of Deep Springs Valley. These base camps served as residential hubs from which groups moved to specialized logistical field camps to collect a narrow range of plants and to hunt large game species. This expansive, narrow-spectrum strategy implies little compettion between groups (Bettinger and Baumoff 1982), allowing free access to critical resources [Delacorte 1990:247]. 27 During the Haiwee and Marana periods in Deep Springs Valley, a reorganization of the subsistence- settlement system occurred: Lowland base camps and temporary seed camps continued to be used, while temporary camps used for hunting decreased in importance. Of greater significane was the appearance of three new settlement types: pinyon camps (similar to lowland occupation sites but having a higher frequency of structures), pinyon caches (largely structures), and alpine (summer) occupation sites (diversified assemblage scatters lacking associated structures). The addition of these settlement types indicated that seasonal movements had become spatially confined, increasing the need to exploit a broader range of lower ranked plant and animal species available from a wider range of environments. COSO REGION The most comprehensive subsistence-settlement studies of the Coso region to date have been completed by Whitley et al. (1987) and Delacorte (1988). Focusing largely on late Holocene adaptations, Whitley et al. (1987) offer interpretations that contrast significantly with those outlined above. During the Newberry period, settlement orgnition included major villages along pluvial Owens River... and the use of various upland environments, as evidenced by multi-function and specialized purpose sites" (Whitley et al. 1987:20). Subsistence pursuits were thought to have been quite general, exploiting a wide range of environments. This adaptive strategy was thought to have persisted into the Haiwee period, differing from the Newberry period only by increases in population as evidenced by the establishment of major villages at a variety of new locations. The Marana period, in contrast, was thought to be marked by a great reduction in the number of sites and the range of environmental zones exploited/occupied. Although changes in subsistence-settlement structure were considered a possible explanation (i.e., land-use shifts reducing archaeological visibility), Whitley et al. (1987:22) hypothesize that "the change was principally one of decreasing population, and that this decrease was in fact very substantial." Delacorte's (1988) review of an expanded data base produced a different set of interpretations for Coso. Represented by a series of limited habitation sites (including milling equipment, cobble tools, bifaces, and points), the lake Mohave period was characterized as a highly mobile settlement system focused on the exploitation of plants and animals in predominantly lowland settings. Little Lake habitation areas, particularly those in riparian settings (e.g., the Stahl site), were thought to have been occupied for longer periods. Similar to characterizations of Newberry period adaptations in Owens Valley (see Basgall and McGuire 1988; Bettinger et al. 1984), Delacorte argues that logistically organized upland hunting cams were used in support of lowland residential bases. Newberry period settet organization was considered similar to the preceding Little Lake system (Delacorte 1988:25-26): "... during the Newberry period groups traveled between a series of lowland occupation sites from which many resources were exploited through use of specialized temporary camps and limited habitation sites." The Haiwee settlement pattern was thought to differ in several fundamental ways from the Newberry pattern. First, occupation sites appear to have moved away from lowland riparian habitats to the desert scrub communities along valley flanks. Accompanying this shift was a decrease in the number of temporary camps (e.g., logistically organized hunting sites) and, most importantly, the inception of pinyon camps in the uplands. The overall stratgy was one of reduced logistical travel, but relatively frequent residential moves over short distances within increasingly smaller territories - a strategy made possible by the exploitation of a wider range of subsistence resources (see Bettinger and Baumhoff 1982). This same basic pattern is thought to have continued into the Marana period, however, the frequency of residential moves appears to have increased later, creating more smaller sites with reduced archaeological visibility. According to Delacorte (1988:30) "The difference between early (pre-Haiwee] and late [Haiwee-Marana] settlement paterns reflects, therefore, a change in the organiztion of aboriginal land-use from large, logistically oriented groups to small, residentially mobile groups, not a decrease in population." DISCUSSION The foregoing review indicates that subsise-settlement pattern strategies rem ed relatively consistent from one area to the next during much of the early Holocene. Begnin in the mid-Holocene, however, interregional variability appears to have progressively increased. As the following discussion seeks to demonstrate, the uniformity of early adaptations is largely due to a subsistence pattern focused on the exploitation of a narrow 28 range of highly ranked resources irrespective of environmental circumstance. Later populations, in contrast, intensified the use of local environments, exploiting a wider range of lower ranked resources, including those regionally restricted in distribution. It follows, therefore, that inter-regional environmental differences had minimal effect on early tool assemblages, while later such differences were amplified, creating greater variability in the archaeological record. Lake Mohave sites reflect a high degree of residential mobility (indicated by significant tool-debitage disjunctions in the frequency of exotic versus local material types [Basgall 1989, 1991]), minimal use of seed resources (based on the near-absence of milling equipment), and a probable emphasis on hunting large and small game animals (the latter particularly prevalent in the more arid parts of the Mojave Desert). With the possible exception of the Stahl site, Pinto/Little Lake period assemblages are quite similar between areas with respect to patterns of flaked stone material acquisition and use, mobility, and hunting adaptations, but differ from earlier assemblages by virtue of showing an increase in the frequency of milling equipment - a shift probably reflecting a broadening subsistence base in response to declining environmental conditions (Warren 1986). The Stahl site differs from most other Pinto/Little Lake period manifestations by the presence of hearths, graves, and perhaps residential structures, and a diverse assemblage of flaked and ground stone tools, all attributes consistent with a long- term residential base. Rather than reflecting a permanent residential base, however, it seems more likely that mobile populations may have occupied the unique environmental setting of Little Lake more frequently and for longer periods of time than was the case in most other localities (Bettinger 1989). During the Gypsum/Newberty period, the settlement system remained mobile but appears to have been less expansive and more regularized. This pattern is well-documented in Owens and Deep Springs valleys (Basgall 1991; Basgall and McGuire 1988; Bettinger 1989; Bettinger et al. 1984; Delacorte 1990), and may also have been the case for Death Valley, but it is difficult to determine with the data at hand (Wallace 1958). In contrast to the central Mojave Desert, where numerous short-term occupations were used to exploit a variety of habitat zones, the Owens- Deep Springs areas evidence greater residential stability. In Owens Valley, for example, groups appear to have moved along a north-south axis, establishing a series of seasonal residential bases, probably occupying Long Valley in summer and southern Owens and Rose valleys in winter. These sites appear to have been occupied and re- occupied for substantial periods of tie judging from the presence of structures, features, and a variety of resources obtained from vertically distant habitat zones (e.g., pinyon, mounta sheep, and marmots found in lowland sites such as Lubkin Creek). This latter phenomena probably reflects exploitation of upland areas by task-specific groups, a stratgy nsistent with archaeological data from the southern Sierra (McGuire 1981), White Mountains (Bettinger 1991), and the uplands surrounding Death and Deep Springs valleys (Delacorte 1990; Wallace 1958). Saratoga Springs/Haiwee and Shoshonean/Marana period adaptations in Owens Valley are characterized by nreased sedentsm, territoriality, and socio-political complexity, developments made possible by more intensive exploitation of local resources within progressively smaller units of land, as well as more extensive use of upland areas used previously for hunting (Bettinger 1989, 1991; Bettinger and Baumhoff 1982; McGuire 1981). In addition to these subsistence shifts, Bettinger (1982a) and Bettinger and King (1971) argue that the exchange of Fish Springs obsidian was also a primaiy contributor to the origin of this system, allowing those in control of the source to trade for subsist resources previously acquired hrugh some form of settlement mobility. Bettinger (1989), however, suggests that other factors, such as increasing population, may also have contributed to the development of this pattern. Although territorial ranges after the Newberry period appear to have been reduced in all areas considered, permanent or semi-permanent residential bases seem to have emerged only in the far western Mojave (e.g., Antelope Valley [Sutton 1988aJ) and perps in Owens Valley (Basgall and McGuire 1988; Bettinger 1989). The more and lands of Deep Springs and Death valleys, as well as much of the Mojave, required numerous short-term residential base camps to exploit a more dispersed set of environmental zones throughout the Haiwee and Marana periods. Continuation of these inter-regional differences are also found in the ethnographic record. Summaries of subsistence-settlement information collected from the Owens Valley Paiute and Coso Shoshone (Delacorte 1988; Delacorte and McGuire 1993; see also Bettinger 1982b) document several organizational differences between the two groups. Rather than residing in large semi-permanent villages like the Owens Valley Paiute, inhabitants of the Coso region were organized into independent family groups who moved frequently from one resource tract to 29 another. In early spring, when stored resources were exhausted, families moved to well-watered areas such as Haiwee Springs to collect greens. This was also a time when larger groups would occasionally form to conduct rabbit and antelope drives. By May, many families traveled to Owens Lake to gather brine-fly pupae, but soon thereafter dispersed to gather seeds. Seed gathering was initiated in the lowlands, and progressively moved up in elevation during the summer. The Coso Range was a favored locality; however, more distant upland areas were also used if necessary. Trips could also be made to Saline or Panamint valleys to collect mesquite in years when local seed production was below nonnal. By late summer groups often concentrated in upland areas to gather and store late ripening seeds and prepare for the fall pinyon harvest. The Coso Range was a primary pinyon collection area, but when local crops failed, families would go to the Inyo and Panamint ranges to collect and store the nuts. After fall harvest and continuing through winter, several families would aggregate in favored lowland settings (e.g., Olancha, Coso Hot Springs, Little Lake) or, in years when pinyon crops were exceptionally productive, smaller groups might choose to winter in the Coso Range closer to their caches of nuts. Alternatively, when local crops were insufficient, families could gather in neighboring uplands, and rely on stored foods and locally available game. As the foregoing discussions of archaeological and ethnographd itoic data indicate, the Coso Volcanic Field comprised only a small component of several larger subsistence-settlement systems. Nevertheless, by virtue of lying within an environment rich in lithic material, but relatively marginal with respect to subsistence resources, the area provides an excellent opportunity to measure the relative importance of obsidian across a variety of adaptive modes and historic circumstances. 30 CHAPTER 3: FIELD AND ANALYTICAL METHODS Extensive excavation was undertaken at 34 sites (Map 3; Table 6). Each locality was systematically surface collected and subjected to a phased excavation strategy. The purpose of this approach was to develop a preliminary evaluation of site/locus structure, followed by extensive excavation focusing on areas identified as potential single component deposits. Potential single component areas were identified in the field as spatially discrete concentrations of material, and typically were confined to areas less than 30-40 m in diameter. This method directed detailed investigations to artifact/debitage concentrations and features; diffuse scatters were less thoroughly sampled. FIELD METHODS Two basic field strategies were applied, one in areas with subsurface deposits (usually off-quarry sites), the other where cultural materials were essentially restricted to the surface (usually lag quarries). If subsurface deposits were suspected, fieldwork began by establishing a 15 x 15 m grid, followed by surface collecting all formed tools from the site (e.g., flaked, ground, and battered stone tools, but not debitage), provenienced relative to 3 x 3 m cells. Subsurface distributions were then evaluated by excavating 1 x 1 m selective recovery units usually in the southwest comer of 15 x 15 m quads. Selective recovery units were dug in 10 cm levels, deposits were processed through 1/4" mesh, and all cultural material other than debitage was retained. Debitage densities were monitored by Controlled Volume Samples (CVS) consisting of approximately 10% of a 10 cm level (1 x 1 m unit - 1/4" mesh). This approach was designed to maximize recovery of information related to site structure, in order to decide the placement of subsequent excavation units. Based on these data and surface distributions, control excavation units (CUs) were placed where artifact and debitage concentrations appeared to be present. Control units were also excavated in 10 cm levels with spoils processed through 1/4" or 1/8" mesh. CU-1/4" units were usually 1 x 2 m in size, while CU-1/8" units were typically 1 x 1 m or 0.5 x 1 m. Where cultural material was largely restricted to near-surface contexts, 3 x 3 m surface scrapes were excavated to a depth of 10 cm. Surface scrape matrices were processed using regular combinations of 1/4" and 1/8" mesh. At lag quarries, because subsurface deposits were usually lacking, a variety of surface collection techniques were used. When such an area was of a manageable size and artifact densities were low, surface collection procedures were the same as those described above, with the additional collection of debitage samples from a 3 x 3 m cell at the corner of each 15 x 15 m quad. When an area was very extensive and artifact densities were very low, debitage and artifacts alike were collected from 3 x 3 m cells placed at constant intervals along widely spaced transects. Interval widths varied from one site to the next in response to the size of the area, the steepness of the terrain, and the perceived density of materials. For areas less than 150 x 250 m (e.g., INY-1816 and INY-1824 Main), sample cells were at 15 m intervals. For even more extensive areas, which usually had even more sparse distributions, 30 m or wider intervals were used (e.g., INY-3200 and INY-3300) and transects were oriented at a right angle to the landform. Temporally diagnostic artifacts or other particularly noteworthy artifacts found outside the sample of surface collection cells were individually collected. When a concentration of materials occurred within an otherwise extensive but sparse quarry, a 15 x 15 m grid was established overlaying the concentration, and all formed tools within the grid were collected (again relative to 3 x 3 m cells). Based on documented surface distributions, areas of further interest were test excavated with 1 x 1 m selective recovery units to verify that subsurface deposits were lacking. Additional excavation was limited to 3 x 3 m surface scrapes overlaying the densest artifact/debitage clusters. Backhoe trenches were also used on occasion to discover features or significant concentrations of artifacts below ca. 40 cm. As field work continued, it became apparent that subsurface deposits were largely restricted to the upper 50 cm (and rarely extended to 80 cm) and correlated with surface distributions. Backhoe trenches were cut only at INY-4267 and INY-4329, they proved to be of limited utility, and further use was discontinued. Upon completing excavation, profiles were drawn of representative exposures, and soil samples were collected so that color, consistency, and other characteristics of depositional strata could be examined under laboratory conditions. Grid stakes, fencing, flagging, etc., were then retrieved, and units backfilled to original contours. 31 0'q Ii m o R v 4 'c 0 m R oo11 t -- W% * W0 - 'N e4 m- U v m N t< 'C GO 0' * W * N '--N 0-- %O- -~ N 'C 0o^ -'G~ '''~~ * 0 * ~ * *- , 0 N i 0 0 i ! i i -, - 'N O' >- U)o @ @ @ @ IiI I- R A00 (.j N t, % . ' o 'C 9 U) m m I , N i m 0't N m - s- N UC '''''- 0 'C ' 'C'C ~ ~N ,N, - , . , .- U)- *'*-- '.' N . . * - * , , ' , , , Nl - - -U ' '' '' II vo I .., - -) I I . i I I I . . I I . I , I i - .I - I * I , . . G I I U . - . I I I I N I- I . * I . - o " I. - I , 0 - . I e . ge) e4 oo r- N 0q .d N 0' .C oo 'C 'C N1 'C a I , , ? ~. . U* * o oo 8 8 ,, ,. , .j .d d -N NO e4 ,~ ,- . , ' ' 0U) i , - ~- ~.00 . ' ' 22 J ~ 'l . . - -fS- ~.;. ,,o Fo nI 4:v 1 1 DZ: it l 32 - C I-I . I . , I I , . . - I - . fC4 . 4-& .- 8 CI) ca . 4W0 D I- "a a9) Co o o q 0 0 C %0 JRO '4 C4 In . S Ul N S 8C 4r. la I i I I I I A la p I A I ,..; I .2 - 02 I = 0 1:2 I I k I I II 4 I II a q Table 6. Materials Recovered from Project Sites (continued). 4244 4246 4252 4267 4319 - - - 3 1 98 26 25 301 109 5 2 - 33 157 5 2 - 14 14 16 9 2 45 15 - -6 - 2 - - S - 6 1 - 13 15 7 - 1 26 24 139 40 28 446 335 4320 4322/H 4324/H 4325 4327 4328 4329 4330 4331 15 25 6 15 1 3 2 67 13 29 7 49 90 93 5 1S 13 216 1 - 1 33 8 70 7 19 73 4 - 4 5 2 11 - 1 1 3 - 1 2 6 13 2 - 11 57 36 185 4 1 - 446 52 18 306 8 15 1 - 58 - - 225 4 13 8 - 2 7 2 1 50 1 4 56 2 1 1161 70 54 21984 2083 1321 25074 448 474 1229 12195 - 3 95 173 1 - - 12 Ground nd Battered Stome Milliptone - Handstone 1 Misc. Ground Stone - Mortar Pestle Bedrock Mortar - Bedrock Slick Non-obsidian Cobble-Core - SUBTOTAL 1 Abrader Anvil - Bead Crystal Historic - Incised Stone Misellnou Shaped Ston - Miseelbums Stone Modified Stone Palett Sherd - SUBTOTAL 2 I 2 1 56 5 1 18 - - 10 - 5 - I - 2 - 14 25* 2 106 30 I 3 4 o 3 3 4 4 I 1 771 400 610 33715 3314 465 3325 185300 7 3 7 251 36 1 - 1033 5 5 3 13 2 2 3 6 9 I I I 2 6 6 25 6 2 108 141 S 3 41 49 3 1 1 4 9 I I 3 3 387 179 73 6 1 1 S 311 963 4 7 250 1 6 1 19 23 1 1 41 354 Artifact Total orwa=s Bone Chmu Excatlm Sismaris On cubic metms) 140 42 30 558 366 71 52 221 70 38 194 1351 80 57 159 7549 2 1 - 20 - - - - - - - 21 3 - - 11 26 - - - - - - - 1 8 - - 315 - - 47 Selective Recovery 1/4' - 0.80 1.90 19.60 6.20 Contol Unit 1/4' 10.00 3.70 3.50 18.40 1.60 Control Unit 1/8" 0.60 0.63 0.70 0.90 0.80 10.60 5.13 6.10 38.90 8.60 2.10 0.60 0.15 2.85 0.70 2.50 0.30 2.95 2.85 0.70 1.40 - 4.90 45.70 11.50 1.50 1.00 3.60 1.90 35.60 14.00 1.20 2.76 0.40 0.30 1.55 0.85 0.15 3.50 6.50 5.16 4.00 7.10 82.85 26.35 2.85 NOTE: * - includes one obsidian core tool. ANALYTICAL METHODS All cultural was washed and catalogued during fieldwork at a temporary lab at Coso Junction, or soon thereafter at Far Western facilities in Davis, California. Accession numbers were obtained from the Environmental Branch, China Lake Naval Air Weapons Station, and any materials previously collected from project sites were loaned to us for re-analysis and incorporation in this study. Analysis during cataloguing was limited to basic artifact description, saving more sophisticated observations for individuals in charge of particular artifact classes. As the first step of analysis, catalogued artifacts were individually reviewed by the authors (flaked stone by Gilreath, ground and battered stone by Hildebrandt) to verify type and condition ascriptions. Reassignments 33 Site (INY-) Faked Stoe Biface/Point Biface Core Drill Flank Blank Flake Tool Formed Flake Tool Projece Point Uniface Uniface-B SUBTOTAL 4378 TOTAL Debt Obsdian Non-obsidian 27 122 2 2 1 3 157 40 2912 1856 2 167 649 35 60 208 303 6232 TOTAL 0.70 194.65 2.90 212.00 0.40 28.71 4.00 435.36 1 1 I I I I based on this full review of the collection were incorporated in the master catalogues. Detailed analyses were then implemented. All recovered materials are now curated at the Naval Air Weapons Station. Field notes, field maps, and project photographs accompany the collection. Master catalogues and analysis files exist in both printed versions and as dBASE IV files on computer diskettes. FLAKED STONE ANALYSIS Our interest in documenting the flaked stone production history for the Coso Volcanic Field resulted in an analysis structured to emphasize recognition of manufacturing activity, intensity, and technology; and artifact morphology in the context of reduction trajectories. Functional analysis was kept to a mimum, in response to the low number of flaked stone "tools," relative to the extraordinary quantities of cores, blanks, and preforms. For descriptive purposes, artifact size, shape, and condition were stressed. The raw material itself sets upper limits on attainable artifact sizes, may influence the shape of some artifact types, and sometimes is responsible for the broken condition of a specimen (e.g., flaws in the stone, not poor execution on the part of the craftsperson might have caused an item to break). Prehistoric flintknappers' habits, techniques, and objectives are, however, more compelling forces responsible for shaping the artifacts, and creating the archaeological record at Coso. The following artifact categories were used for flaked stone at the catalogue level: projectile points, biface/points, drills, bifaces, uniface-B's (items reduced like a biface but with modification largely restricted to one face), flake blanks, formed flake tools, simple flake tools, cores, and unifaces. In addition to classifying each artifact by material (obsidian, cryptocrystalline, basalt, etc.), basic measurements were recorded (length, width, thickness, weight) for all analyzed specimens, as well as the condition or portion of the item represented. All projectile points and biface/points were analyzed; analysis of all other flaked artifacts was largely restricted to those occurring in temporally meaningful contexts. Projectile points are bifacial implements fashioned by pressure flaking and having hafting elements present, denoted by notches, shoulders, and/or a stem. Most can be readily classified according to Thomas' (1981) Great Basin projectile point key; the remaing are large, shouldered forms considered Great Basin Stemmed and various leaf-shaped items. A total of 60 projectile points was recovered (Table 6; see Plates 1-9 in Chapter 5); 112 additional points found during survey in the Volcanic Field supplement the sample. Attributes recorded for all 172 projectile points are those used in Thomas' key. Non-diagnostic pieces of projectile points were classified as biface/points, of which 40 were recovered (Table 6). In addition to recording the same attributes as those for bifaces, biface/point fragments were subjectively identified as deriving from arrow or dart-sized points, based primarily on size. Only two of the nearly 6,250 flaked stone items were classified as drills. Both exhibit elongate, bi-convex bits fashioned by bifacially pressure flaking converging edges. Bifaces exhibit percussion and/or pressure flake removal scars along opposing sides of a continuous margin. Nearly 3000 bifaces were recovered (n=2912, Table 6). Two other artifact categories related to biface production - uniface-B's (n=303) and flake blanks (n= 167) - are considered biface-equivalents. The former are typically large flake-based items that were percussion flaked, removing thinning flakes from a single surface. Minor modification sometimes occurs on the original ventral surface, but is confined to a narrow zone around the perimeter, and related to edge and platform preparation. Flake blanks are complete or near complete flakes which were edge trimmed by a series of short percussion flake removals, usually making a triangular form with rounded basal comers. Planar and cross-section views indicate that bifaces and these two variants were systematically rather than haphazardly formed. The analytical utility of differentiating reduction stages is well established by such works as Callahan (1979), Collins (1975), Crabtree (1973), Frison and Bradley (1980), Muto (1971), Newcomer (1971), and Sharrock (1966). Bifaces, uniface-B's and flake blanks received comparable analysis, sing wth classification into five stages based on size, reduction technique, appearance, and quality. Stage 1 items are typically large, very thick items with percussion worked edges rendering an irregular planar form, and extremely sinuous and often 34 discinuous worked margin. Approximately 60% of the item's margin has been bifacially worked, suggesting its affiliation with a biface reduction trajectory, rather than a flake-production core trajectory. The distinction between Stage 1 bifaces and bifacial cores is admittedly often arbitrary. Modification on Stage 2 items, a term used irchangeably here with bifacial blanks, is also restricted to percussion flaking, but they have less sinuous margins and slightly thinner cross-sections than Stage ls. At least 80% of the margin has been bifacially modified (for uniface-B's at least 80% of the margin has been unifacially worked), with flake removals having served to effectively thin the item (Figures 3 and 4c). Stage 3 items, also referred to as preforms, are thinner and display more regular, less sinuous, margins than Stage 2s. Evenly spaced, systematically removed percussion flakes have rendered a symmetric, uniformly bi-convex cross-section on the bifaces (Figure 4a-b); plano-convex shapes on uniface-B's; and slightly concave-convex cross-sections on the edge-trimmed flake blanks (Figure 5). Stage 4 items are small, irittently pressure flaked bifaces, representing either fragmented projectile points or items which broke early during tool finishing activities. Finally, Stage 5 bifaces are extensively pr flaked ragents of final tool forms such as projectile points and bifacial knives. Margins are heavily flaked, and entire opposing surfaces are completely covered by closely spaced pressure flake removal scars. In addition to classifying bifaces to stage of reduction and condition, a single edge-angle mauent was taken, plana and cross-sectonal shape were recorded, and the reason for discard/rejection was assessed. As previously mentioned, high technological standardization and high rates of success have been listed as indicators of production specialization. A high number of bifaces standard in form and shape, combined with manufacturing failure patterns could provide evidence that such specialization occurred at Coso. For the most part, uniface-B's and flake blanks will not be distinguished from true bifaces in discussions hat follow, in order to minimize confusion and unnecessarily protracted assemblage characterizations. For this study, a core is a mass of stone minimally exhibiting two flake removals. At least a portion of the margin from which flakes were removed remains intact, so that direction of removals can be evaluated. This definition served to eliminate the amorphous ore-like" pieces of shatter produced during initial reduction and the natally f ed chunks of glass common in the Volcanic Field. A total of 1856 cores was recovered (Table 6); they were classified as unidirectional, bidirectional, bifacial (Figure 6a), non-p(Figure 6b, d), or cobble/chunk tests (Figure 6c), ding on the orientation of flake removals and worked margins relative to one another. Other attributes recorded include: the basic form of the mas reduced (tabular cobble, globular cobble, angular cobble, flake, piece of shatter), type of cortex evident on the piece, and planar and cross-sectional shape of the object. Simple flake tools (n=649) are flakes that show minima amounts of deliberate flaking and/or damage extending along a continuous segment of the perimeter. Edge modification occurs nearly exclusively as microchipping, most commnly uicial age, and rarely intrudes more than two or three millimeters from the a . As a result, edge damage has little chad the size and shape of the orginally detached flake (Figure 7a). Analysis iluded mber of edges modified, shape of the modified edge (straight, convex, concave, etc.), length of the mdified portion of an edge, type of modification (unifacial/bifacial microchipping, edge grinding), as well as spine-plane and working edge anges. Formed flake tools (n=35) are flake-based tools where edge modification was extensive and deliberate, resulting in reshaping of the originally detached flake. Recorded attributes parallel those for simple flake tools. Unifaces (n=208) display steep, unidirectional, percussion flake removal scars underlaying a 2-5 mm zone of extensive edge modification (Figure 7b-d). Edge modification, apparently from use, takes the form of non- invasive, percussive edge trimming flakes and/or mi Hipp . They differ from unidirectional cores in that they tend to display fairly regularized planar shapes, keeled cross-sections, and inentional edge age; they differ from uniface-B's in that percussion flake removals did not effectively thin the piece, rather they created a steep platform/face juncture. Flake stone debris lacking post-detachment modification (i.e., no subsequent edge damage or flake removals) was considered debitage. Excavation and surface collection rendered over 185,000 pieces of obsidian debitage, compared to about 1,000 pieces of non-obsidian chipping debris (Table 6). Preliminary review of the collections revealed two major manu ring trajectories, both largely confined to percussion flaking: biface production and core production. Consequently, flake types developed for debitage classification sought to monitor 35 a b c 0 Figure 3. Obsidian Stage 2 Bifaces from INY-3456 (a, b) and INY-4252 (c, d). (987-44-216); c. Base (987-93-96); d. End (987-93-68). d 5 cm . Whole (987-44-104); b. Whole 36 a b J 0 5cm Figure 4. Obsidian Stage 3 Bifaces (a, b) and Stage 2 Uniface-B (c) from INY-4252. a. Complete (987-93-17/89); b. Base (987-93-71); c. End (987-93-7). 37 a b c 0 5cm Figure 5. Complete Obsidian Stage 3 Flake Blanks from INY-4239. a. 987-38-93; b. 987-38- 1; c. 987-38-172. the relative importance of these two basic strategies. The initial separation of biface from core/flake percussion reduction debris relied on complexity of the striking platform, platform to dorsal surface angle, complexity and configuration of dorsal surface scars, and flake curvature as distinguishing characteristics. Biface percussion reduction debitage categories include: primary and secondary decortication flakes, and early (large interior flakes which tend to be thick, of variable shapes, and only slightly curved from proximal to distal ends), middle (smaller than early tinning flakes, thinner, and slightly more curved from proximal to distal end, with moderately complex dorsal surfaces), and late stage thinning flakes (thin, regularly shaped biface thinning flakes of uniform size, complex dorsal surface, moderately curved from proximal to distal end, and multi-faceted, sometimes ground striking platforms). Cortical tinning flakes and early interior thinning flakes were subsequently combined as early stage biface reduction debris. Core/flake percussion reduction debitage categories include: primary and secondary decortication flakes, simple interior (displaying two or fewer dorsal arrises), complex interior (displaying three or more dorsal arrises), and rectangular/linear flakes. Other flake types recognized include: angular shatter (angular chunks displaying sufficient flake attributes such as compression rings, hackles, or partial bulbs of force, signifying that they were culturally produced), edge preparation flakes (small [ < 1.5 cm], usually whole percussion flakes), indeterminate percussion flake fragments, pressure flakes, edge preparation/pressure flakes, and indeterminate flake bits. GROUND AND BATTERED STONE ANALYSIS Milling equipment is dominated by millingstones (n=387), and handstones (n= 179), followed by fewer miscellaneous fragments (n=73), and the rare bedrock slick, mortar, and pestle. These tools were analyzed using methods applied to similar assemblages from Owens Valley and the central Mojave Desert (Basgall and McGuire 1988; Basgall et al. 1988), but modified to some extent to deal with analytical problems unique to our project's goals (see also Fratt and Adams 1993). 38 a b C d o 5cm Figure 6. Complete Obsidian Cores from INY-2826 (a, b,); INY-4327 (c); and INY-4319 (d). a. Bifacial (987-29-70); b. Non-patterned (987-29-64); c. Cobble/Chunk Test (987-101-64); d. Non-patterned (987-97-352). 39 a b d c 0 5cm Figure 7. Complete Obsidian Flake Tool from INY-2103 (a), and Unifaces from INY-4267 (b) and INY-4327 (C,7 d). a. 987-27-385; b. 987-88-52; c. 987-101-17; d. 987-101-18. 40 Millingstones Aboriginal decisions regarding materials used were often influenced by the natural form and consistency of locally available rock. Rock ypes found only as boulders and cobbles, for example, were rarely selected when a portable, thin slab millingstone was desired. Hardness also influences wear patterns, as softer rocks develop diagnostic wear patterns more quickly than resistant stone. Finally, the presence of exotic rock types (usually in the form of handstones and thin slab millingstones) can also provide information regarding settlement mobility, particularly if the source locality is known. For all of these reasons, material type was the first major attribute used to classify millingstones. The vast majority of millingstones fall into four basic rock types: rhyolite, granite, basalt, or dike rock. Rhyolite is least resistant of the four, and is common as natural blocks and slabs on the sides of rhyolitic domes. It is quite light and friable, often exhibiting pumice-like qualities. Granitics are much harder, and various forms found in the Volcanic Field include granite, granodiorite, diorite, and gabbro. Local basalts are also relatively hard, and range from boulders to thin slabs. Most of this material is fme-grained, dense, and heavy; however, vesicular basalt occurs in a few nearby areas. Dike rock is a fine-grained material extruded from fissures within the granitic outcrops. Harder than the other three, it routinely occurs as thin tabular pieces. Once identified by material, millingstones were classified according to maximal thickness. Block millingstones are more than 13.0 cm thick (Figure 10b), and are believed to represent essentially stationary tools. Slab millingstones measure between 6.0-13.0 cm thick, and could reasonably have been transported short distances (Figures 8, 9b, 10a). Thin slabs, those less than 6.0 cm thick, were readily transportable (Figure 9a). In addition to other morphological measures such as length, width, and weight, several important functional attributes were recorded for each worn surface. Surface configuration measures the depth of the working surface, and was recorded as flat (no visible concavity), slightly concave (depth < 1.0 cm), concave (depth > 1.0 cm), and irregular/convex. Concave forms rarely achieved depths greater than 2.0 cm; deeply basined millingstones do not occur. Wear type refers to some combination of polish, pecldng, and striations, while surface wear attempts to measure how extensive wear has been (smooth/complete, irregular, incipient). Other observations include length and width of the grinding surface, whether the implement was intentionally shaped, the number of worn surfaces, and artifact condition (complete, nearly whole, margin fragment, interior fragment). Material-specific morphological forms outlined above (i.e., blocks, slabs, and thin slabs of granite, rhyolite, etc.), were cross-tabulated with functional attributes to identify correlations between artifact form, function, duration of use, condition, etc. Patterning among the millingstones, combined with attributes of the larger assemblage (e.g., flaked stone, features, plant remains), contribute to a variety of inferences regarding subsistence-settlement strategies. Handstones Handstones were analyzed using an approach similar to that of millingstones. Material types include the four outlined above, plus sandstone. Conceding that any and all handstones are portable, morphological types were not explored. Instead, material type was cross-tabulated with a series of functional attributes including surface configuration (flat, slightly convex, convex, rounded), wear type (polish, pecking, striations), and surface wear (smooth/complete, irregular, incipient). Secondary wear patterns (edge battering, edge flaking, edge grinding) were noted, as was tool condition. On bifacial (and the few trifacial) handstones, each milling surface was recorded separately, including dimensions of the worn area. Representative unifacial and bifacial handstones of granite are shown in Figures 11 and 13, rhyolite ones in Figures 11-13, and a basalt handstone in Figure 12a. Miscellaneous Ground Stone Miscellaneous ground stone does not represent a distinct functional class, but rather fragments of either millingstones or handstones too small to classify. Analytical methods, therefore, were essentially equivalent to those applied to millingstones and handstones, recognizing the same material types and functional attributes. 41 0 20 cm Figure 8. Complete Unifacial Slab of Granite (987-92-1799) with a Planar Surface from INY-4329. Other Milling Equipment Other kinds of milling equipment were rare, comprising seven mortars (including one bedrock mortar), one pestle, and a few bedrock slicks. All of these artifacts were recorded using the same basic techniques outlined above. Non-Obsidian Cobble-Cores Included in this category are 311 various percussion flaked and/or edge modified cobbles or chunks of non- obsidian material. Some no doubt were hammerstones (e.g., Figure 14d) used by the flintknappers in working obsidian, but many operated as part of the milling assemblage based on context and modification type. Material types parallel those represented in the milling equipment. In addition to basic metrics (length, width, thickness, weight) and condition, we recorded the type and prevalence of use-related damage. Most commonly, edge flaking (Figure 14a), edge battering (Figure 14b), or both (Figure 14c) occur on dike rock or granitic cobbles. Basic shape of the original object (tabular, globular, angular cobble, etc.) was also documented. MISCELLANEOUS ARTIFACTS Unusual items are collectively referred to here. Those readily identified as ornaments include 247 glass beads, three shell beads (Figure 15), and perhaps the one piece of incised stone. A single quartz crystal was recovered, as was a thin, lightly surface-worn stone, provisionally identified as a palette. Pebbles or rocks obviously 42 a b 0 20 cm Figure 9. Select Complete Unifacial Granite Millingstones from INY-1906 (a) and INY-1984 (b). a. Thin Slab with a Planar Surface (987-105-131); b. Slab with a Slightly Concave Surface (987-86-27). 43 Figure 10. Select Complete Unifacial Millingstones from INY-3456 (a) and INY4252 (b). a. Slab of Rhyolite with a Deeply Concave Surface (987-44-68); b. Block of Dike Rock with an Irregular Surface (987-93-76). 44 b a d c 0 5cm Figure 11. Select Handstones from INY-1906. a. Bifacial Granite Handstone with Bi-planar Surfaces (987-105- 165); b. Unifacial Granite Handstone with a Planar Surface (987-105-286); c. Bifacial Granite Handstone with Planar and Slightly Convex Surfaces (987-105-169); d. Unifacial Rhyolite Handstone with a Planar Surface (987-105-171). 45 b a C d 0 , 5cm Figure 12. Select Handstones from INY-3012. a. Bifacial Basalt Handstone with Planar and Slightly Convex Surfaces (987-9-116); b. Bifacial Rhyolite Handstone with Slightly Convex Surfaces (987-9-115); c. Bifacial Rhyolite Handstone with Planar and Slightly Convex Surfaces (987-9-236); d. Unifacial Rhyolite Handstone with a Slightly Convex Surface (987-9-227). 46 a b c 0 5 cm d Figure 13. Select Handstones from INY-2103 (a), INY-4244 (b), and INY-4329 (c, d). a. Unifacial Rhyolite Handstone with a Slightly Convex Surface (987-27-370); b. Bifacial Granite Handstone with Slightly Convex Surfaces (987-46-121); c. Bifacial Granite Handstone with Slightly Convex Surfaces (987-92- 1525); d. Unifacial Granite Handstone with a Slightly Convex Surface (987-92-1626). 47 a d C 0 5cm Figure 14. Selected Complete Non-Obsidian Cobble-Cores from INY-4330 (a-c) and INY-4378 (d). a. Edge Flaked Piece of Dike Rock (987-91-9); b. Edge Battered Cobble of Dike Rock (987-91-81); c. Flaked and Edge Battered Chunk of Dike Rock (987-91-209); d. Battered Sandstone Cobble (987-34-190). 48 b 0'' a 0 .q':? b 5 cm Figure 15. Olivella Beads from INY-1824 (a) and INY-4378 (b). a. 987-26-90; b. 987-34-149. carried into the site by humans, since they are non-local materials, were catalogued as miscellaneous stones (n=23). If such pebbles or rocks had been altered or shaped in some fashion, they were catalogued as miscellaneous shaped stones (n= 19; Figure 16a-b). Peck-marks on seven items identified them as anvils. Four pumice abraders were also recovered (Figure 16c). Finally, only 41 pieces of pottery were found, all identified as Owens Valley Brown Ware. a b 0I c 0 - - cm Figure 16. Miscellaneous Artifacts from INY-1907 (a) and INY-3012 (b, c). a. Miscellaneous Shaped Stone of Fine-grained Igneous Material (987-8-784); b. Miscellaneous Shaped Stone of Quartzite (987-9-488); c. Pumice/Rhyolite Abrader (987-9-163). 49 FLOTATION ANALYSIS (by E. Wohlgemuth) Eighty-six flotation samples were collected from project sites, all from feature contexts. Each sample was water separated in a large washtub, by skimming and decanting to recover both floating and less buoyant plant remains. Burnt macrofossils were collected using 40 mesh/inch (0.42 mm) screen, and rinsed thoroughly with a fine water spray. The heavy fraction sample was washed through graduated mesh of 3.0 mm and 1.0 mm to recover additional non-floating macrofossils and obtain a sample of other micro-constituent debris. Subsequent analyses were limited to 53 of the original samples; ones that yielded only trace amounts of carbonized material were given no firther consideration. Light fractions were size-sorted using 1.75 mm (10/inch), 1.00 mm (16/inch), and 0.7 mm (24/inch) mesh. The materials were passed through each screen individually, as opposed to a nested manner, to minimie damage to the fragile particles. Plant remains retained in the 1.75 mm screen were sorted with the naked eye, while debris in the 1.0 and 0.7 mm grades was examined under a microscope at 10-20X. Seeds were separated from remaining charcoal, and subsequent effort concentrated on identifying seeds to the most specific level possible (i.e., species, genus, family). E. Wohlgemuth and E. Honeysett performed all seed separations and identifications, consulting with specialists at the Seed Identification Laboratory, California Department of Food and Agriculture, Sacramento, for independent verification and identification of particularly troublesome specimens. Flotation sample results are summarized in Chapter 7. FAUNAL ANALYSIS In spite of the high volume of site deposit excavated, only 315 pieces of faunal remains were recovered. All vertebrate fauna collected was initially segregated between bone fragments that could be characterized to taxon and element, and those too fragmentary for identification. The identifiable portion was then classified as specifically as possible, using the comparative faunal collections provided by the Department of Anthropology, University of California, Davis. In cases where some degree of segregation was possible, but species-level identiflcations could not be made, specimens were placed within higher order taxonomic units (e.g., genus, family, order, etc.). Remaining fragmentswere separated into two broad categories: "large mamnnl," including animals of canid size or greater; and "small mammal/bird," encompassing all smaller forms. Finally, all bone was also segregated according to the presence/absence of burning. Bird and mammal remains were identified by Hildebrandt, augmented by D. Simons. Each category was quantified by count and uncorrected weight. Minimum number of individuals was not computed due to the relatively small samples available within discrete occupational areas. No human remains were encountered. 50 CHAPTER 4: ENVIRONMENTAL PRODUCTIVITY Several hard and soft natural resources occur in the Volcanic Field in great enough abundance to have attracted prehistoric Native Americans to the area. Obsidian is the most noteworthy hard resource, and we have already discussed its distribution and abundance (see Chapter 2). We turn our attention here to the distribution and availability of those plants and animals that were important subsistence resources. Annual precipitation is quite low throughout this region, averaging about 14 cm, and ranging from 0.5 to 2.0 cm a month. Low rainfall and extremely warm temperatures throughout much of the year (Table 7) combine to produce sparse, drought-resistant vegetation. According to Kuchler's (1977) general vegetative reconstruction (Map 7), the Volcanic Field was dominated by the Mojave Creosote Bush community, an association that covers much of the Mojave Desert to the souti and east. Juniper-Pinyon Woodland occurs several kilometers to the west in the southemn Sierra, as well as in the higher elevations of the Coso and Argus ranges, north and east. Further north, in Owens Valley, vegetation associations were more diverse. Before the Los Angeles Department of Water and Power Aqueduct was built, the Owens River and its well-watered tributaries were bounded by riparian vegetation. Desert Saltbush covered areas set back from lthe lake and river margins, and Sagebrush Steppe extended up the valley flanks to the Juniper-Pinyon zones of the Sierra Nevada to the west and the Inyo Mountains to the east. Leimer and Leitner (1988) have documented more specific vegetative patterns for the Volcanic Field, and defined four basicplant associations: Desert Saltbush Scrub, Mojave Mixed Woody Scrub, Mojave Creosote Bush Scrub, and an herbaceous subunit of the Desert Saltbush Scrub community (Map 8). The Desert Saltbush Scrub community usually occupies basins and is often dominated by a single species. Fourwing saltbush (Arniplex canescens) is dom nt in uplands, while allscale s (A. polycarpa) or e (A. confertifolia) are common in lower elevational basins. Associated shrub species include bud sage (Artemisia spinescens), spiny hopsage (Grayia spinosa), and cheesebush (Hymenoclea salsola). The herbaceous layer is more diverse and more responsive to annual variations in rainfall, and often includes coreopsis (Coreopsis bigelovii), pincushions (Chaenactis spp.), Pringle's eriophyllum (Eriophylnum pfinglei), yellow pppergrass (Lepidiumfnaum sEpp. favum), and apricot mallow (Sphaeralcea ambigua). Mojave Creosote Bush Scrub is similar to the previous community but iuncludes creosote bush (Larrea tidentata) as a dominat or co-dominant species. Usually crosote is co-dominant with spiny hopsage, cheesebush, fourwing Iterft (Krascheninnikovi lanata), indigo bush (Psorothamnus arborescens var. minutifolius), mand/or desert tomato/boxthor (Lyum andersonii and L. cooperi). On deep cinder or sandy soils in the uplands, creosote bush and cheesebush are the only dominate shrubs, and are usually associated with an understory strongly domiated . by coreopsis. A third phase is found on alluvial fans with a cemented subsoil hardpan. Dominated by creosote and burrobush (Ambrosia dumosa), this phase is rare in the study area but common elsewhere in the Mojave Desert. Mojave Mixed Woody Scrub is quite variable in structure, usually forming one of two vegetation phases depending on substrate condition. Deep alluvial nts support associatons of spiny hopsage, fourwing saltbush, cheesebush, and desert tomato. In addition to these species, steeper rocky soils produce a rich assemblage of winterfat, buckwheat (Eriogonumfasciculum var. poliofolium), Mormon tea (Ephedra nevadensis), and indigo bush. Joshua trees (Yucca brevifolia) are the most conspicuous plant in this community, though not a particularly common one, tending to be restricted to deep alluvial soils on gentle slopes. The herbaceous subunit of the Desert Saltbush Scrub community occurs in some of the small playas sporadically distributed across the study area. It contains many herbaceous species found in the Desert Saltbush Saub zone (e.g., coreopsis, pincushions, and apricot mallow), but lacks the corresponding shrubs. The lack of shrub cover appears to be the result of occasional soil saturation and/or the accumulation of dissolved salts in the soil. 51 1. MOJAVE CREOSOTE BUSH SAGEBRUSH STMPPE k DESERT SALTBUSH JUNIPER-1INYON WOODLAND L CONIFER FOREST JOSHUA TREE SCRUB ALPINE COMMUNTIES Adapted from Kuchler 1977. 1A15'A~ A A V. v '. ....a ?ii*-- ?,'-- v...' v..; *-; .v'. "Jv1 ' ..v. '.:v'"'. :V. V ,,, ,, ~ i'i . .l,J'll''''... v' v. v. A-. A. K AA ? A AAA A A ~A All ~ '1 A' . . -v .' - .? a--a 'l :. ' .''"'' '?V'' '' ' ? " ,:i,A,d Al i J J W ." 1'J v.Y. .V??v' .??'' .~. . ,4 .' Al. .... aA a..' A 4, A . A A OWNS7'w A A A A 'AA A ''' ".." '. A:,'. .' ,-e1,'lA Al, *...iy ..:....:....A... ,;;..... ...~,.:;.... '~ i L z.:^ % ? ?.1.....*.' A' ? r ' A A A ? .,.-.,, ? A i j-z... i .'' ~.'".;4,Y:.y.v. . v""Av ' ' ~"'f"" '"''.- ? 0- A l .......... ,,",lv... v . A, A-"ti..... / .A ." .'" . .L . . '. .. " .. . ...- 'a..... 'v' -'..'"" " - A A A AI AA' AA A A AA ~ L K A A AA AAV~ A A A A AA A A A A A A A A A ' . ' ' v.'. .. . A .,-'..,.: , , A 'A'~A~,,', .'' V...Vk~''-","~v AA A A A A A A A A A A A~A A A A A ~~~A ..AaVi~Y A A i'V:~?..~v\ 'i .V.d, V A?.V.' .V..........:.. A A AAAA A A A A.,:A AL A A A AAA AA, A A A A A A ~ A ~ .~A & A A A A A A A A AA AL V1 L.,....-.,:,.~A., v,,....,.. .... ...~....-...-- A A A A A A A , A AA, . A A l A AA h V* V A":";'*',':,.',. . A. A A A. ALA A :' IAAAA: ...".':-,k-,,-.'. '..V &&, L * ? 41~:;~ , w ..T,~ .:.v-v: A AAA4ALA AAL L ? A A A ~ A~ A A A A A AlA& aAA Al 'A L * A,A A A A AA. A 1A U.T..:.' A A A A l a ' Atv y V *: A A A ~~~A AAA LAIAIA . A A A A ALA 1n1L~~ AAAAAjV V V JI Y A A L,A?A AlAAA 1Ak V ~ VI!f' V oA A l A AA .L'. ..... A:....~ A A A ' . . : :.4diJ Ai ALAAA ..L V.. :. V L VL?..:'..'.. A A A A........ A A A A SA :A? AA A ,V V V A A, L, Ll . .. .*. .. v A A A A A A A.l....L..A A . . A:;. A A A: A'':..A A A Al V ' v . V AA A A A IAAI AlA AAIIA AIL A 10 0 10 20 MLES - 1- -0 W iO-- a1 == 0 10 = 0 KLOMETEW ,._.. *70 Map 7. Regional Vegetation. 52 - MOJAVEMMD WOODY SCRUB D MOJAVE CREOSOTE BUSH SCRUB DESERT SALTBUSH L ISCRUB D HERBACEOUS AFTER LEITNER AND LEITNER 1988 o 1000 FEET o 500 METERS CONTOUR INTERVAL 100 FEET Map 8. Local Vegetation. 53 Table 7. Average Monthly and Mean Annual Precipitation and Temperature at Select Locations in the Region. Average monthly and mean annual precipitation (in cm) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Independence 2.5 2.0 1.2 0.6 0.5 0.2 0.4 0.4 0.4 0.8 0.9 2.8 12.7 (elev. 1200 m) Lone Pine 3.1 2.6 2.7 0.5 1.0 0.2 0.2 0.4 0.6 1.1 0.3 1.7 14.4 (elev. 1136 m) Haiwee 1.7 2.0 1.7 1.2 0.5 0.5 0.9 0.8 0.5 0.9 1.4 2.0 14.1 (elev. 1146 m) Average monthly and mean annual temperature (in ?C) Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Annual Independence 4.1 6.2 9.6 13.7 17.7 22.6 25.8 24.6 20.7 14.9 8.9 4.4 14.4 (elev. 1200 m) Lone Pine 4.4 6.7 9.9 13.3 16.3 20.9 23.8 22.4 18.4 13.6 8.7 4.2 13.6 (elev. 1136 m) Haiwee 4.6 7.5 10.3 14.6 19.0 23.4 27.3 36.2 22.1 16.3 10.1 5.4 15.6 (elev. 1146 m) PLANT RESOURCES The above communities provide a variety of subsistence resources that were exploited by local Native American peoples. Following Delacorte (1990), these resources are organized according to shared similarities in seanal availability and techniques used to exploit them (Table 8). Plants listed derive from ethnographic reviews by Bettinger (1982b), Delacorte (1990), Fowler (1986), Steward (1933, 1938), and Zigmond (1981), cross- referenced with regional and local plant inventories developed by DeDecker (1984) and Leitner and Leitner (1988). Five plant complexes are defined: greens and shoots, roots and tubers, seeds, fruits and berries, and tree crops. Most greens and shoots were available for a short period of tie in spring, providing important food at the end of winter when stored resources were depleted. Most commonly exploited species were those concentrated in moist settings near permanent water (Delacorte 1990). These settings not only produce the highest density and diversity of plants, but also have a growing season prolonged by the availability of water. The study area lacks wetland-adapted plants (e.g., watercress, tule, cattail) and other mesic species, but several dryland species exist (Table 8). Of these, coreopsis is particularly abundant, common i the Desert Saltbush Scmb (including the herbaceous subunit) and Mojave Creosote Bush Scrub commuities. Collecting these and other greens required little time and effort, as most were either eaten raw or minimally blanched/leached with boiling water (sometimes extensively to eliminate toxicity), characteristics leaving little evidence of their use in the archaeological record (Delacorte 1990). Plants with edible roots and tubers include a variety of species adapted to a wide range of habitats (Table 8). As with greens and shoots, wetland species such as cattail, tule, chufa ft sedge, spike-msh, and other moisture dent taxa (e.g., buttercup, wood brownies), were not available in the Volcanic Field. Four important dryland forms are present (Inyo onion, Sego lily, thistle, and blue dicks), but none is considered a major component of the four plant comues defined. The four taxa present could have been collected year-round with a digging stick, but they are most nutritious after the spring/summer growing season. Subsurface roots and tubers accumulate large reserves of carbohydrates during the growmg season that are depleted by growth in the following spring (Delacorte 1990). The difficulty of locating these plants in winter (i.e., the dying back of surface vegetation, and periodic snow cover), indictes most collection was in summer and fall. Many required little or no processing, while others were roasted or pulped to remove edible starch. If the latter techniques were used, archaeological indicators would include hearth/oven features or heavy-duty flaked and battered stone tools (Delacorte 1990). 54 Table 8. Important Economic Plants of Southeastern California. Species EDIBLE GREENS AND SHOOTS Allium atrorubens Amnsinckia tessellata Anisocoma acaulis Caulanthus sp. Claytonia perfoliata Coreopsis bigelovii Nasturtium officinale Phacelia sp. Scirpus nevadensis Stanleya elata Tnifolium womskioldii Typha domingensis Yucca brevifolia EDIBLE ROOTS AND TUBERS Allium atrorubens Calochortus sp. Cirsium sp. Cyperus esculentus Dichelostemma capitatum Eleocharis sp. Fritillaria atropurpurea Lewisia rediviva Ranunculus cymbalaria Sdrpus nevadensis Typha domingensis SEEDS: Perennial Grasses Agropyron sp. Agrostis sp. Bromus sp. Deschampsia sp. Eymus sp. Glyceria striata Hilaria jamesii Melica inperfecta Muhlenbergia sp. Oryzopsis sp. Poa sp. Sitanion hystrix Sporobolus sp. Stipa speciosa Trisetum spicatum SEEDS: Perennial Herbs Asclepias sp. Cymopterus sp. Eriogonum sp. Juncus balticus Polygonum sp. Common name Inyo onion * Fiddleneck * Scale bud * Wild cabbage * Miners' lettuce Coreopsis * Watercress Phacelia * Tule Prince's plume * Clover Cattail Joshua tree * Inyo onion * Sego lily * Thistle * Chufa flat sedge Blue dicks * Spike-rush Wood brownies Bitterroot Buttercup Tule Cattail Wheatgrass Bentgrass Bromegrass * Hairgrass Wild rye * Fowl meadow grass Galleta Melic grass * Scratchgrass Ricegrass * Bluegrass * Squirreltail * Dropseed Needlegrass * Milkweed * Buckwheat * Wire grass Smartweed Season of Availability April-November March-May March-May March-June April-June March-May February-November March-June April-July April-June April-September March-May April-May June-November June-November August-December December-May May-November October-April June-November May-November July-November July-April July-April July-September July-September May-August August-September July-September August-September June-July May-June July-October May-September July-August August-September June-October May-July August-September July-October May-August April-December June-September July-October 55 Table 8. Important Economic Plants of Southeastern California (continued). Common Name SEEDS: Perennial Herbs (continued) Ruppia sp. Scirpus nevadensis Solidago sp. Stachys albens Suaeda torreyana Trifolium wormskioldii Triglochin debilis SEEDS: Annual Grasses and Herbs Amaranthus sp. Chenopodium sp. Descurainia sp. Eragrostis orcuttiana Festuca sp. Gilia sp. Helianthus sp. Lappula redowskii Lepidium sp. Mentzelia sp. Oenothera sp. Panicum capillare Salvia columbariae SEEDS: Shrubs Artemisia sp. Atriplex sp. Dalea polyadenia Ephedra nevadensis Haplopappus macronema Sphaeralcea sp. EDIBLE BERRIES Amelanchier utahensis Lycium andersonii Opuntia sp. Physalis hederaefolia Prunums virginiana Ribes sp. Rosa woodsii Sambucus caerulea Shepherdia argentea Vitis arizonica TREES Pinus monophylla Prosopis glandulosa Prosopis pubescens Quercus chrysolepsis Ditch grass Tule Goldenrod * Hedge-nettle Seep-weed Clover Alkali arrow grass Pigweed * Goosefoot Tansy mustard * Love grass Fescue * Gilia Sunflower Stickseed Peppergrass * Blazing star * Evening primrose * Panic grass Chia * Sagebrush * Saltbush * Nevada Dalea Mormon tea * Goldenrod Globe mallow * Service berry Desert tomato * Prickly pear * Ground cherry Western chokecherry Currant/gooseberry Wild rose Elderberry Buffalo berry Canyon grape Pinyon pine Mesquite Screw bean Canyon live oak Species NOTE: * - present in the Coso Volcanic Field. 56 Season of Availability May-August July-September August-October June-October June-September June-October June-October August-October July-October April-September June-November May-July May-July August-October May-August April-June May-August July-October August-October May-July September-November July-October June-October April-May August-October April-July June-July May-June May-July June-August June-July July-August July-September July-October May-June July-August September-November August-September August-September September-November Small seeds were obtained from a wide range of grasses, herbs, and shrubs (Table 8). For most of these species, there is only a short interval between when the seeds are ready for harvest and when they are taken by animals or lost on the ground. Successful collection, therefore, required carefully monitoring local stands and quicldy harvesting the crop (Delacorte 1990). As discussed in more detail below, full use of these kinds of resources could only be accomplished by people with a thorough knowledge of the distribution and condition of local seed- bearing plants. Seven of the 15 perennial grasses regularly used in the region are present in the Coso Volcanic Field (Table 8). Five of these seven (ricegrass, bluegrass, squirreltail, needlegrass, and wild rye) are listed by Bettinger (1982b) among the most important producers of perennial grass seed in the larger Inyo-Mono area. These grass seeds were available during summer and processed in various ways. Seeds were either detached from the plants using seed beaters and collected in baskets in the field, or entire plants were cut, bunched, and transported in large baskets to a residential base where seeds were then detached. Following collection, most grasses were winnowed and parched, then ground into a flour, and either eaten immediately or used in stews, gruels, and breads (Steward 1933; Zigmond 1981). Like some herbaceous seed producers, processing these resources often creates rather conspicuous archaeological remns. Seed grinding is evidenced by illing equipment, while fire hearths could indicate parching, particularly if they contain plant macrofossils lost or discarded during processing. Of the 12 perennial herbs considered regionally important seed producers, only milkweed, buckwheat, and goldenrod are found in the Volcanic Field (Table 8). Anmnal grasses and herbs are much more andant, with seven of 13 regionally important species growing here. Two of these, blazing star and chia, are also classified as major, region-wide subsistence resources by Bettinger (1982b). Some of these seed-bearing herbs and annual grasses produce during the spring-summer (buckwheat, tansy mustard, ess, blazing star, chia), while for others seed production begins in sumer and extends into fall (milkweed, goldenrod, pigweed, evening pnmrose). All species were harvested with a seed beater and basket. Most seeds were ground into a flour (some after parching), and some were mixed with water to make a beverage (e.g., buckwheat, tansy mustard, peppergrass, chia). Blazing star was ground into a paste similar to peanut butter (Zigmond 1981). Seed producing shrubs are also abundant at Coso, represented by four of the six listed regional taxa (Table 8). Each of these four is dominant in at least one of the local plant communities; however, none appears to have been a staple. Sagebrush, due to its bitter taste, was mixed with other materials only during periods of resource stress (Steward 1933), while saltbush, Mormon tea, and globe mallow seed are only mentioned in passing (Irwin 1980; Steward 1933; Zigmond 1981). The latter is particularly abundant in some prehistoric sites (e.g., Crater Middens [cf. Bettinger 1989:287-288]). Fruit-bearing plants, although relatively numerous in the larger region, are limited at Coso to desert tomato and pricldy pear cactus (able 8). Both species are quite common here, particularly desert tomato which is a major component of the Mojave Mixed Woody Scrub commity. Fruit becomes available in late spring and early summer on both plants. Desert tomato was eaten raw, juiced, or pulverized into a paste. Cactus fruit was knocked from the plant with a stick, cooked, then eaten (Zigmond 1981). Regionally important tree crops include pine nuts, acorns, and mesquite beans. None of these exists in the study area. Pine nuts occur along the southern Sierra and in the Coso and Argus ranges, acorns are confined to the southern Sierra, and mesquite occur deeper in the Mojave Desert to the south and east. PLANT RESOURCE SUMMARY Clearly, only a small portion of the plants used regionally by aboriginal peoples existed within the Coso Volcanic Field. Beginning in spring, when a marginal resource base developed in the form of edible greens and shoots, nearby wetland areas such as Owens Valley or Little Lake would have provided a greater abundance and diversity of hese plants. With the advent of unmme r, local resource productivity peaked when seeds from several pernial grasses and other herbaceous plants could be harvested. This was also when roots and tubers, and fruit from a few additional taxa were collected. Although other areas in the region were also at peak production during the sa months, the Volcanic Field no doubt attracted a fair number of people on at least a short-term basis. By 57 fall, subsistence productivity was much higher in the nearby mountains (pine nuts and acorns) and in the desert to the east (mesquite beans). ANIMAL RESOURCES Economically important animal resources can also be grouped according to similarities in behavior, seasonal distribution, and exploitation techniques. Bettinger (1982b) has proposed six such groups, including pronghorn and jack rabbits, deer and bighorn sheep, small game, insects, waterfowl, and fish. Pronghom antelope (Antilocapra americana) and jack rabbits (Lepus calfornicus) were common throughout the desert scrub and upper sagebrush areas of eastern California. Because of their excellent eyesight and speed, these animals could not be successfully acquired through stalking or ambush, but were most effectively obtained through communal drives. Pronghom are predominately a herding il, the size and density of their aggregations varying according to season. During summer, herds usually are dispersed, due to the wide range of available browse. With the advent of winter, feed becomes more restricted, causing populations to congregate in localized areas of abudnce, makdng them, therefore, more susceptible to communal drives (McClean 1944; Steward 1938). During such drives, animals were usually herded between two brush fences which ultimately converged into a brush and wood corral where they were taped and dispatched. Jack rabbits were also driven by large groups of people. Rather than brush enclosures, fiber nets were used to form a large semi-circular trap where animals could easily be killed once entangled (Steward 1938). Both pronghorn and jack rabbits were, of course, also taken individually. Mule deer (Odocoileous hemionus) and bighorn sheep (Ovis cunadensis) are entirely different from pronghom and jack rabbits with respect to behavior, distribution, and capture techniques used. Deer would have been rarely present in the Volcanic Field, favoring the mountainous areas of the Sierra and, to a lesser extent, the Coso and Argus ranges. Bighorn sheep, on the other hand, have a much greater tolerance for arid environments, allowing them to occupy the rocky slopes of the study area. Preferring rugged country with good visibility, small groups of these animals followed a transhumant pattem, foraging along the mountain-valley interface most of the year, while spending e summer at more productive, higher elevations. Both species were hunted by small groups or by individuals. Bighor sihep were most nmmnly obtaed by slowly heing a grou toward a mountain pass, where a few men in hiding dispatched them with arrows (Irwin 1980). Single animals were stalked by an individual hunter, somet using rock decoys to guide animal movement (Irwin 1980; Steward 1933). Bighorn sheep were an important prey from a regional perspective, but of more particular concern for inhabitants of the Coso region, judging from the thousands of sheep petroglyphs that exist here (Grant et al. 1968; Heizer and Baumhoff 1962). Small game was probably the most plentiful a important quanry in the study area. In ition to cottontail rabbits (Sylvilagus sp.), numerous species of rodents were actively pursued, including vanous squirrels, rats, mice, and chipmunk. These animals were taken using a variety of snares and taps, as were small game birds such as quail (Steward 1933). Desert tortoise and a variety of lizards were also important resources, the latter largely resticted to desert areas to the east and south (Schneider and Everson 1989). Insect larvae were significant in the diet, however, the most important of these (brine fly larvae [Hydropyrus hians]) was available only from Owens Lake. Steward (1938:83) makes menton of piugi gatherng at Little Lake, perhaps using the term to broadly refer to large grubs, larvae, or army worms (cf. Fenega and Fisher 1978, Sutton 1988c), not just Pandora moth larvae. What the grub/larvae available at Little Lake might have been remains unknown. Fish, shellfish, and waterfowl were also non-local, essentially limited.to the Owens Valley and Little Lake areas (Steward 1933). DISCUSSION Environmental characteristics of the Coso Volcanic Field provide an interesting context for evaluating exisdng models of prehistoric adaptation. The incredible abundance of high-quality obsidian in the area placed no limitaions on the amount of erial reduced for pesoal use, exchange, or any other purpose. Variable intensities of lithic production must therefore be accounted for by ging demand among consumer groups, changes in flaked stone reduction technology, shifts in settlement organizaton, and the like. Subsistence resources, in contrast, are limited both in an absolute sense, and relative to other areas in the region. Small seeds, rabbits and other small 58 game, and in lesser degree, large game such as pronghorn and bighorn sheep are the only locally available subsistence resources worth mentioning. Conspicuously absent are a wide range of upland (acorns, pine nuts, deer) and lowland (fish, shellfish, brine fly larvae, waterfowl, greens and shoots, roots and tubers) resources - crucial components of the overall subsistence system. These data correspond quite closely to contrasts observed in the ethnographic records from adjacent geographic contexts (Bettinger 1982b; Delacorte 1990). Whereas the Owens Valley Paiute occupied large semi- peranent villages supported by locally present dryland roots and seeds, as well as temporary camps from which more distant resources such as pine nuts, large and small game, and fish were obtained; the Coso Shoshone system required a higher degree of residential mobility of smaller, independent family groups. 59 CHAPTER 5: CHRONOMETRICS Obsidian hydration data provide the prevalent means for assessing the age of archaeological deposits in the Volcanic Field. To date, hydration rim values are available for over 4600 specimens from numerous discrete artifact and debitage concentrations at 381 sites in the Volcanic Field. Nearly 1400 of those rim values are from the 34 excavated sites (Table 9), an additional 2484 come from limited work at some 317 other sites within ca. 15,000 acres in the surrounding vicinity (summarized in Gilreath and Hildebrandt 1991:Appendix D), and the remaning were obtained by other researchers' projects in the Volcanic Field (e.g., Elston and Zeier 1984). Over 900 of the hydration specimens have also received XRF analysis, which identified 99% as Coso glass (359 Sugarloaf, 353 West Sugarloaf, 163 Joshua Ridge, 15 Cactus Peak, and 29 undifferentiated Coso), three as known exotic glasses, and seven as unknown. Given the incredibly low number of non-Coso pieces of obsidian in the XRF sample, non- geochemically sourced hydration specimens are assumed to be of local glass. A chronological scheme was developed by Gilreath and Hildebrandt (1991) for the Volcanic Field by applying the Coso obsidian hydration rate developed by Basgall (1990), to hydration cluster samples from over 450 individual artifact and debitage concentrations within the Coso Volcanic Field. Single period deposits were identified as those loci with a hydration cluster sample having a coefficient of variation (CV, derived by dividing the standard deviation by the mean value) of 0.25 or less, resulting in the recognition of 138 quarry and 233 off-quarry single period loci (Gilreath and Hildebrandt 1991:Table 18). The temporal distribution of single period quarry and off- quarry loci were graphed (Figure 17), and using breaks or shifts in their frequencies through time, eight periods were defined (Table 10). Deposits at numerous other loci appear to have been generated during multiple periods, but they do not represent magnet locations on the landsape (e.g., areas next to springs), only similar kinds of loci simply overlapping in space. They do not contain dfen s or densities of artfacts or other occupational debris (i.e., they are not villages or substantial occupational sites), and, in fact, including multiple-period loci does little to alter the relative proportions of the quarry and off-quarry loci through time. Period names applied here were chosen with regard to the Bettinger and Taylor (1974) nomenclature. The temporal distribution of loci within the Coso Volcanic Field indicates that during the earliest period, single period off-quarry loci considerably out-number quarry loci. In the subsequent temporal interval the two occur with near-equal frequencies. In the Little Lake period, the number of off-quarry sites changes little from prior intervals, but the frequency of quarry loci rises slightly. The Early Newberry period is marked by a spike in the number of off-quarry loci, with a substantial trough occurring between 3000-2750 B.P. Quarry loci are at maximal frequency during the Middle Newberry period correponding with a secondary peak in the frequency of off-quarry loci. During the Late Newberry period off- quarry loci are at record numbers, though the number of quarry loci remains relatively constant. A dramatic decline in the frequency of off-quarry single period loci occurring with a concomitant decrease in the number of quarry loci marks the Haiwee interval. Finally, in the Marana period, the number of quarry loci is at a record low, coresondingwith coparatively few off-quarry loci. For a hydration-based chronology such as this, we remind the reader that poorly represented periods indicate intervals of minimal obsidian exploitation, and need not necessarily indicate minimal exploitation of other resources in the Volcanic Field. The initial chronological assessment of deposits at project sites considered depositional context, the horizontal and vertical distribution of cultural materials, the homogeneity of the assemblage from within an area, and oter relevant clues of depositional integrty. Discrete concentrations of features and/or cultural debris within a site were teaed as potential single period component areas. Hydration sles from those areas and radiocarbon assays from the rare locations containing organic material were forwarded for analysis. Results of the hydration and radiocarbon analyses were used to re-evaluate the hypothesis that materials from that area were deposited during a single period. n interpreig he hydration data, rim values in excess of 24.0 IL were uniformly dismissed from c ation as valid chronological data: the largest reported rim values on projectile points made of Coso glass are 21.1 and 21.5 p from items in the Volcanic Field, and rims thicker than 30.0 it are not uncommon on gravelly- looking flakes. For many areas, available indicators suggested a single period deposit when the hydration sample produced tightly clustered values, ssociated temporally diagnostic artcts were consistent with the age-estimate provided by the hydration samle, the deposit was spatially discrete, and horizontal and vertical distributions of material were 61 Table 9. Chronological Data from Project Sites. Number of Hydration Samples Number of Number of Projectile Points Radiocarbon Assays Other Chronologically Sensitive Items 1816 1824 Main 1824-B 1824-X 1906 1907 1984 2103 2825 2826 3004/3005 3012 3015 3299 3300 3456 4239 4240 4243 4244 4246 4252 4267 4319 4320 4322/H 4324/H 4325 4327 4328 4329 4330 4331 4378 TOTAL 37 113 5 10 61 22 55 79 20 37 82 59 11 40 56 38 22 11 13 25 16 14 74 50 48 29 26 19 20 28 187 35 33 20 20 uncomplicated. For other areas, chronological data indicated material from more than one period was present, and these areas were handled in a variety of manners depending on the magnitude and nature of the temporal mixing. In some instas there was a surface or near-surface hearth, usually with milling equipment associated, overlaying a very sparse obsidian debitage scatter. Characteristically, the hearth would provide a radiocarbon date consistent with the Marana period (ess than 650 B.P.), while hydration rim values indicated the debitage in the arae was older and of variable age. These hearthks with associated milling stations were treated as Marana period component areas, and the sparse, dispersed debitage was excluded as part of their Marana-period assemblage, as it was understood to be background noise. That is, trace amounts of redeposited obsidian debitage often existed where late period hearths were made. In other istances these suface or near-surface milling stations occurred with 62 Site INY- 3 5 2 2 2 8 2 8 1 1 4 3 2 2 5 3 1 6 2 1 1 1 2 1 1 2 1 1 1 1 6 2 1 shell bead 246 glass beads, 1 shell bead 1 glass bead 41 sherds 1 shell bead 1395 63 a - 0 LL m z q e. 0 iiaiii!ii YEARS B. P. Figure 17. Temporal Distribution of Single Period Loci in 250 Year Increments (from Gilreath and Hildebrandt 1991:56). 63 Table 10. Chronological Periods for the Coso Volcanic Field. Bettinger and Taylor 1974 Gilreath and Hildebrandt 1991 Coso Glass Designation Interval Designation Period Interval Hydration Range Marana 650 B.P.-contact Marana VIII 650-200 B.P. 4.2-2.5 ,t Haiwee 1350-650 B.P. Haiwee VII 1275-650 B.P. 5.65-4.2 u Late Newberry VI 2300-1275 B.P. 7.3-5.65 / Newberry 3150-1350 B.P. Middle Newberry V 2800-2300 B.P. 7.9-7.3 , Early Newberry IV 3500-2800 B.P. 8.7-7.9 y Little Lake 6000-3150 B.P. Little Lake III 5500-3500 B.P. 10.6-8.7 z Mohave pre-6000 B.P. Early I and II pre-ca. 5500 B.P. >10.6 A no hearth associated. These milling stations were identical to the Marana-period ones associated with hearths in terms of the condition, type, and other over-all characteristics of the equipment, and in their basic structure. Again, a sparse scatter of obsidian debitage frequently co-occurred that rendered variable pre-Marana age hydration values. These milling stations lacking hearths were cross-dated as Marana period compnts; the small quantity of obsidian found in their proximity was treated as background noise and excluded from the assemblage inventory. A different set of problems pertained to dating deposits from the early end of the occupational sequence. Part of the problem relates to the fact that older modified obsidian tends to displays quite variable hydration bands, perhaps as a result of the hydration rim sloughing off, or perhaps simply resulting from the vagaries of time. One ine of reasoning indicats that e longer an obsidian artifact has been left in an environment, it is more likely to have fallen subject to events that adversely influenced the constancy of rim development (Rhode 1992). An opposite line of reasoning suggests that rim width variability should diminish not increase with time, because hydration m development slows down over te, placing greater constraits on te amount of variability possible. Empirical data support the former position. Figure 18 is based on Coso hydration data generated by Gilreath and Hildebrandt (1991), graphing only those loci classified as single period deposits because they produced hydration samples having a CV value of less than .25 (the mean values range from 1.4-20.7 ,u, standard deviations ranged between 0.2-3.9). As this figure indicates, the standard deviations associated with cluster samples increase as mean values increase. The second part of the problem has to do with the overall veracity of Basgall's (1990) Coso hydration rate for the early end of the sequence. The Coso hydration rate developed by Basgall is robust for the past ca. 5500 years, but appears to be less firm for earlier dates. This is not particularly surprising since most radiocarbon age- hydration pairings figuring into the rate formulation post-date 5500 B.P. - an unfortunate reality due to the low- number of early radiocarbon age-hydration pairings available. A second factor relates to the nature of logarithmic rates in general, which produce a Coso curve that begins to flatten out at about 5500 B.P. (see Figure 1). The effect of using a logarithmic rate is such that a 0.1 & change in the hydration value is amplified with time. For instance, the dierence between 4.9 and 5.0 Iucalculates to 44 years, while the difference between 14.9 and 15.0 t equates to 189 years (using the Haiwee EHT correction factor). Bause isistencies n the actual rim formation process appear to become pronounced with antiquity, and because the Coso hydration rate is perhaps less robust after ca. 5500 B.P., we simply chose to classify deposits with a mean hydration value greater than 10.6 u and CV values typically less than 0.25 as Early period components, with no finer chronological separation. A fourth group of potential components corresponds to areas used during two consecutive periods, signaled by hydration values sp g those periods; or during two temporally discrete events, indicated by a bimodal hydration value distribution. In these instances, the assembly was assigned, intact, to both periods. 64 5 10 15 MEAN VALUE: MICRONS 20 25 Figure 18. Hydration Data Summary from Single Period Deposits in the Coso Volcanic Field (Adapted from Gilreath and Hildebrandt 1991:Appendix D). 65 4 3.5 3 i2.5 2 1.5 1 0.5 0 0 -0 * v S 0 0 * 0 00 * . .0 e- 0 * em. 0* 00- *0 *em oe * m oem m e 0 0 *_ 0 00 0 0..0.. 0 00 @0 00 @0 0 0 . m- v- 0- v -* *-m e e - e -m - 1 - - I For all remaining areas, diverse hydration rim values on samples were interpreted to mean that materials had accumulated during numerous intervals. Such areas were subsequently considered Residual, since their temporally mixed deposits were of very restricted analytical utility in addressing the questions of interest to us. The temporal assignment of component areas at project sites is presented in Table 11. Table 11. Chronometric Summary from Component Areas by Site. Hydration Data n mean s.d. NOE -- -- 2 Uncorrected Projectile Radiocarbon Points Age B.P. -- modernma Marana Period 1816 C 1816 D 1816 E 1816 F 1816 G 1816 H 1906 A 1906 E 1906 F 1906 H 1984 E 2103 D 2825 B 3004/5 G 3300 D 3456 B 4239 B 4243 B 4267 A 4329 A 4329 D 4329 N 4329 R 4329 W 4329 V 4330 B Haliwee Period 4239 A 4240 (all) 4322/H C 4325 B 4378 B 3004/5 P 3012 Cd 4329 Oc 7 4 1 1 1 23 10 10 10 12 9 15 1.5 1.3 2.2 3.7 3.1 4.7 5.5 4.3 5.1 11.7 6.5 9.5 0.2 0.1 0.5 0.1 0.3 1.3 1.9 1.3 2.9 6 5 10 6 4 5 6 9 5 3 7 4 1 4 2 1 2 1 1 1 2 1 1 160+70 110 ?70b 110-?70b 370?70 110?60 130?80 moderna 140?50 and 580?60 310?50 and 360?50 530?60 290?50 600?50 930?80 860?65 66 Historic 4322/H B Other glass beads sherds Table 11. Chronometric Summary from Component Areas by Site (continued). Hydration Data n mean s.d. NOE Uncorrected Projectile Radiocarbon Points Age B.P. Late Newberry Period 1816 A 1824 Main D 1824 Main F 1824 Main Z 1824-B (all) 1824-X (all) 1906 B 1907 A 1984 A 2103 A 2103 B 2826 B 2826 C 3012 D 3012 E 3456 A 4243 A 4243 C 4244 (all) 4252 (all) 4267 N 4378 A Middle Newberr Period 1824 Main J 1824 Main K 1824 Main M Early Newberry Period 1824 Main A 1824 Main Y 3004/5 A 4327 A 4329 K 4331 A 1984 BC Little Lake Period 4319 (all) 4322/H A 4325 A 4330 A 3299 AC 4320 CC 2103 HC 12 11 17 8 5 9 24 15 17 25 13 9 14 4 4 28 9 3 24 14 8 8 10 4 7 8 4 13 5 22 6 19 9 2 1 1 3 14 5 7.0 0.2 6.6 0.6 6.3 0.5 5.7 0.5 6.3 0.1 6.6 0.3 6.5 0.1 5.8 0.3 7.0 0.9 7.0 0.8 6.9 0.3 5.9 0.4 6.3 0.4 6.3 0.5 5.7 0.6 7.2 0.7 6.9 0.8 6.7 0.4 6.5 0.5 5.8 0.5 6.4 1.2 5.8 1.2 7.6 0.9 7.5 0.3 7.2 0.1 8.5 0.5 8.6 0.5 8.5 1.0 8.7 0.7 8.5 1.1 8.7 1.1 11.2 4.5 9.9 1.2 8.8 1.1 10.6 2.4 8.8 1.8 10.4 2.8 10.6 3.8 10.8 1.3 1 4 1 2 2 shell bead shell bead 2 3 6 1 1 1 2 2 1 2 9 2 1 3 3 9 2 1 1 3 1 6 3 3600?280 3080?70 2450?90 67 Other Table 11. Chronometric Summary from Component Areas by Site (continued). Uncorrected Hydration Data Projectile Radiocarbon n mean s.d. NOE Points Age B.P. Other Early Period 1824MainB 4 11.5 2.5 -- -- -- -- 1984 C 5 13.6 2.0 5 -- -- -- 1984 D 2 15.8 4.6 -- 1 .. 3004/5 J 9 12.0 1.7 2 -- -- -- 3299 B 16 12.5 4.3 2 -- -- -- 3300A 19 11.5 2.8 4 1 -- -- 3300 B 19 12.0 3.8 1 -- -- -- 4267 M 2 13.0 2.0 2 -- -- -- 4320 A 12 11.8 1.8 8 -- -- -- 4320 B 7 11.1 1.9 5 -- -- -- 4324/H B 14 11.5 2.4 -- -- -- -- 4327 C 4 11.8 0.9 -- -- -- -- 4329 G 18 11.3 1.7 3 -- -- -- 4329 L 6 18.4 2.2 6 -- - -- 4331B 14 11.9 2.9 4 1 -- -- NOTE: - "modem" reported as a percentage; b - also reported as "modemn"; c - also Early; d also Late Newberry; NOE - number of hydration rim values excluded from mean calculation. RADIQCARBON ASSAYS Twenty radiocarbon dates were obtained from project sites (Table 12). All samples were collected from feature contexts through the use of water flotation. Fifteen features from nine sites produced dates corresponding to the Marana period (i.e., <650 years B.P.), including four classified as modem. Obsidian tools and debitage genuinely associated with these features was rare, precluding the potential for a radiocarbon age-hydration paring. Single features at two sites provided radiocarbon age estimates falling between 1275-650 B.P., corresponding to the Haiwee period. The first, from INY-4325 B, is not associated with an obsidian assemblage; the other, from INY- 3004/5 I, is from a rock concentration with an obsidian assemblage occurring in the same general vicinity but evidendy not contemporaneous, since hydration data indicate it dates to the Early period (Table 13). The single radiocarbon date obtained from a feature in INY-1907 A produced a radiocarbon date of 3080?780 B.P. The bydration specimens from this area, however, cluster tightly around a mean of 5.8 p (s.d. =0.3), are comparable for artifacts and debitage alike, and show no ppreciable difference with depth. The mean hydration value equates with an age of 1360 B.P., falling at the recent terminus of Late Newberry, and the range in hydration values is so narrow that it suggests the obsidian assemblage dates from a single occupational episode. The vertical distribution of materials here provides additional indication of a single period deposit. Using the Coso hydration rate we have adoped, values in the low eights would be consistet ith the radiocarbon date obtaied from the INY-1907 feature, and no rims of this mapitude were recorded. While the possibility remans that the feature pre-dates the dense obsidian assemblage-bearing deposit (in effect, maldng it an isolated feature), they are, in our judgement, more likely to be contemporaneous, which makes the radiocarbon date enigmatic. A betr pairing was obtained from INY-1824 Main-M, where radioar and hydration data from a single feature rendered Middle Newberry period dates of 2450?950 B.P. and 2246 B.P., respectively (Table 13). A single feature at INY-4330 A, provided the earliest radiocarbon date for the project, at 3600?280 B.P., corresonding to the magtude of the associated hydration rim values, allowing temporal assignment to the Little Lake period. 68 Table 12. Uncorrected Radiocarbon Dates from Project Sites. Site and Component 1816 C 1824 M 1906 H 1906 F 1907 A 2103 D 3004/5 G 3004/5 I 4243 B 4267 A 4322/H B 4325 B 4329 A 4329 A 4329 V 4329 D 4329 D 4329 N 4330 B 4330 A Laboratory Number Beta-39765 Beta-33258 Beta-36258 Beta-36259 Beta-36261 Beta-39766 Beta-39361 Beta-36260 Beta-39763 Beta-59992 Beta-39363 Beta-39764 Beta-39360 Beta-39357 Beta-39364 Beta-39358 Beta-39359 Beta-39356 Beta-39362 Beta-30427 Catalogue Number 31-286 26-942 26-943 105-267 105-393 87-176 87-177 27-734 27-735 109-917 109-920 43-51 88-700 98-87 102-73 102-69 92-1671 92-1950 92-1951 92-2437 92-2426 92-1878 92-1879 92-2617 92-2620 91-207 91-234 Provenience N30/W25 N174/E107 N28/W151 N78/E335 N18/E38 S56/W18 S360/E49.8 S288/E191 N34/W69 S60/W20 N84.5/W56.9 S17/E80 N22/W189.5 N176/W157 N122/W57 N9/W56 N7/W55 N74/W83 S29/W122 N 1 l/E3 Nl l/W5 cm bs 8-12 40-60 0-10 0-10 40-48 0-20 0-10 0-10 0-10 0-10 10-20 0-15 10-20 0-30 3040 4-20 0-30 8-17 20-30 3040 69 Flotation Volume (liter) 7.0 30.0 11.5 22.0 9.5 24.0 14.0 7.0 6.5 17.5 5.0 12.5 8.0 6.5 9.0 10.0 19.5 10.0 7.0 5.5 Radiocarbon Age B.P. 160?70 2450?90 110?70 Modem 110?70 Modem 3080?70 370?70 110?60 860?65 130?80 101.6% ?0.6% Modem 99.1%?0.7% Modem 930?80 140?50 580?60 290?50 310?50 360?50 530?60 600?50 3600?280 Context Feature 2 Feature 1 Feature 10 Feature 9 Feature 1 Feature 1 Feature 1 Feature 4 Feature 1 Feature 1 Feature 1 Feature 2 Feature 2 Feature 1 & la Feature 17 Feature 20 Feature 6 Feature 14 Feature 2 Feature 3 Table 13. Radiocarbon and Hydration Pairings from Project Sites. Hydration Rate Site and Radiocarbon Age Hydration Pairing Equivalent Component (B.P. uncorrected) Mean Value (years B.P.) 1906 H 110+70 1.5 59 3004/5 I 860+65 11.7 6927 1907 A 3080?70 5.8 1360 1824 Main M 2450+90 7.2 2246 4330 A 3600?280 8.8 3577 NOTE: The hydration mean value for each pairing except INY-1907 is calculated with select values excluded (see Table 10). PROJECTILE POINTS In total 172 projectile points have been collected from excavations and prior work in the project area: 153 of obsidian and 19 of non-obsidian materials (Table 14). Consistent with the trend reported by Hughes (1988) at the Stahl and Rose Spring sites, and by Basgall and McGuire (1988) at INY-30, West Sugarloaf obsidian tends to be the prevalent Coso material subgroup, followed by substantial quantities of those identified to the Sugarloaf subgroup, but comparatively few of either Joshua Ridge or West Cactus Peak material subgroups. The remanimg obsidian points include one each made of Fish Springs, "Queen imposter," and Mono Glass Mountain obsidian; and two have not been geochemically analyzed. Most non-obsidian projectile points are made of basalt (n= 10), followed by cryptocrystalline materials (n=8), and one provisionally identified as tuff. The projectile point assemblage contains examples of each type commonly identified throughout the Great Basin. Organized from what is generally considered most recent to oldest, they include two Desert Side-notched points and 10 Cottonwood Triangular; six Saratoga Spring and 21 Rose Spring; 10 Humboldt Basal-notched, 31 Elko Corer-notched or Eared, and seven Gypsum; 17 Little Lake, 42 Great Basin Stemmed (14 Silver Lake, 22 Lake Mohave, and six undifferentiated); and three Paleoindian style Concave-based points. Those of less certain chronological placement include eight non-basally notched Humboldt points, ree Elko Side-notched, and eight leaf- shaped forms. The balance of the assemblage consists of two small notched items (either Desert Side-notched or Rose Spring points), one provisionally classified as a Cottonwood, and one provisionally classified as a Little Lake. Summary metrics for the obsidian points for each type are provided in Table 15; individual point measures are presented in Gilrath and Hildebrandt 1995, Appendix A, Volume m; representative examples are shown in Plates 1-9. Hydration rim measures were made for all Coso obsidian projectile points (Table 16). As shown in Figure 19, the mean value of the various point styles increases in magnide from the Desert series, to Saratoga Spring and Rose Spring, to Humboldt Basal-notched, to Elko, and then Gypsum; with Little Lake, Great Basin Stemmed, and Paleoindian points having substantially larger rim values. Using one standard deviation from the mean value, the hydration data indicate very little temporal overlap between Desert series, Rose Spring and Elko series points, with Humboldt Basal-notched temporally falling between Rose Spring and Elko. As shown in Table 17, the hydration rim measures on the majority of the Desert series, Saratoga Spring, Rose Spring and Humboldt Basal-notch items falls within the range identified for the various temporal periods. The hydmtion range for the Marana period, for example, is identified as < 4.2 i, and 69% of the Coso obsidian Desert series points had rim values less than 4.2 I. In sum, the hydration data for Desert series, Saratoga Spring, Rose Spring, and Humboldt Basal-notched points are consistent with the position held by many Great Basin archaeologists regarding their chronology. 70 Table 14. Projectile Points by Material Type. Type Desert Side-notched Cottonwood Triangular Saratoga Spring Rose Spring Humboldt Basal-notched Other Humboldt Elko Corner-notched and Eared Gypsum Elko Side-notched Little Lake Great Basin Stemmed Silver Lake Lake Mohave Other Stemmed Concave-based Others Leaf DSN/RS? Cottonwood? Little Lake? Coso Obsidian Exotic Obsidian WSL SL JR WC Fish Qu-Imp Mono 1 6 3 16 7 3 11 1 2 4 5 7 3 5 1 1 1 1 2 2 3 2 - 1 1 - 2 3 2 ioa 3 2 1 - 1 8 1 3a 5 2 1 1 - 1 2 1 1 - UNK. Non-Obsidian OBS Basalt CCR Tuff Total - - - - 2 - - - - 10 1 1 4 - 1 1 3 1 6 - - 21 - 10 - 8 3 - 31 1 1 7 - - 3 2 1 - 17 1 -- 5 1 1 1 1 - - 1 1 1 1 14 22 6 3 - - 8 - - 2 - - 1 - - 1 77 43 18 10 1 1 1 2 10 8 1 172 NOTE: WSL - West Sugarloaf; SL - Sugarloaf; JR - Joshua Ridge; WC - West Cactus; Fish - Fish Springs; Qu -Imp - Queen Imposter; Mono - Mono Glass Mountain; UNK - not sourced; CCR - Cryptocrystalline; DSN/RS? - Desert Side-notched/Rose Spring?; a - includes one unknown, probably Sugarloaf. The hydration data provide little comfort, however, to those who hold that Great Basin Stemmed points pre- da Little Lake/Pinto series points, and even less comfort to those who contend that the Little Lake/Pinto points largely pre-date Elko points. Rather, the hydration data indicate that in the Volcanic Field, these three point styles were contemporaneous for a long period of time. The chronological significance of Elko points throughout the Great Basin is controversial, and the fact remains that many who key their projectile points through the Monitor Valley scheme, often apply some measue of subjectivity when deciding if an item is referenced as an Elko or Little Lake. It is not uncommon that objecdve application and strict conformiy with the key would requiire classification into one category, but subjectivity over-mles such assigent. Researchers' familiarity with the material culture in the area where they work allows them to recognize the less than perfect examples, and those pieces are often given provisional classification, referred to as "atypical" or "aberrant," often described as non-conforming because of the crudeness of craftsmanship. Consequently, and not infrequently, these irregular items are placed in a category other than that to which they keyed. Elko and Little Lake points from the Volcanic Field raised all of these concerns. Subjectively there was considerable overlap in their appearance, size, and quality of craftsmanship, and they appeared to represent a conim rather than two readily disiguihable styles. For that reason, it came as no surprise to learn that their hydration rim values indicated considerable temporal overlap (Figure 19). Sorted by hydration values, however, it became apparent that Elko points with a maximal thickness less than 6.5 mm routinely displayed quite narrow 71 TOTAL - Table 15. Summary Metrics of Obsidian Projectile Points. ML AL MW TH WT BW NW PSA DSA NOA STL Desert Side-notched (n=2) n mean s.d. max. min. Cottonwood (n= 10) n mean s.d. max. min. Saratoga Sring (n=6) n mean s.d. max. nmin. Rose Svring (n=21) n mean s.d. max. mmn. Humboldt Basal-notched (n=9) n mean s.d. max. min. Elko Comer-notched and Eared (n=28) n mean s.d. max. Gysum (n=5) n mean s.d. max. min. Elko Side-notched (n=3) n mean s.d. max. min. Litte Lake (n= 14) n mean s.d. max. mnn. - 2 - 14.3 - 2.8 - 17.2 - 11.5 6 6 8 26.1 24.0 13.4 8.7 8.0 2.6 39.3 36.6 17.2 15.8 15.0 9.7 4 4 6 39.1 39.0 19.2 10.6 10.7 4.3 56.7 56.7 25.2 30.6 30.6 12.3 10 10 18 29.5 28.3 15.9 5.4 4.3 2.3 37.4 37.4 22.0 22.2 22.2 12.7 2 2 6 63.8 56.4 29.8 10.2 11.4 3.7 73.9 67.8 36.2 53.6 45.0 25.7 14 14 22 48.3 46.2 28.7 8.9 8.1 4.3 59.1 59.1 36.0 28.6 28.6 20.7 1 1 5 47.2 47.2 24.7 0.0 0.0 3.8 47.2 47.2 29.6 47.2 47.2 17.8 1 1 3 50.4 42.0 22.9 0.0 0.0 1.2 50.4 42.0 24.0 50.4 42.0 21.2 10 10 14 40.3 37.8 28.6 6.4 6.1 4.3 51.5 47.8 37.0 29.9 28.1 20.8 2 3.0 0.2 3.3 2.8 10 3.2 0.5 4.0 2.1 6 4.8 1.1 7.1 3.7 21 4.0 0.7 5.9 2.6 9 8.2 1.7 11.2 5.6 28 6.1 1.5 10.5 4.0 - 2 - 14.3 - 2.8 - 17.2 - 11.5 5 7 0.6 11.8 0.2 2.7 0.9 15.0 0.3 7.5 4 5 2.5 19.1 1.3 4.7 4.7 25.2 1.4 12.3 10 15 1.2 8.4 0.3 2.5 1.8 14.5 0.8 7.0 2 6 16.3 28.2 6.4 23.6 22.7 80.0 13.8 13.2 12 18 7.4 21.0 3.4 4.8 12.4 30.7 2.7 11.2 1 2 5.3 10.0 0.0 1.6 5.3 11.6 5.3 8.4 2 7.6 0.1 7.7 7.5 NA NA NA NA NA NA NA NA NA NA 20 7.9 2.1 10.1 5.8 NA NA NA NA NA 25 15.1 3.1 21.2 9.9 5 11.6 2.0 14.0 8.9 3 14.5 2.2 16.8 11.6 14 19.2 3.8 25.9 10.7 2 159.5 0.5 160.0 159.0 5 5.6 0.9 7.0 4-5 3 6.5 0.8 7.6 5.6 14 8.4 1.3 10.6 5.8 2 228.0 2.0 230.0 226.0 - 1 - 20.9 - 0.0 - 20.9 - 20.9 10 13 7.8 19.6 3.6 5.1 15.4 26.8 3.7 8.4 2 68.5 1.5 70.0 67.0 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 24 43.2 22.0 110.0 12.0 5 103.8 26.2 130.0 70.0 4 7.3 1.4 9.7 6.3 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 25 126.8 15.9 150.0 95.0 5 63.8 8.1 80.0 59.0 3 149.7 3.7 154.0 145.0 14 100.0 7.8 112.0 85.0 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 15 4.9 1.0 6.7 2.7 NA NA NA NA NA 26 8.7 2.0 12.7 5.2 NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA NA 26 169.7 16.8 210.0 140.0 5 167.6 21.7 190.0 135.0 3 201.7 23.9 235.0 180.0 14 196.5 21.8 220.0 143.0 3 3 52.0 113.3 27.0 141.5 90.0 313.4 30.0 10.0 14 14 96.5 12.1 22.1 3.5 125.0 21.5 50.0 6.8 72 Table 15. Summary Metrics of Obsidian Projectile Points (continued). ML AL MW TH WT BW NW PSA DSA NOA STL Great Basin Stemmed- Silver Lake Variant (n=8) n 6 6 7 8 7 8 8 NA 8 NA 8 mean 40.3 40.1 25.1 7.3 6.3 16.8 17.1 NA 215.6 NA 12.1 s.d. 8.7 8.8 1.8 0.8 2.2 7.2 4.3 NA 9.2 NA 2.2 max. 56.5 56.5 27.4 8.6 10.3 24.5 21.9 NA 230.0 NA 16.6 min. 27.8 27.4 21.7 6.5 3.3 1.4 7.4 NA 200.0 NA 9.0 Great Basin Stemmed- Lake Mohave Variant (n=20) n 7 7 15 18 9 6 11 NA 10 NA 10 mean 45.9 45.9 28.8 7.8 9.9 19.9 23.5 NA 210.4 NA 24.9 s.d. 6.4 6.4 3.3 1.3 2.2 2.9 2.2 NA 22.2 NA 8.2 max. 59.3 26.3 36.4 10.0 13.6 23.0 26.4 NA 243.0 NA 43.0 min. 40.0 40.0 23.3 4.9 7.0 16.0 19.0 NA 176.0 NA 17.6 NOTE: NA - not aWlicable; ML - maximum length; AL - axial length; MW - maximum width; TH - thickness; WT - weight; BW - basal width; NW - neck width; PSA - proximal shoulder angle; DSA - distal shoulder angle; NOA - notch opening angle; STL - stem length; all measurements in mm. Table 16. Hydration Data Summary of Coso Obsidian Projectile Points from the Volcanic Field. Desert Series Rose Spring Saratoga Spring Elko <6.5 mm thick Elko >6.5 mm thick Elko Side-notched Humboldt Basal-Notched Gypsum Little Lake/Pinto Great Basin Stemmed Concave-Base Leaf-shaped n mean s.d. Rim Values 12 3.0 1.2 1.3, 1.5, 1.9, 2.1, 2.8, 2.8, 2.9, 3.0, 3.8, 4.3, 4.5, 4.7, (6.7) 20 5.2 0.8 (-2.5), 3.6, 4.2, 4.4, 4.5, 4.7, 4.8, 4.8, 4.9, 5.0, 5.2, 5.3, 5.3, 5.3, 5.6, 5.7, 5.8, 5.8, 6.2, 6.3, 6.9 6 4.8 0.8 4.2, 4.3, 4.3, 4.8, 4.9, 6.3 12 7.4 1.0 6.0, 6.6, 6.8, 6.9, 6.9, 7.0, 7.3, 7.6, 7.6, 8.1 9.1, 9.3, (11.6, 16.2, DH, DH) 8 12.3 3.3 (6.2), 8.7, 9.4, 10.7, 11.1, 12.2, 12.5, 14.9, 18.9, (DH) 2 8.6 1.3 7.6, 9.5, (DH) 8 6.3 1.0 (2.3), 4.8, 5.7, 5.7, 5.9, 6.0, 6.8, 7.5, 7.7 4 10.1 2.5 7.9, 8.0, 12.1, 12.4 12 14.2 4.3 (6.1), 9.1, 9.5, 9.5, 10.6, 11.2, 13.6, 14.3, 15.0, 17.7, 18.3, -19.5, -21.5, (DH, VW) 21 12.9 2.7 (5.5), 8.7, 8.9, 9.4, 9.9, 10.3, -11.0, 11.1, 11.5, 11.6, 12.5, 12.9, 13.1, 13.6, 14.2, 14.2, 14.3, 15.2, 15.6, 16.7, 17.5, -17.8, (DH, DH, DH, DH, DH, VW) 2 17.3 5.4 13.4, 21.1 6 11.1 4.0 (4.7), 7.5, 8.7, 8.7, 9.7, 14.0, 17.9 NOTE: () - value not included in calculations; measure. DH - diffuse hydration; VW - variable width; - approximate 73 a DESERT SIDE-NOTCHED 987-18-139 INY-3033 987-62-1 INY-3442 DESERT SIDE-NOTCHED 987-85-6 INY-1969 h COTTONWOOD 987-105-249 INY-1906 i COTTONWOOD 987-105-262 INY-1906 n SARATOGA SPRING 987-18-124 INY-3033 SARATOGA SPRING SARATOGA SPF 987-38-130 987-38-143 INY-4239 INY-4239 0 q SARATOGA SPRING 987-92-1306 INY-4329 r SARATOGA SPRING 987-157-12 INY-1928 5 cm Plate 1. Desert Side-notched (a,b), Cottonwood (c-m), and Saratoga Spring (n-r) Projectile Points. 74 b c COTTONWOOD 987-15-390 INY-3015 d COTTONWOOD 987-18-11 INY-3033 e COTTONWOOD 987-18-137 INY-3033 I COTTONWOOD 987-153-1 INY-4279 f COTTONWOOD 987-61-2 INY-3433 m COTTONWOOD 987-153-4 INY-4279 j COTTONWOOD 987-137-17 INY-4307 k COTTONWOOD 987-137-25 INY-4307 a b c d e f ROSE SPRING ROSE SPRING ROSE SPRING ROSE SPRING ROSE SPRING ROSE SPRING 986-1-1 987-9-353 987-16-4 987-16-24 987-18-132 987-31-170 INY-1816 INY-3012 INY-3017 INY-3017 INY-3033 INY-1816 h ,!,iii!!?!~ :iU J ROSE SPRING SLE N ROSE S'PRING ROSE SPRING ROSE SPRING 987-48-6 987-57-2 9731987-84-1 987-66- 987-69-1 ISOLATE INY-3427 IN INY-3435 INY-3441 ISOLATE mI n) RoSE SPRING ROSE SPRING .ROSE SPRING 987-27-315 ROSE SPRING ROSE SPRING ROSE SPRING ROSE SPRING 987-157-11 INY-2103 987-75-1 987-95-1 987-146-2 987-157 -10 INY-1928 ISOLATE INY-3007 INY-4316 INY-1928 0 5cm Plate 2. Rose Spring Projectile Points. 75 a HUMBOLDT BASAL-NOTCHED 987-27-424 INY-2103 b 'I,c~I HUMBOLDT BASAL-NOTCHED HUMBOLDT BASAL-NOTCHED 987-48-18 987-58-3 ISOLATE INY-3428 d HUMBOLDT BASAL-NOTCHEC 987-67-1 ISOLATE HUMBOLDT BASAL-NOTCHED HUMBOLDT BASAL-NOTCHED 987-103-70 987-157-16 INY-4328 INY-1928 0 HUMBOLDT 987-109-416 INY-3004/5 Plate 3. Humboldt Basal-notched (a-f) and Undifferentiated Humboldt (g-j) Projectile Points. 76 M UMMUL.J I 987-102-44 INY-4325 g HUMBOLDT 987-92-1524 INY-4329 5 cm j HUMBOLDT 987-157-14 INY-1928 - i a b c d ELKO ELKO ELKO ELKO 986-1-47 986-3-183 987-8-6 987-17-43 INY-1816 INY-2824 INY-3011 INY-3016 e f g h.K SIDE-NOTCHED ELKO ELKO 987-26-587 987-26-79 987-26-96 987-26-433 INY-1824 INY-1824 INW-1824 INY-1824 ELKOL ELKO j ~~~987-29-42 i 987-46-10 ELKO INY-2826 INY-4244 6ELKO 987-27-310: :9 : 746-119 INY-2103 INY-4244 0 5 cm E e , Plate 4. Elko Series Projectile Points. 77 ELKO b ELKO 9878-1 ISOLATE T ELKO 987-87-2 INY-1907 -w d c ELKO C g87-85-8 ELKO SIDE-NOTCHED 987-84-1 INY-1969 INY-4236 ELKO 987-88-308 INY-4267 h ELKO 987-91-1 INY-4330 I ELKO 987-92-2064 INY-4329 ELKO 987-92-2012 INY-4329 9 2 ELKO "8987-95-7 NINY-4317 Pa nts m lELKO 987-102L 5 ELKO 987-137-1 ELKO INY-4325 987-109-149 INY-4307 987-138-1 INY-3004/5 IY-4306 ELKO 0 5 cm 987-146-1 INY-4316 Plate 5. Elco Series Projectile Points. 78 a ELKO 987-48-29 ISOLATE e ELKO 987-87-1 INY-1907 a GYPSUM 987-27-32 GYPSUM INY-2103 987-29-9 INY-2826 b GYPSUM 987-27-491 INY-2103 GYPSUM GYSU tt l0i: ::t .:: 0e0 GYPSUM GLITTLE LAKE 987-109-317 987-146-11 987-27-41 INY-3004/5 INY-4316 INY-2103 k LITTLE LAKE LITTLE LAKE UTTLE LAKE 987-48-19 987-91-2 987-109-124 ISOLATE INY4330i INY-3004J5 INY-4300/ 0 5 cm Plate 6. Gypsum (a-f) and Little Lake (g-m) Series Projectile Points. 79 d GYPSUM 987-109-151 INY-300415 LITTLE LAKE 987-27-368 INY-2103 L I IL= LA,MN 987-161-3 ISOLATE UTTLE LAKE 987-44-19 INY-3456 i a GREAT BASIN STEMMED SILVER LAKE 987-58-2 ' INY-3428 GREAT BASIN STEMMED SILVER LAKE 987-88-14 INY-4267 GREAT BASIN STEMMED SILVER LAKE 987-110-3 ISOLATE GREAT BASIN STEMMED SILVER LAKE 987-151-5 INY-4301 e f GREAT BASIN STEMMED GREAT BASIN STEMMED LAKE MOHAVE LAKE MOHAVE 987-6-15-1 987-48-20 INY-4325 ISOLATE GREAT BASIN STEMMED GREAT BASIN STEMMED LAKE MOHAVE LAKE MOHAVE 987-58-1 987-88-40 INY-3428 INY-4267 i GREAT BASIN STEMMED LAKE MOHAVE 987-88-163 INY-4267 i GREAT BASIN STEMMED LAKE MOHAVE 987-90-6 INY-4331 0 k GREAT BASIN STEMMED LAKE MOHAVE 987-106-1 INY-1923 5 cm I GREAT BASIN STEMMED LAKE MOHAVE 987-110-19 ISOLATE Plate 7. Silver Lake (a-d) and Lake Mohave (e-l) Variants of Great Basin Stemmed Projectile Points. 80 b C Ill ........... II a GREAT BASIN STEMMED LAKE MOHAVE 987-110-21 ISOLATE e GREAT BASIN STEMMED LAKE MOHAVE 987-151-3 INY-4301 b GREAT BASIN STEMMED LAKE MOHAVE 987-138-2 INY-4306 f GREAT BASIN STEMMED LAKE MOHAVE 987-151-4 INY-4301 0 C GREAT BASIN STEMMED LAKE MOHAVE 987-138-3 INY-4306 g GREAT BASIN STEMMED LAKE MOHAVE 987-157-13 INY-1928 d GREAT BASIN STEMMED LAKE MOHAVE 987-141-2 INY-4304 h GREAT BASIN STEMMED LAKE MOHAVE 987-157-28 INY-1928 5 cm Plate 8. Lake Mohave Variants of Great Basin Stemmed Projectile Points. 81 a CONCAVE BASE 986-7-1 INY-3431 LEAF-SHAPED 987-26-97 INY-1824 LEAF-SHAPED 987-27-112 INY-2103 h LEAF-SHAPED 987-109-136 INY-3004/5 0 Plate 9. Concave-based (a-c) and LEAF-SHAPED 987-109-413 INY-3004/5 f LEAF-SHAPED 987-57-28 INY-3427 EAF-SHAPED 987-151-1 INY-4301 5 cm Leaf-shaped (d-j) Projectile Points. 82 9 LEAF-SHAPED 987-92-825 INY-4329 21 20 19 18 17 16 15 6| | |PALEO INDIAN n=2 14 13 12 GREAT BASIN STEMMED n=21 z =?10 1O t I LITTLE LAKE n=12 8 9-------------- GYP-SUM n=4 ----- 3300B.P. 8 7 ELKO SERIES n=22 6 1350B.P. 6 - - ~-l HUMBOLDT n=8............P 5 -- lROSE SPRING n=20 --..---------------- 650 B.P. 4 SARATOGA n---=6 3 2 DESERT SERIES n=12 * MEAN ONE STANDARD DEVIATION Figure 19. Coso Obsidian Hydration Values on Projectile Points from the Volcanic Field. 83 Table 17. Frequencies of Coso Obsidian Projectile Points by Type and Period. Coso Hyd. Desert Saratoga Rose Humboldt Elko Gypsum Elko Little Great Basin Period Range (A) Series Spring Spring Basal-notch Thin Thick Contracting Side-notch Lake Stemmed Total Marana < 4.2 9 - 3 1 - - - - - - 13 Haiwee 4.2-5.65 3 5 12 1 - - - - - 1 22 Late Newberry 5.65-7.3 1 1 6 5 7 1 - - 1 - 22 Middle Newberry 7.3-7.9 - - - 2 2 - 1 1 - - 6 Early Newberry 7.9-8.7 - - - - 1 1 1 - - 1 4 Little Lake 8.7-10.6 - - - - 2 1 - 1 4 4 12 Early > 10.6 - - - - 2 6 2 - 8 16 34 Total 13 6 21 9 14 9 4 2 13 22 113 hydration rim bands, while those with a maximal thickness equal to or exceeding 6.5 mm showed a strong tendency to display substantially wider hydration bands, quite comparable withi the Little Lake points. Further, as shown in Figure 20 and Table 16, the temporal overlap of Thin Elko points compared to either Thick Elko or Little Lake points, using one standard deviation from the sample mean, is minimal. Thin Elko points are concluded to have a restrictive temporal range, calculating to ca. 3200-1700 B.P. (using one standard deviation) using Basgall's Coso hydration rate, which makes them a good chronological indicator. On the basis of this attribute, 71% of the Thin Elko points fall within our Late, Middle, and Early Newberry periods, collectively; while 77% of the Thick Elko points fall within our Little Lake and Early periods. It is worth noting, too, that the hydration rim values on Thick Elko and Little Lake points from the Volcanic Field are comparable to values reported for a large sample of points from the Stahl site (Figure 20), included in that sample here are 'Corner-notched, tangs visible, notched base" (Meighan 1981:Figure la-4), "Notched stem, open sides" (Meighan 1981:Figure 2c-t), and "Side notches, slightly concave base without basal notch" (Meighan 1981:Figure 3a-d) variants. The Thick Elko and Little Lake points from the Volcanic Field, however, on average have a higher mean hydration value than comparable forms at the Stahl Site (Figure 20). To this point we have shown that hydration data from Desert Side-notched/Cottonwood, Saratoga Spring/Rose Spring, Humboldt Basal-notched, and Thin Elko points from the Volcanic Field correlate well with the chronological scheme for the project area; and are largely consistent with the point sequence applied throughout the westem Great Basin. Our data suggest, however, that Thick Elko, Little Lake, and Great Basin Stemmed points were used contemporaneously throughout the Little Lake and Early periods. The range in values for each of these latter three types is quite large, paralleing the charactristic of the cluster samples from Early period deposits (compare Figure 18). The higher incidence of im asa bly diffse hydration (D)H) and variable width (VW) bands on Great Basin Stemmed points (recorded on 21%), compared to their lower incidence on Thick Elko (15%) and Little Lake (13%) points might signify the greater tiquity of Great Basin Stemmed points, but this is at present only conjecture. Alternatively, one could argue th hydration rims have poor chronological resolution at this early date. Only two Coso obsidian Paleoindian concave-based points have been recovered, precluding assessment of the temporal span, though both had quite wide hydration rim bands (13.4 and 21.2 I). Table 18 is presented to underscore the importance of Effective Hydration Temperature (EHT) on the rim formaton rate of Coso glass. Rim values recorded on Coso projectile points from six other projects in the westem Great Basin are mmrized here; geochemical XRF analysis has bee conductd on all specimens included in this tabulaton. Fort Irwin, located north of Barstow, CA, currently has the highest EHT factor, having a mean annual temeratre higher than the other areas. The Kern Plateau, at the south end of the Sierra Nevada Range, has the lowest EHT factor, being substantially cooler than the other areas listed. The Lubkin Creek site (INY-30), apro tely six km south of Lone Pine, has an EHT value in to Fort Irwin and the Kern Plateau, while the EHT in the Volcanic Field and the Stahl Site at Little Lake are likely to be quite similar, falling between the values for INY-30 and Fort Irwin. Contel project sites are adjacent to U.S. Highway 395 between Big Pine and Little e, through Owens and Rose valleys (Delacorte and McGuire 1993), indicating that the EHT value from the north end is comparable to that for INY-30, while the EHT for the south end is similar to the Volcanic Field and the Stahl Site. Put simply, listed from high to low EHT values (hot to cold) the areas are: Fort Irwin, the 84 21 20 19 18 17 16 15 14 13 12 LITTLE LAKE n=12 11. 6000 B.P. z U THICK ELKO n=8 9 -9STAHL SITE n=35 3300 B.P. 8 - -- -- -- -- -- -- -- -- -- -3300 B.P 7 THIN ELKO n=12 6 - - ---------- 1350B.P. 5 - -- -------- 650B.P. 4 3 2 * MEAN ONE STANDARD DEVIATION Figure 20. Coso Obsidian Hydration Values on Elko and Little Lake Points (Stahl Site Sample Based on Meighan 1981; All Others are from the Volcanic Field). 85 Table 18. Hydration Data on Coso Obsidian Projectile Points from Various Projects. n mean (A) s.d. range (/A) Desert Series Fort Irwin 5 3.2 1.5 1.5-5.0 Coso 12 3.0 1.2 1.34.7 Stahl Site 2 3.2 2.1 1.74.6 Contelb 12 3.1 1.9 1.3-5.9 INY-30c 44 3.0 0.9 0.7-5.2 Kern Plateau 40 2.0 0.7 1.04.7 Rosegate Series Fort Irwin' 6 4.7 1.1 3.5-6.1 Coso 20 5.2 0.8 3.6-6.9 Contelb 11 5.1 1.2 4.0-7.9 INY-30c 13 4.1 0.8 2.6-5.4 Kem Plateau 20 2.8 0.8 1.04.0 Humboldt Basal-notched Fort Irwin (none) Coso 8 6.3 1.0 4.8-7.7 Contel 2 6.2 0.4 5.9-6.5 INY-30 12 5.5 0.7 4.7-6.9 Kern Plateaud d d d d Elko Series: Thin < 6.5 mm maximal thickness; thick are > 6.5 mm maximal thickness. thin - Coso 12 7.4 1.0 6.0-9.3 dthin - Contel 3 7.1 1.3 6.3-8.6 thin- INY-30 2 6.6 0.4 6.3-6.8 thin - Kem Plateau 2 5.4 1.5 4.3-6.4 thick - Coso 8 12.3 3.3 8.7-18.9 thick - Contel 2 8.1 0.2 7.9-8.2 thick - INY-30 (none) thick - Kern Plateau (none) Little Lake/Pinto/ Gatecliff Split-stem Series Fort Irwin 4 12.5 1.5 11.1-14.1 Coso 12 14.2 4.3 9.1-21.5 Stahl Site 35 11.1 2.5 8.1-17.3 INY-30 1 14.0 - - Kern Plateau 2 9.9 1.1 9.1-10.7 Great Basin Stemmed Fort Irwin 11 13.3 2.5 10.0-16.9 Coso 21 12.9 2.7 8.7-17.8 INY-30 (Lake Mohave) 1 11.8 - Concave-based Coso 2 17.25 5.4 13.4-21.1 NOTE: Fort Irwin data derived from Basgall 1991-.51; ' - excluded a value of 2.7 for a Rosegate and 5.4 and 16.8 values for Pinto items. Stahl site data derived from Meighan 1981; Desert series items correspond to Meighan 1981:Figure 4h and j, Little Lake/Pinto/Gatecliff items correspond to Meighan 1981:Figure la-l, Figure 2c-t, and Figure 3a-d. Contel data derived from Delacorte and McGuire 1993; b - excluded values of 7.2 and 7.3 from Desert series items, and a value of 9.4 on a Rose Spring item. INY-30 data derived from Basgall and McGuire 1988; ? - excluded multiple rim values of 5.8/8.5 and 5.2/9.3 on two Desert series items, and multiple rim values of 2.6/7.0/7.5 on a Rose Spring item; d - Kern Plateau data was intentionally omitted because of confusion surrounding Sierran Concave-base forms. 86 Volcanic Field, and Stahl site; the Contel project area; INY-30; and the Kern Plateau. Though the mean hydration values from one area to the next for any particular series do not perfectly correspond to the temperature gradient, the trend is strongly supported. Where similar point styles are represented at both Fort Irwin and the Kern Plateau, the sample from the latter area consistently produced a narrower mean hydration rim value than was recorded for the Fort win sample - the matude of the difference icreasingrom 1.2 for the Desert series samples, to 1.9 , for the Rosegate series, and to 2.6 i for the Little Lake/Pinto/Gatecliff Split-stem series. Comparing samples from the Volcanic Field to INY-30, samples from the latter environment, with the exception of the Desert series, have consistently lower mean values than the corresponding sample from the Volcanic Field, generally averaging between 0.8 to 1.2 . narrower. As indicated in Tables 11 and 19, projectile points were recovered from comparatively few component areas. Had temporal assignment relied on projectile points very few areas (only 18) would have been dated; and a high propotion of those would likely have been considered multi-period accumulations (Table 19). Only two of the six points from Marana period deposits are Desert series forms, the reder are Rose Spring, Great Basin Stemmed (Silver Lake vaiant), Huniboldt Concave-base, and an undifferentiated arrow-sized fragment. Ten of the 19 points recovered from Newberry componets are styles other than Gypsum or Thin Elko: two Thick Elkos, three Little T ake, one Great Basin Stemmed, three leaf-shaped points, and a Humboldt Concave-base. On the other hand, Desert series points are always from deposits concluded to be Marana period; four of the five Rose Spring/Saratoga Spring points are from Haiwee period deposits; both Gypsum and all Thin Elkos are from Newberry period deposits; and three of five Great Basin Stemmed items are from Early deposits. In sum, projectile points are not common in the Volcanic Field, but hydration data from them corroborate the temporal placement of the point types. OTHIER TEMPORALLY DIAGNOSTIC ITEMS The remaining tporally diagnostic items recovered from project sites are limited to Olivella shell beads found singly at three sites, a single lass tade bead at one site and nearly 250 from another site, and comparatively few Owens Valley Brown Ware sherds at another (Table 9). Each of the Olivella beads is a complete split specimen,, two of which are punched, the other drilled. These are classified as Types C2 or Types Dla (Bennyhoff and Hughes 1987), and are tentatively dated to the Late Middle and Late prehistoric intervals in the Great Basin, though their chronological significance is weakly documented. One of both variants was collected from Late Newberry period deposits, reinforcing the temporal placement for Type C2, but suggesting that Type Dla beads occur during the "Middle" period. Three varieties of glass wade beads were recovered, the most numerous (n=27) being blue monochrome, multi-sided tubular beads with chopped ends (Class Dtum Type lc, Ross 1993:Appendix V). This type is fairly ommon tughout the Desert West in ptohistoric and early historic times. By comparison, only a few of those co nly and eoneously refred to as Russian" trade beads were recovered; these being similar to the previous pe, but having ground facets (n= 15; Class Dtum Type If, Ros 1993:Appendix V). he final variety coresponds to Ross' (1993:Appendix V) Class Dtfp Type IVa; they are small red-on-white cylinders shaped with a hot-tumble finish (n= 10). This latter type has also been referred to as "cornaline d'Aleppo" or "Hudson's Bay Company" beads (e.g., Jenkins 1975; Mille 1975). The first variety descnrbed above generally dates to the early to mid-1800s; the second variety is slightly more recent, typically post-dting 1860; and the final variety is dated ca. 1820-1860.- Al of the tade beads at INY-1906 are from a Marana component area (H) where a very recent radiocarbon date was obtained, as were Desert series points, and quite narrow hydration rims. All 41 pieces of pottery were recovered from INY-4329. They consist of six rim fragments, 34 wall f n, and one basal fragmt deriving from Owens Valley Brown Ware vessels. Regional data suggest that this pottry is a comparatively late phenomenon, dating within the past 700 years (Basgall and McGuire 1988; Bet 1989). To date, moderate quantiies of Owens Valley Brown Ware pottery have been found at only a few other sites in and adjacent to the Volcanic Field (fewer than 20 pieces each from 13 sites, and over 200 pieces from one site on Coso Wash). Thirty-one of the pieces from INY-4329 were recover from Marana-age Component W, where an arrow-sized projectile point was also collected, and a single hydration rim value of 3.1 was obtained, lezling additional surto t his late emporal placement. The remainder are from a residual area (E), adjacent to Component D, another Marana-age deposit identified at the site. 87 Desert Rose S Series Saratol Table 19. Diagnostic Projectile Points in Components. pring/and Humboldt Thin Thick Lil ga Spring Basal-notched Gypsum Elko Elko La lde Great Basin ake Stemmed Marana 1816 C 1906 H 4267 A 4329 D 4329 W 2 subtotal 2 Haiwee 3012 C' - 4239 A 4329 O subtotal Late Newberry 1816 A 1824 MainF - 1984 A 2103 A 4244 (all) - subtotal - Eary Newberry 3004/5 A 4329 K subtotal Early 1984 D 3300 A 4331 B subtotal 1 1 1 2 1 4 1 2 1 I 1 1 1 - - 1 - - 2 1 - 1 - 1 1 - 1 1 1 2 6 I - 1 - 2 - 1 - 4 1 1 3 2 2 1 1 10 1 1 - 1 3 - - 1 4 - 1 - - 1 - - 1 - - 1 3 3 6 - 3 3 9 - 1 - 1 - 1 - 3 TOTAL 2 5 2 7 3 2 5 6 32 SUMARY Investigations at the 34 sites led to the discovery of 21 spatially discrete Early period concentrations of material at 12 sites; seven Little Lake period component areas at as many sites, seven Early Newberry period components at six sites, three Middle Newberry components at one site; 23 Late Newberry period concentrations at 16 sites, eight Haiwee period components at as many sites, and 26 Marana period component areas at 13 sites (Table 11). Of the 6232 flaked stone artifacts recovered from project sites, 3594 are within component areas, as are 359 of the 652 pieces of ground stone, 155 of the 311 cobble-cores, three of the four abraders, five of seven anvils, 31 of 41 sherds, all but two of the beads, and the single incised stone. In the subsequent chapters, assemblage characterizations for the different time periods are based on the artifacts found in these well-dated deposits. 88 Other Total 1 4 I 2 2 CHAPTER 6: LITHIC PRODUCTION PATTERNS This chapter presents period-specific flaked stone assemblage characterizations, organized from early to most recent. Because reduction activities greatly differed between quany and off-quarry locations, they are considered separately within each period. We have also focused on attributes shared by components of the same age (Table 11), rather than seeldng out variability between them. For this reason, emphasis is given to composite assemblages created by combining materials across contemporaneous component areas. The chapter concludes with a discussion of diachronic trends in obsidian exploitation practices in the Volcanic Field. EARLY PERIOD QUARRY ASSEMBLAGES Early period areas occur primarily at lag quarries (n= 16), of which three (INY-2103H, -3299A, -4320C) contain materials from both Little Lake and Early period occupations, while one (INY-3004/51) contains an Early flaked stone assemblage associated with a Haiwee period ground and battered stone assemblage (Table 20). The combined chipped stone assemblage is dominated by cores (52.2%), with bifaces (inclusive of uniface-B's and flake blanks) maidng a significant secondary contribution (37.4%). Other artifacts occur in persistently low frequencies, including flake tools (5.7%), unifaces (3.5%), and formed flake tools and projectile points ( < 1.0% each). Cores. Cores are the "type" artifact of Early period quarries, found at nearly every Early lag quarry, but particularly abundant at INY-3300. Similarly high proportions of bifacial, non-patterned, and unidirectionally flaked cores were recovered, with a substantial nmber of bidirectional specimens also found (Table 20). The vast majority (85.9%) were discarded whole. Size differs little between these (Table 21), indicating that one core form does not give way to another with subsequent reduction, excluding cobble/chunk tests. Whole cobble/chunks tests have an average weight of 218.4 g, while average weight of the other forms ranges from 137.7 to 188.5 g. Wide-shouldered slightly outumber triangular shapes among bidirectional and non-patterned cores, while the reverse is true among unidirectonal cores. The two shapes are equally represented among bifacial cores; blocky-rectangular shapes are best represented among non-patterned cores. Bifaces. Bifaces were recovered from nearly each Early lag quarry (Table 20). Stage 2 forms are most numerous in virtually every one of these contexts, accounting for 72.4 % of all bifaces. Stage 1 bifaces follow at 14.7%, and Stage 3 at 11.3%. As indicated in Table 22, size tends to decrease as reduction increases, with Stage Is averaging 78 x 52 x 25 mm (86.9 g), and Stage 2s 73 x 48 x 19 mm (64.2 g). Stage 3s are longer, but narrower and thinner than preceding forms, averaging 83 x 45 x 15 mm. Over half of the bifaces discarded at the lag quarries are whole (54.7%), most frequent among Stage Is (79.7%) and declining to 45.1% among Stage 3s (Table 23). Most whole bifaces are wide-shouldered (33.9%) and tiangular (29.1%) shaped; the most common cause for discard is a poor cross-section (divots or high-spots, 32.9%), followed by human error during manufacture (31.4%), and structural flaws in the stone (18.0%, Table 23). Projectile Points and Biface/Points. The rem g bifacial implements are limited to a few projectile points and a biface/point fragment. Single projectile points were found in only three of 16 Early lag quarry deposits (Table 20). Eah is an Early type. The point frm INY-1984D is a complete basalt Silver Lake; the one from INY- 3300A is a complete Lake Mohave made of cryptocrystalline material; and the one from INY-4331B is a proximal fragment of a Lake Mohave (Plate 7j) made of Sugarloaf obsidian. The single biface/point is a medial fragment of an intermittently pressure-flaked obsidian specimen from INY-3004/51. Unifaces. A low number of unifaces occunred at the majority of the Early lag quarries, with a total of 44 recovered (Table 20). Averaging 67 x 48 x 20 mm (69.0 g), the unifaces are comparable in size to discarded Stage 2 bifaces from the same contexts (compare Tables 22 and 24), and like the bifaces, they are mostly discarded whole (79.5%). Rather than representing a desired separate artifact type, the size, condition, and shape of the unifaces suggest they represent a particula kind of failure in the biface production trajectory that resulted from using strildng angles too steep to effectively thin the base implement. Consequently, the blank remained usuitably thick, and surmounting this problem to produce a regularized cross-section would have required reducing the item's length and/or width to a degree apparently judged unsuitable. 89 C 4 *q eq C@ % N 4 I -( ''' I II eyj 0 q* CY ,iql o a U m*0 s a eq 'n C * eq eq - eq eq * 00  eq 0%0eq Sn 0 1SSn 0'0 eq * eq * '0- ro 0  Sn eq * 0 eq eq * e4'0 eq 'eq * Sn0%F * 0 ('4 0 Sn * '0 eq 'eq * eq 0% * ' 0 * - ' * '0 * ' Sn ('4- ("(' , , , fS ,x , . . ... ''0 , - m~ . - , , - ~~ , , -0SS - '-s'4- - - ~ . . - ~ , - * V' - 0- Sn** * - ' ' 5 - eq - . '0eq 'o - j;;gflS w 0 o *000%0o%0- .* - -(40 50 - S n'0' '0% * . . * . ' . . (. ., * ., eq 1" ~O .- .., 0% 'eqS' 'X WI oo 0 'i o.,, O ,o - o- -~ oo g,q pp 0 .- e . * eq- *. - w e ? '5.' 0 ^B * *1I II A -9 i I . I . I I< < I I I I . I I I # I I . I I I I - - I - en I I I o o CA a) 0 04 0 00 0 .S w 0) CIO.d .- 0 aL) Ca) CU2 a) 6q q4 -0o F-4 I' %0 I * c% - . I I 90 0 . . . * i ',0' . I I . I I . - I I I .... I Y ;" I ' I 3 3. i i . eq. - C i I i to I I I i I= * 1jaIfl ii =0 %O 8 8 - I V C4 p * Sn * 0 ' e 1 1 t i ! i i i on 0 0 .4- 0 0 o oS CU I , i 0 CU ......) a) o .0 .....a) 0 . ' ' ' ' . . . . 0 * .9 0 z # I I- - I I I I I a < I I < I . - I - I I I I I Table 21. Summary Metrics of Cores in Early Period Components. Bifacial length width thickness weight edge angle Bidirectional length width thickness weight edge angle Non-patterned length width thickness weight edge angle Unidirectional length width thickness weight edge angle Lag Ouarries n mean s.d. max. mn. 127 99 87 18 170 47 119 60 13 110 28 125 31 8 68 19 93 142.9 106.4 976.7 22.9 127 69 13 100 40 90 77 79 15 132 52 89 56 12 98 34 89 34 11 78 13 77 151.0 117.4 686.0 41.4 90 72 14 120 50 123 122 81 19 150 45 122 56 14 109 32 122 41 13 83 16 119 188.5 173.5 1171.9 34.8 123 76 13 110 40 118 12 80 15 133 43 9 57 12 100 26 110 33 9 68 17 100 137.7 85.7 562.8 37.8 118 68 14 105 40 Cobble and Chunk Test 39 length 39 77 24 152 41 width 39 59 21 150 29 thickness 39 42 14 96 18 weight 38 218.4 326.5 2009.0 30.6 edge angle 39 79 15 115 45 Not analyzed Grand Total 167 664 Off-Quarry Contexts n mean s.d. max. min. 9 8 64 9 74 45 9 42 9 61 32 9 26 6 40 18 8 68.2 32.4 144.4 40.9 9 71 11 80 45 9 7 62 14 83 39 9 46 8 62 35 9 27 7 37 16 7 87.1 38.0 155.4 44.6 9 76 6 85 65 9 9 66 16 89 42 9 45 8 58 35 9 34 6 44 25 9 101.0 36.2 142.0 39.3 9 80 19 110 50 8 5 63 7 74 55 5 53 7 62 44 7 28 8 42 14 5 91.8 41.0 173.2 65.6 8 61 16 80 40 5 5 81 28 136 61 5 49 19 86 33 5 39 21 81 22 5 232.4 308.2 847.5 51.0 5 71 7 80 60 42 NOTE: length, width, thickness in mm; weight in grams. Formed Flake Tools. Formed flake tools occur in very low frequencies at about one-third of the Early quarries (Table 20). This artifact type is, however, primarily restricted to Early period quarries. Of the 15 recovered from component areas, 12 were from Early lag quarries, and only three from later contexts. Averaging 61 x 50 x 16 mm, the formed flake tools at the Early lag quarries typically have modification along two or three steeply shaped edges (worked edge angles average 85.80) and the majority (n=9) are whole. The shape of edges usd shows little patteming, with a few examples of each with straight, convex, and concave-shaped worked edges. 91 Table 22. Summary Metrics of Bifaces, Uniface-B's and Flake Blanks in Early Period Components. Bifaces n mean s.d. max. min. 63 49 78 16 133 50 61 52 9 75 36 63 25 6 44 14 49 86.9 39.8 204.5 29.7 63 62 14 90 40 265 127 73 13 104 46 231 48 10 76 10 251 19 5 36 8 124 64.2 28.5 174.1 15.5 264 56 13 90 30 46 20 83 21 146 53 44 45 12 76 28 45 15 7 32 5 46 45 10 70 25 Stage 1 length width thickness weight edge angle Staze 2 length width thickness weight edge angle State 3 length width thickness weight edge angle StaRe 4 width 2 34 thickness 2 11 edge angle 2 40 Other 3 Not analyzed 7 Grand Total 386 Lag Quarries Uniface-B's n mean s.d. max. numin. 1 1 85 - - - 1 57 - - - 1 28 - - - 1 111.4 - - - 1 75 - - - 59 38 71 12 93 50 54 51 11 77 31 58 20 4 28 11 38 64.8 27.9 132.8 16.8 59 64 10 80 40 6 3 63 14 81 47 6 53 8 63 40 6 15 3 20 12 2 44.0 18.2 62.6 25.9 6 56 11 70 40 Flake Blanks n mean s.d. max. min. 5 5 81 17 106 63 5 58 15 86 42 5 25 4 30 20 5 107.1 76.3 256.2 51.6 5 52 8 60 40 15 11 75 22 139 43 15 52 14 92 28 15 19 6 30 11 12 69.5 80.7 329.3 12.5 15 51 11 70 35 1 1 90 - - - 1 57 - - 1 19 - - - 1 88.7 - - - 1 40 - - 1 35 33 2 12 9 - 40 40 2 68 2 23 Bifaces n mean s.d. max. min. 6 4 84 7 94 75 5 40 10 49 21 6 19 7 29 8 4 86.3 38.8 141.5 31.9 6 63 13 80 40 84 32 63 11 89 37 70 43 10 70 37 82 14 4 26 7 29 39.7 20.0 101.0 16.1 84 54 15 80 6 55 6 53 8 61 38 46 40 10 59 16 55 9 3 23 4 4 16.2 6.3 24.8 8.9 55 36 7 55 25 Off-Quarry Contexts Uniface-B's n mean s.d. max. min. 13 6 65 16 90 37 13 47 13 87 32 13 16 4 23 9 5 51.7 23.9 97.1 30.4 13 57 9 75 45 2 2 41 2 8 2 33 3 43 38 1 8 7 3 35 30 Flake Blanks n mean s.d. max. min. 3 2 55 3 58 52 3 55 4 60 50 3 21 3 25 19 2 41.7 0.9 42.6 40.7 3 62 5 65 55 3 2 57 5 62 52 3 31 2 33 29 3 7 1 7 6 2 10.7 0.7 11.4 10.0 3 30 - 30 30 Stare 4 1 width 1 31 thickness 1 9 edge angle 1 25 Other 4 Not analyzed 2 Grand Total 152 15 NOTE: length, width, thickness in mm.; weight in grams. 92 2 Stage 1 length width thickness weight edge angle StaRe 2 length width thickness weight edge angle Stage 3 length width thicmkness weight edge angle 6 Table 23. Summary Attributes of Bifaces in Early Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeterminate All Lag Quarries Off-Quarry Contexts Lag Quarries Off-Quarry Contexts Lag Quarries Off-Quarry Contexts 79.7 66.7 51.0 36.0 45.1 10.0 20.0 N Total 54.7 254 468 26.9 46 171 Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Irregular Total % Total N 29.1 33.9 13.0 13.0 1.6 9.4 100.0 254 32.6 23.9 10.9 10.9 2.2 19.6 100.0 46 Cause of Reiection Human Error Structural Cross- Poor Outrepasse Flaw Section Planar Shape 18.0 19.3 32.9 8.2 3.6 5.8 31.4 6.0 54.4 7.0 Unknown Total % Total N 8.1 5.3 100.0 468 100.0 171 Simple lake Tools. Simpleflake tools were recovered from 75% of the Early lag quarries, but never in much abundance (Table 20). The majority (n=61, 84.7%) of simple flake tools at Early lag quarries are whole. They typically display modification along one or two edges that usually are straight (36.1 %) or convex-shaped (40.3%). Their low frequency, the high percentage found whole, and the unimpressive degree of wear on them, combine to suggest that most are edge-damaged flakes rather than functional tools. Table 24. Summary Metrics of Unifaces in Early Period Components. Lag Quarries n mean s.d. max. min. 44 24 32 35 24 67 48 20 69.0 14 14 6 47.3 89 79 37 198.0 39 23 10 13.1 Off-Quarry Contexts n mean s.d. max. min. 12 5 7 8 5 64 43 20 50.0 8 7 3 8.2 76 56 25 55.9 51 35 16 34.1 NOTE: length, width, thickness in mm.; weight in grams. Debitage. The intensity of stone tool manufacture varies greatly from one Early period lag quarry to the next, with minimal and maximal subsurface debitage densities of 17 and 1095/m3 (CU-1/4"), respectively (Table 20). At all but one Early period lag quarry, core manufacturing debris is either equal to or more common than biface man g debris. The degree to which core production prevails varies between component areas, but the combined average of core/flake debris and angular shatter accounts for 78.4% of the diagnostic percussion debitage (Table 25). Cortical flakes typically make up at least half of the diagnostic core-production debris, followed by simple interior flakes. The low frequency of complex and other interior percussion flake types signals minima concem with dressing cores, and little intent to produce cores of much sophisticated form. The small amount of 93 length width thickness weight Table 25. Debitage Summary Data for Early Period Components. Sample Surf. 1/4" 1/8" Total % Diag. Lag Quarries 1824 B 1984 C 1984 D 2103 H 3004/5 I 3004/5 J 3299 A 3299 B 3300 A 3300 B 4320 A 4320 B 4320 C 4327 C 4329 G 4331 B Off-Quarry Contexts 1984 B 4267 M 4324/H B 4329 L 4329 O 89 - 43 - 10 - 15 - 203 - 469 - 438 - 226 - 167 - 37 - 12 -100 - 810 - 83 352 287 - - 1167 47 22 201 82 119 808 632 99 12 1 13 530 15 84 3313 80 136 65 10 216 285 588 1246 858 266 49 13 113 1340 98 436 287 3313 1247 % Biface % Core % Reduct. Reduct. Shatter 35.3 38.5 20.0 18.5 46.0 72.3 59.1 57.9 48.2 75.0 100.0 47.8 39.3 36.7 49.8 64.1 16.3 41.9 75.0 8.0 50.0 30.0 9.2 2.6 6.4 8.7 3.1 25.0 23.1 25.9 5.1 30.6 46.5 59.2 75.9 73.8 25.0 60.0 50.0 42.5 64.1 40.5 71.2 63.8 58.6 52.8 30.8 66.7 60.1 58.3 47.0 31.5 18.1 21.2 Biface Reduction Early Middle Late 32.0 27.5 26.7 56.9 22.4 27.5 38.3 22.2 46.1 7.4 34.8 11.1 6.4 9.2 5.9 5.0 13.9 50.0 100.0 8.3 25.0 45.5 55.3 32.6 50.0 44.4 33.3 7.1 64.5 18.2 12.9 31.3 7.8 14.5 55.6 50.0 66.7 75.0 36.4 36.2 53.5 50.0 55.6 78.6 32.3 27.3 43.6 62.4 46.9 72.3 Core Reduction Conrt. S.Int. C.Int. Other 30.6 25.0 18.2 8.5 14.0 66.7 14.3 3.2 54.5 43.6 6.4 45.2 13.2 16.7 40.0 100.0 58.8 55.9 56.4 71.5 64.6 46.7 47.4 25.0 19.4 75.0 61.9 21.6 53.5 27.5 64.8 66.7 60.0 29.4 15.5 34.9 19.1 28.4 44.0 15.8 25.0 58.3 11.3 28.6 36.3 24.1 28.6 19.8 8.3 11.8 22.6 7.6 9.2 5.0 5.3 10.5 25.0 19.4 12.6 19.6 22.4 22.4 12.6 NOTE: Surf. - surface; Diag. - diagnostic; complex interior. Reduct. - reduction; Cort. - cortical; S.Int. - simple interior; C.Int. - biface manufachre at took place at Early period lag quarries typically represents earlier stages of manufacture. Late outnumber early and middle stage thinning flakes at only two component areas - both being areas with minimal debitage samples. EARLY PERIOD OFF-QUARRY ASSEMBLAGES Five Early period components are in off-quany contexts (Table 20), and two of these contain a mixture of cts dating from other periods: Early and Early Newberry period materials are mixed together at INY-1984B, Early and Haiwee period chipped stone are mixed together at INY43290. In contrast to Early quarries, the flaked stone assemblage from off-quarry components is dominated by bifaces (62.0% including uniface-B's and flake blanks) followed distantly by flake tools and cores (17.9% and 15.0%, respectively, Table 20). Unifaces and point fragments are rare. Bifaces. Bifaces were recovered from each of the five Early off-quarry components (Table 20). The cumulative biface assemblage is doated by Stage 2 specimens (58.5%), but unlike the Early quarries, Stage 1 items are rare (3.5%), and Stage 3 items occur in significant quantities (35.1%). Comparatively few whole bifaces were discarded at secondary reduction locations (26.9%, Table 23), made up mostly of Stage ls and 2s. Whole bifaces tend to be triangular (32.6%) or wide-shouldered shapes (23.9%, Table 23). Over half (54.4%) were discarded due to human eor during manufacture, followed distantly by structural flaws (19.3%). Bifaces from secondary reduction locations are considerably smaller than their counterparts at lag quarries. Whole Stage 2 bifaces at the early lag deposits, for example, weigh on average 64.2 g, while ones in Early off-quarry contexts average 39.7 g (Table 22). 94 8.3 6.0 1.2 0.2 1.9 4.0 26.3 25.0 2.8 1.1 9.5 22.5 21.4 2.7 Projectile retie poPoints and Bface/Points. Projectile points and bifacepot fragments are rare in Early period off-quarry components: iNY43290 produced a nearly complete Saratoga Spring point made of Coso obsidian (Plate lq), clearly associated with later use of the site. Both biface/point fragments were found at INY-1984B: one is a cryptocrystalline distal tip weighing 8.3 g, showing a slight shoulder indicating it probably derived from a stemmed projectile point; the other is a medial fragment, weighing 4.0 g, of an apparent dart-sized obsidian point. Unifaces. A few unifaces occurred in each Early off-quarry component (Table 20). These artifacts are similar in size to co-associated Stage 2 bifaces, averaging 64 x 43 x 20 mm, with an average weight of 50.0 g (Table 24). Some 42% of the unifaces are whole, presumably discarded due to an inability to effectively thin the base item. Cores. Cores are much less abundant at Early period off-quarry components than at Early period quarries. However, the relative propordions of various core types is similar between the two contexts, with near equal amounts of bifacial, bidirectional, non-patterned and unidirectional cores found (Table 20). The majority (81.0%) were discarded whole. Cobble/chunk tests are the largest form, averaging 232.4 g; others are much smaller, averaging 68.2-101.0 g, depending on type (Table 21). The ones from off-quarrty components are roughly half the weight of those at the Early period quarries (Table 21), and the more extensive reduction has altered them such that near equal amounts of triangular, rhomboidal, and rounded-elongate forms occur. Simple Flake Tools. Simple flake tools are present at all but one of the Early off-quarry components. About two-thirds (n=34, 68.0%) of these are whole (Table 26). Typically they display one or two modified edges, most commonly straight (38.0%) or convex-shaped (36.0%). Debitage. Debitage densities in Early period off- quarry component areas are about three times as high as at the qualrries. Densities are, however, just as variable, rangng from 138 to 2346 pieces/m3 (CU-1/4"; Table 20). Site INY-4324/H-B produced the extraordinarily high density; however, densities at the three other loci with subsurface samples are < 740 pieces/. In these contexts, biface production tends to be only slightly more prevalent than core reduction, onting anywhere from 46.5% to 75.9% of the diagnostic percussion debris (Table 25). At two components, middle e ifa and late stage theinning flakes are equally represented, and at the other two, middle stage thinning flakes heavily dominate. Debitage from core production typically contains a fairly high amount of complex interior flakes (between 12.6-22.4% of the diagnostic debris), and in two locations, rectangular flakes are particularly abundant, constituting about 22% of the core-reduction debris. The high incidence of interior and rectangular percussion flakes suggests the occasional use of a fairly systematic (i.e., patterned) reduction strategy and a preference for linear flakes. In general, a variety of reduction strategies are represented in Early period off- quarry Tcom s, with biface and core-reduction activities well represented. Bifacial blanks and preforms were both produced with some frequency, while the cores imported from the lag quafries were worked into more patterned forms at contemporaneous off-quarry locations. EARLY PERIOD DISCUSSION Early period occupations occur pririly at lag quarries typically having low qntities of stone, which were worked in moderate levels. A wide assortment of artifacts types were made at these components, with about 50% cores and 38% bifaces, and small quantities of unifaces, flake tools, projectile points, and formed flake tools. Formed flake tools occur nearly exclusively in Early period lag quarries. Cores are of similar size regardless of form, and near equal quanites of bifacial, non-patterned, and unidirectional ones occur. Stage 2 forms dominate the bifaces (72%), with substantally fewer but near equal numbers of earlier and later stage items. Debitage reflects emhases on ini reduction, and core over biface manufacture. Of the few associated projectile points, two of the four are of non-loca mateials. A high percentage of the artifacts were discarded whole (86% cores, 85% flake tools, 80% unifaces, and 54% bifaces), in roughed-out forms judged unacceptable for one reason or another. In moving to the Early period off-quarry compons, reduction activities shift to an emphasis on bifaces. This is reflected in the artifact assemblages as well as the debitage samples. Bifaces account for 62.0% of the artifacts, with flake tools making a conibution at 17.9%, and cores king third (15.0%). The majority of bifaces are blanks (Stage 2, 58.5%) followed by preforms (Stage 3, 35.1 %), and the blanks average about half to two-thirds the size of their quarry counterparts. About 75% of the bifaces are fragmented, also reflective of their 95 Table 26. Summary Attributes of Simple Flake Tools at Off-Quarry Contexts by Period. Average Number Number of Number of Period Total Whole Modified Edges Modified Edges Early 50 34 70 1.4 Little Lake 4 2 5 1.3 Early Newberry 26 18 33 1.3 Middle Newberry 12 8 16 1.3 Late Newberry 79 48 123 1.6 Haiwee 11 6 16 1.5 Marana 5 2 6 1.2 Modified Edge Shape Period Straight Convex Concave Spokeshave Beaked S-Shaped Total Early 19 18 6 2 1 4 50 Little Lake 1 2 - - - 1 4 Early Newberry 7 11 2 - - 6 26 Middle Newberry 5 5 1 - - 1 12 Late Newberry 32 25 9 4 6 3 79 Haiwee 6 3 - - 2 - 11 Marana 1 1 2 - - 1 5 Average Metrics Period N Length Width Thickness Weight Edge Angle (0) Early 34 52 39 10 23.6 35 Little Lake 2 53 33 5 7.1 28 Early Newberry 18 59 41 12 31.1 36 Middle Newberry 8 45 41 8 11.4 29 Late Newberry 48 50 39 9 24.6 31 Haiwee 6 54 39 7 12.3 40 Marana 2 58 39 8 16.6 37 Average Length of Modified Edges by Shape Period N Straight Convex Concave Spokeshave Beaked S-Shaped Early 50 28 38 33 8 7 41 Little Lake 4 24 - - - - 35 Early Newberry 26 39 42 43 - - 49 Middle Newberry 12 22 44 - - - 42 Late Newberry 79 33 43 30 12 37 45 Haiwee 11 29 19 - - 30 - Marana 5 29 - 31 - - 51 NOTE: Length, width, and thickness in mm.; weight in grams. more advanced state of reduction. Cores are few in number, but like the quarries, near equal numbers of assorted types exist, and a high proportion (81%) were discarded whole. They are, however, smaller than those at quarries, at about half their weight. Debitage is also indicative of the cores more advanced state of reduction, with a definite preference shown for linear, blade-like flakes. High proportions of flake tools (68.0%) were discarded whole, 96 showing no inclination to curate such tools in this obsidian rich zone. A few projectile points and unifaces complete the flaked stone assemblage. LITTLE LAKE PERIOD QUARRY ASSEMBLAGES Litdtle Lake period component areas tend to occur at lag quarries (Table 27). Five Little Lake period lag quarries exist of which three contain Early period materials intermixed. Their combined flaked stone assemblages contain a high proorton of cores (48.2%) and bifaces (43.3%, inclusive of uniface-B's and flake blanks), while flake tools, unifaces, and formed flake tools contribute no more than 5% each. Site INY-4319 yielded a more ssantial assemblage than the other four locations, and in large part the following characterizations conform to this component. Cores. Variable quanties of cores were recovered from each of the five components (Table 27). All core forms are well represented with bifacial and unidirectional cores slightly more numerous than bidirectional or non- patterned ones. Size differs little between these four forms, with the average weight of whole ones per class ranging from 153.0 to 189.8 g (Table 28). Cobble/chunk tests are, not surprisingly, much larger, averaging 610.2 g. Triangular and wide-shouldered shapes are most common, though other forms are well represented. Bffaces. Bifaces are relatively common at four of the Little Lake quarries (Table 27). Most are Stage 2 (67.5%) followed by Stage 1 (21.5%), then Stage 3 (11.0%) forms. Slight reductions in all dimensions and edge angle acuteness track through the reduction stages. Stage ls average 80 x 51 x 23 mm (80.8 g); Stage 2s 76 x 47 x 18 mm (63.8 g); and Stage 3s average 40 mm wide and 9 mm thick (Table 29). A high proportion are whole (Table 30), most frequent among Stage ls (80.5%), declining to 45.5% among Stage 3s. Most whole bifaces are triangular (34.7%) or wide-shouldered (28.2%). The most common reason for discard is human error (31.7%) or poor cross-section (23.1%), though 25.6% displayed no apparent reason for discard (i.e., it was not obvious why reduction was discontinued on many of the whole ones). Projectile Points and Biface/Points. Only one biface/point was found at one Little Lake quarry, a squared- off base of a large (11.3 g), broad-bladed (35 mm) pressure-worked fragment. Workmanship, robustness, and general shape suggest affinity with Paleoindian points. Hydration analysis recorded an immeasurably diffuse hydration band. Unifaces. Only one of the five Little Lake lag quarries had a sizeable number of unifaces, and few were found at two others (Table 27). These items appear to be rejected, unifacially worked items intended as Stage 2 bifaces. All but one of the 20 unifaces are whole, and they generally match the size of co-associated Stage 2 uniface-B's (compare Tables 29 and 31): Stage 2 muniface-B's average 23 mm thick, unifaces average 25 mm thick. The high incidence of discarded whole unifaces, many with poor cross-sections, suggests that a formidable error prompted discard early in the manufacturing sequence. Formed Flake Tools. The single obsidian formed flake tool from one Little Lake lag quarry is whole, measures 84 x 42 x 17 mm, and has a working edge angle of 860. It has two convex-shaped working edges, each about 50 mm long. Simple Flake Tools. Simple flake tools are present at three Little Lake quarries, but only in an appreciable number at one (Table 27). Nearly 95% are whole. Modified edges are primarily convex (77.8%), and typically only one edge per tool is modified. The low frequency of simple flake tools at Little Lake lag quarries, the high prportion found whole, and the minimal extent of edge attrition indicate that they are probably naturally damaged flakes. 97 Table 27. Assemblage Inventories in Little Lake Period Components. Flaked Stone Bifaces, Uniface-B's, and Flake Blanks Stage 1 Stage 2 Stage 3 Stage 4 Stage 5 Indeterminate Subtotal Projectile Points and Biface/Points Cores Non-Patterned Unidirectional Bidirectional Bifacial Other Not Analyzed Subtotal Flake Tools Unifaces Formed Flake Tools Drills Debitage ( "/m3) 2103 H* 3299 A* QL QL 7 24 3 4319 4320 C* 4322/H A Subtotal 4325 A 4330A Subtotal QL QL QL QL OQ OQ OQ 30 102 15 34 147 2 6 2 2 12 - - 1 1 1 8 9 9 10 1 37 - 2 - 4 - 1 405 435 17 28 18 21 10 63 157 15 15 111 5 1 1 3 10 1 4 3 2 43 135 22 - 2 9 202 - 1 4 4 4 8 20 1 34 42 32 43 11 63 225 18 20 1 60 34 2 3 1 31 9 1 33 12 5 41 46 - 1 1 1 1 1 1 2 1 1 5 4 1 1 1 2 1 1 6 4 2 20 239 Ground and Battered Stone Millingstones Thin Slab Block Subtotal Handstones Bifacial Unifacial Subtotal Mortars Pestles Misc. Ground Stone Cobble-Core Tools Anvils Other Materials Features Beads Sherds Incised Stone Miscellaneous Stone Modified Stone Pebble Bone Seeds 5 5 5 A 5 A 1 - 1 1 1 1 1 1 2 25 2 A 1 29 1 1 1 3 1 - 3 3 3 NOTE: * - dual period assignnts (see Table 10); QL - lag quarry; OQ - Off-quary; A - one item each reassigned to Residual. 98 Table 28. Summary Metrics of Cores in Little Lake Period Components. Lag Quarries Off-Quarry Contexts n mean s.d. max. min. n mean s.d. max. min. Bifacial 43 1 length 35 90 19 164 61 - - - - - width 43 60 13 99 28 1 57 - - - thickness 42 32 7 60 23 1 23 - - - weight 33 153.0 105.8 675.4 22.9 - - - - - edge angle 43 76 12 100 50 1 65 - - Bidirectional 32 2 length 30 82 17 114 57 2 85 1 86 84 width 32 59 12 105 44 2 61 - 61 61 thickness 32 33 11 67 18 2 31 1 31 30 weight 30 154.8 126.6 749.5 53.1 2 150.8 6.5 157.3 144.3 edge angle 32 75 16 105 55 2 83 3 85 80 Nonpattened 34 1 length 32 87 23 158 52 - - - - - width 33 58 12 95 35 - - - - - thickness 33 40 10 70 27 1 17 - - - weight 32 189.8 129.7 599.5 48.7 - - - - - edge angle 34 83 15 110 50 1 85 - - - Unidirectional 42 1 length 38 86 22 153 51 - - - width 41 59 14 100 31 1 88 - - - thickness 41 33 9 68 18 1 25 - - - weight 38 162.6 130.8 803.4 42.8 - - - - - edge angle 42 75 13 105 40 1 60 - - - Cobble and Chunk Test 11 1 length 11 105 27 165 66 - - - - - width 11 73 25 109 31 - - - - - thickness 11 55 23 103 13 1 (33) - - - weight 10 610.2 560.3 1941.5 38.0 - - - - - edge angle 11 89 15 110 70 1 100 - - - Not analyzed 63 - Grand Total 225 6 NOTE: length, width, thickness in mmn.; weight in grams. Debitage. Debitage densities suggest that comparatively little stone-tool reduction occurred at any Little Lake period lag quarry. Subsurface densities (CU-1/4") average only 209 pieces/m3, and never exceed 435 pieces/m3 (Table 27). Core reduction is the prevalent activity at each component, with core/flake debris and angular tater combining to account for no less than 76.9% of the diagnostic percussion debris at each (Table 32). Where adequ samples were recovered, more than half of the core reduction flakes are cortical pieces, and complex and other interior flakes are minimally represented. Of the small quantity of biface reduction debris, decreasing quantities of early reduction flakes, middle stage thinning flakes, and late stage thinning flakes are present. Typically though, late stage thinning flakes still account for between 12.2-25.0% of the biface reduction debris, indicating all biface percussion reduction stages occurred. 99 Table 29. Sunmuazy Metrics of Bifaces, Uniface-B's and Flake Blanks in Little Lake Period Components. Bifaces n mean s.d. max. min. 39 30 80 10 103 54 36 51 8 68 38 37 23 5 37 13 29 80.8 21.5 137.6 42.4 37 59 9 75 45 105 53 76 13 100 38 98 47 10 72 23 105 18 5 36 9 55 63.8 29.8 133.4 18.0 105 51 9 80 30 Sta2e 1 length width thickness weight edge angle Stane 2 length width thickness weight edp angle State 3 length width thickness weight edge angle 9 54 28 1 12 7 3 40 30 Lag Quarries Uniface-B's n mean s.d. max m. 4 4 71 50 187 64 4 56 7 66 46 4 23 3 27 20 4 84.0 21.7 120.3 63.5 4 48 6 55 40 23 17 102 17 124 51 19 54 10 76 29 21 23 6 33 9 17 91.1 45.1 196.4 18.0 21 58 12 85 40 2 1 83 - - - 2 59 10 68 49 2 17 3 20 14 1 62.6 - - - 2 43 3 45 40 Flake Blanks n mean s.d. max min. 7 6 71 19 100 43 7 48 11 60 28 7 17 5 26 9 6 55.2 43.7 143.7 12.5 7 43 12 65 30 9 8 72 11 90 54 9 42 12 57 16 9 14 3 19 9 8 41.7 20.7 88.7 14.8 9 38 3 45 35 Other Grand Total 155 Bifaces n mean L.d. max. mn. 1 1 71 - - - 1 27 - - 1 50 - - - Off-Quarry Contexts Uniface-B's n mean s.d. max. mm. Flake Blanks n mean s.d. max. min. Sta_ 2 31 length 2 66 3 68 63 width 15 50 8 64 36 thicess 31 14 4 24 6 wight 2 70.2 2.3 72.4 67.9 edgeangle 31 39 7 55 25 Stra3 12 width 7 41 10 60 31 thickness 12 9 2 11 5 edge angle 12 31 4 40 25 Grand Total 44 1 1 94 1 46 1 26 1 97.8 1 65 1 1 1 106 1 90 1 23 1 173.7 1 45 1 NOTE: length, width, thickness in mm.; weight in grams. LiTTLE LAKE PERIOD OFF-QUARRY ASSEMBLAGES Only two Little Lake period off-quany components were identified, and one of these (INY4330A) produced nearly 90% of their combined flaked stone assemblage (Table 27). Bifaces strongly dominate (78.0%), followed distantly by cores (10.2%) and flake tools (6.8%). Bifaces. Some 71.7% of the bifaces are Stage 2s, followed by 26.1% Stage 3s. Only one of the 46 bifaces is a Stage 1; only four are whole. Given the fragmented condition of bifaces found in these contexts, little can be inferred about their size or preferred morphologies. More than two-thirds of the bifaces were discarded due to human error during manufacture, followed distantly by structural flaws (19.6%; Table 30). 100 11 1 71 11 40 11 9 1 33.8 11 33 StaRe 1 width thickness edge age 2 18 29 Table 30. Summary Attributes of Bifaces in Little Lake Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeterminate All N Total Lag Quarries Off-Quarry Contexts 80.5 59.3 12.1 45.5 100.0 62.3 124 8.7 4 Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Lag Quarries Off-Quarry Contexts 34.7 28.2 50.0 17.7 50.0 10.5 1.6 Irregular Total % 7.3 100.0 - 100.0 Human Error Outrepasse Structural Flaw Cause of Rejection Cross- Poor Section Planar Shape Unknown Total % Total N Lag Quarries Off-Quarry Contexts 31.7 69.6 1.0 8.7 12.6 19.6 23.1 6.0 2.2 25.6 100.0 100.0 199 46 Projectile Points and Biface/Points. Only one biface/point fragment was found at one Little Lake off- quarry component. It is a distal tip too small (0.9 g) to determine whether it is part of a dart or arrow-sized point. Unifaces. One uniface was recovered from each Little Lake off-quarry context. Their low frequency is consistent with the notion that they are early stage manufacturing rejects. The one uniface that was whole is comparable in weight to the one whole uniface-B from these two areas (95.0 g and 97.8 g, respectively, compare Tables 31 and 29). Table 31. Summary Metrics of Unifaces in Little Lake Period Components. Lag Quarries mean s.d. 83 50 25 93.3 14 8 6 38.6 max. mm. 117 68 37 185.7 Off-Quarry Contexts n mean s.d. 2 59 37 15 43.6 2 1 1 1 max. mn. 64 7 70 57 61 - - - 11 - - - 95.0 - - - NOTE: length, width, thickness in mm.; weight in grams. Cores. The six cores recovered from the two component areas represent all types (Table 27). Most are isagmented, but the two whole ones are similar in size to their quarry counteiparts (Table 28), suggesting they were discarded in much the same condition as they entered the site. In light of the quantity of cores produced at Little Lake period lag quarries, their absence in contemporaneous off-quarry contexts is particularly noteworthy, suggesting that many cores were transported outside of the Volcanic Field for consumption/use elsewhere. 101 199 46 Total N 124 4 length width thickness weight n 20 19 20 20 19 Table 32. Debitage Summary Data for Little Lake Period Components. Sample % Biface % Core % Biface Reduction Core Reduction Surf. 1/4" 1/8" Total % Diag. Reduct Reduct Shatter Early Middle Late Cort S.Int C.Int Other Lag Quaries 2103 H - - - - - - - - - - - - - - 3299 A - 469 119 588 72.3 2.6 40.5 56.9 45.5 36.4 18.2 56.4 34.9 7.6 1.2 4319 all - 184 159 343 60.3 19.8 42.0 38.2 60.9 26.8 12.2 62.1 12.6 18.4 6.9 4320 C - 12 1 13 100.0 23.1 30.8 46.1 33.3 - 66.7 25.0 25.0 25.0 25.0 4322/H A - 62 33 95 35.8 11.8 32.3 55.9 50.0 25.0 25.0 81.9 18.2 - - Off-Quarry Contexts 4325 A - - 20 20 5.0 100.0 - - 100.0 - - - - - - 4330 A - 672 414 1086 33.3 90.4 6.6 3.0 15.9 37.3 46.8 62.5 12.5 8.3 16.7 NOTE: Surf. - surface; Diag. - diagnostic; Reduct. - reduction; Cort. - cortical; S.Int. - simple interior; C.Int. - complex interior. Flake Tools. Only four flake tools were recovered from Little Lake period off-quarry components. The two whole ones average 53 x 33 x 5 mm (7.1 g; Table 26). Modified edge shape varies, with convex, straight, and s-shaped edges represented. Debitage. Debitage densities were low at both Little Lake period off-quarry contexts, equaling 20 pieces/m3 at one and 239/m3 (CU-1/4") at the other. Only INY4330A yielded a subsurface debitage sample large enough to warrant consideration, and here, 90.4% of the diagnostic pieces were from biface manufacture, with only 9.6% identified as core-reduction flakes or shatter (Table 32). Of the biface production debris, late stage inning flakes are most abudnt (46.8%), with decreasing quantities of middle (37.3%) and early stage (15.9%) thinning flakes recovered. While the proportions of bifaces by reduction stage indicate bifacial blank manufacture was prevalent, the debitage profile clearly indicates preform (Stage 3) manufacture was of greater incidence. Core reduction, whether measured by the number of cores or the relative proportion of core-reduction debitage, is poorly represented. LITLE LAKE PERIOD DISCUSSION Patterns established in the Early period continue essentially unchanged through the Little Lake period. Occupations occur primarily at lag quarries; assorted core forms dominte heir assemblages, followed by early stage bifaces; and high proportions of both are whole rejects. Debitage profiles are consistent with this emphasis on initial reduction of the lag cobbles into cores and Stage 1 and 2 bifaces. In moving to off-quarry contexts, reduction emphases shift to biface blank and preform manufture. Small quantities of flake tools, unifaces, projectile points, biface/points, and formed flake tools occur intermittently at both the quarry and off-quarry components. Cores are noticeably absent in off-quarry contexts, and core-reduction debris is also poorly represented, indicating that most cores were carried out of the Volcanic Field in the form made at the quarries. EARLY NEWBERRY PERIOD QUARRY ASSEMBLAGES Of the seven Early Newberry components, four occur at lag quarries (Table 33). These four quarries generated a collection dominated by bifaces (53.0%), followed distantly by cores (18.1%), and flake tools (14.1 %). Our characterization of Early Newberry quarries largely derives from materials at INY-3004/5A, which yielded 70.5% of the combined flaked stone artifact assemblage. The three other components offered comparatively meager assemblages. Cores. Cores are surprisingly scarce at Early Newberry lag quarries, with only small quantities at each component (Table 33). Of the 27 cores, unidirectional and non-pattemed forms co-dominate at 33.3% each. 102 Bifacial, bidirectional, and non-pattemed forms are similar in size, averaging between about 160-170 g (Table 34), while unidirectional ones are narrower and thinner, and, therefore, considerably lighter, averaging 110.0 g. Unidirectional cores generally take rounded-elongate shapes, while non-patterned ones are irregularly shaped. Table 33. Assemblage Inventories in Early Newberry Period Components. 1824 Main A 3004/5 A 4327 A 4331 A Subtotal 1824 Main Y 1984 B* 4329 K Subtotal QL QL QL QL QL OQ OQ OQ OQ Flaked Stone Bifaces, Uniface-B's, and Flake Blanks Stage 1 - - - 2 2 - - 8 8 Stage 2 6 23 1 4 34 - 43 54 97 Stage 3 - 34 - - 34 - 23 34 57 Stage 4 - 3 - - - 3 Stage 5- - - - - Indeterminate 1 - - - 1 - 2 22 24 Not Analyzed 1 4 - - 5 - - 1 1 Subtotal 8 64 1 6 79 - 68 119 187 Projectile Points and Biface/Points - 7 - - 7 - 2 3 5 Cores Non-Pattemed 2 1 - 6 9 - 3 - 3 Unidirectional 2 6 1 - 9 - 1 1 2 Bidirectional 1 1 1 1 4 - 2 2 4 Bifacial 1 1 1 1 4 - 3 2 5 Other - 1 - - 1 - - 6 6 Subtotal 6 10 3 8 27 - 9 11 20 Flake Tools 18 - 3 21 1 10 15 26 Unifacs 3 5 3 2 13 - 4 2 6 Formed Flake Tools - 1 - 1 2 - - - Drls - - - - . - Debitage ('/"/m3) 135 540 23 45 NS 138 4129 Ground and Battered Stone Millingstones Thin - 4 - - 4 - 4 1 5 Slab - 2 - - 2 - 6 - 6 Block - - - - - - Subtotal - 6 - - 6 - 10 1 11 Handstones Bifacial ... . . . . . . Unifacial - 2 - - 2 - - - - Subtotal - 2 - - 2 - - - - Mortars .- - -. Pestles - - - - - MiscellaneousGround Stone - 5 - - 5 - - 2 2 Cobble-Core Tools 5 2 2 - 9 - 4 8 12 Anvils - - - - - - - Other Materials Features - - - - - - - - - Beads - Sherds - - - - - - - - - Incised Stone - - - - - - - - - MiscaneousStone - 9 - - 9 . Modified Stone . .- - - Pebble .- - - - Bone - - - - - - 1 2 3 NOTE: * - dual period assignments (see Table 10); QL - lag quarry; OQ - off-quany; NS - no sample. 103 Table 34. Summunary Metrics of Cores in Early Newberry Period Components. Lag Ouarries Off-Quarry Contexts n mean s.d. max. minun. n mean s.d. max. min. Bffacial 4 5 length 2 112 1 113 111 4 61 10 71 45 width 4 63 7 72 53 5 47 7 57 37 thickness 4 34 15 58 19 5 26 3 30 22 weight 1 168.7 - - - 4 65.7 15.9 89.6 49.7 edge angle 4 66 10 80 55 5 54 11 71 41 Bidirectional 4 4 length 2 88 8 96 80 2 76 4 80 72 width 4 67 12 76 46 4 48 8 55 35 thickness 4 33 13 49 17 4 28 9 40 16 weight 2 158.6 24.8 183.3 133.8 1 83.6 - - - edge angle 4 69 16 85 45 4 64 15 80 40 Non-patterned 9 3 length 9 79 14 109 62 3 58 7 68 50 width 9 61 12 80 42 3 46 6 55 40 thickness 9 41 8 50 28 3 38 5 44 33 weight 8 172.4 51.7 178.3 114.2 3 102.7 28.1 142.0 78.2 edge angle 9 76 13 95 55 3 83 17 95 60 Unidirectional 9 2 length 7 84 12 104 70 1 39 - - - width 9 51 10 65 29 1 32 - - - thickness 9 28 4 34 21 2 32 - - - weight 7 110.0 45.9 180.5 45.0 1 46.2 - - - edge angle 9 74 8 90 60 2 70 - - - Cobble and Chunk Test 1 6 length 1 88 - - - 5 73 13 93 55 width 1 65 - - - 5 60 15 88 44 thickness 11 51 - - - 6 36 9 48 21 weight 1 354.4 - - - 5 173.9 82.5 318.3 68.3 edge angle 1 95 - - - 6 64 5 70 55 Grand Total 27 20 NOTE: length, width, thickness in mm.; weight in grams. Bifaces. Bifaces occurred at each of the four components, and are especially numerous at INY-3004/5A (Table 33). Stage 2 and Stage 3 bifaces co-dominate at 46.0% each. It should be noted, however, that only Stage is and 2s occur at thee component areas, and all Stage 3s occur at INY-3004/5A. As indicated in Table 35, size decreases as rduction stage increases, with whole Stage 2s averaging 79 x 42 x 15 mm (61.0 g), and Stage 3s 59 x 34 x 11 mm (22.1 g). Overall, 20.5% of the bifaces were discarded whole (Table 36), and of these, rounded- elongate (36.8%) and triangular shapes (21.1 %) are most frequent. Discard of nearly two-thirds of the bifaces is ascribed to human error during manufacture. Projectile Points and Biface/Points. The six projectile points and one biface/point recovered from Early Newberry lag quarries all came from INY-3004/5A. The points represent a variety of types, including one each Gypsum (Plate 6d), Elko (Plate 5n), Humboldt (Plate 3i), and Little Lake (Plate 61); and two leaf-shaped (Plate 104 Table 35. Summary Metrics of Bifaces, Uniface-B's and Flake Blanks in Early Newbeny Period Components. Bifaces n mean s.d. max. m. 2 1 107 2 58 2 29 1 149.7 2 63 6 63 52 2 31 28 3 65 60 StaRe 1 length width thickness weight edge angle Stane 2 length width thickness weight edge angle Sta~c3 length width thickness weight edge angle Sta~e4 length width thickness weigb edge angle Other Not analyzed Grand Total 1 4 73 Bifaces n mean s.d. mnax. min. 5 1 114 - - - 4 57 7 63 47 5 21 4 26 15 1 157.2 - - - 5 74 16 90 45 89 20 66 13 89 37 62 49 12 80 31 76 14 4 28 8 18 48.8 23.7 103.6 17.0 89 53 13 80 30 54 7 64 15 95 52 46 46 11 85 25 51 9 2 15 3 6 27.4 13.5 54.9 14.9 54 36 7 55 25 24 172 Lag Quarries Uniface-B's n mean s.d. max. min. 4 3 81 17 96 58 4 51 5 59 45 4 17 5 24 12 3 79.8 40.5 134.3 37.3 4 64 18 90 45 1 5 Off-Quarry Contexts Unifaac B's n mean s.d. max. mn. 6 3 56 14 72 37 6 49 17 87 34 6 17 3 23 13 2 36.5 6.1 42.6 30.4 6 53 7 65 45 2 2 47 2 9 2 33 9 56 38 1 10 8 3 35 30 8 Flake Blanks n mean s.d. max min. I 1 42 1 15 1 40 1 Flake Blanks n mean s.d. max. mm 3 2 67 6 ' 72 61 3 58 10 72 48 3 25 4 28 20 3 88.0 38.7 141.7 51.8 3 47 2 50 45 2 2 74 4 78 70 2 47 12 59 35 2 15 4 19 11 2 45.2 26.4 71.6 18.8 2 35 5 40 30 1 1 62 1 29 1 6 1 11.4 1 30 1 7 NOTE: length, width, thickness in mm.; weight in grams. 105 29 6 79 14 101 60 22 42 9 55 20 23 15 5 25 9 6 61.0 26.0 111.1 34.7 29 51 15 75 25 34 10 59 15 94 40 31 34 7 52 18 33 11 3 18 6 8 22.1 7.5 38.2 14.2 34 42 14 75 35 3 1 55 - - - 3 26 1 26 25 3 10 1 11 8 1 12.4 - - - 3 50 12 65 35 _Sta_e_1 length width thickness weigt edge angle StaRe 2 length width thickness weight edge ange Stan 3 lengt width thickness weigbt edge angle Other Not analyzed Grand Total Table 36. Summary Attributes of Bifaces in Early Newbeny Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeterminate All N Total Lag Quarries Off-Quarry Contexts Lag Quarries Off-Quarry Contexts Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Irregular Total% Total N 21.1 15.8 36.8 15.8 - 10.5 100.0 19 27.3 24.2 6.1 18.2 6.1 18.2 100.0 33 Cause of Rejection Human Structural Error Outrepasse Flaw Cross- Poor Section Planar Shape Unknown Total % Total N Lag Quarries Off-Quarry Contexts 65.4 11.5 56.5 11.3 2.6 18.3 7.7 6.5 1.3 2.7 11.5 100.0 4.8 100.0 9h and 9i). Each is whole or nearly complete. The Elko was geochemically identified as an unknown" obsidian; the rest have been acribed to various Coso subgroups. Hydration rim values on the Coso specimens range from 7.9 to -15.5 ; however, rim values on five of the six items range from 7.9-9.7 , suggesting that the "unknown" might simply be made from an aberrant piece of Coso glass. The single biface/point fragment at 3.9 g, is a distal tip of a dart-sized point. Unifaces. A few unifaces occurred at each Early Newberry period lag quarry (Table 33). They account for 8.7% of the cumlative flaked stone assemblage - the highest value they obtain in any period or context. Six of the 13 unifaces are whole. Averaging 59 x 44 x 18 mm (45.0 g. Table 37), they are intermediate in size to co- associated Stage 2 and 3 bifaces. They are also substantially smaller than the unifaces recovered from lag quarries used in other time periods. These factors combine to suggest that Early Newberry unifaces were functionally separate from the biface reduction trajectory. Formed Flake Tools. One formed flake tool was found in each of two Early Newberry lag quarries (Table 33). Both are broken (18.5 g and 34.8 g), and have steep (85-90?) working edges. One has a single convex-shaped modified edge, the other has two straight worked edges. Il i i i Table 37. Summary Metrics of Unifaces in Early Newberry Period Components. - J vsLU LA n mean s.d. 59 44 18 45.0 10 10 6 27.1 13 6 8 10 6 max. mn. 74 60 29 81.8 42 31 10 9.0 Off-Ouarr Contexts n mean s.d. max. min. 6 2 58 3 46 4 19 2 42.2 7 9 3 8.1 64 56 23 50.3 51 35 16 34.1 NOTE: length, width, thickness in mm.; weight in grams. 106 50.0 50.0 26.5 22.7 23.5 12.3 33.3 20.5 16 17.7 33 78 186 78 186 lengt width thickness weight Simple Flake Tools. In contrast to earlier lag quarries, flake tools are well represented at Early Newberry lag quanries, and they also tend to be more extensively modified. The average number of worked edges per piece is two. Shaped margis are most commonly straight (33.3%), followed by convex (28.7%) and concave (23.8%). As in preceding periods, a very high proportion (81.0%) are whole. The high number of worked edges per tool and the diversity of their edge shapes suggest that edge attrition is quite likely the result of use-related damage. All but three were recovered from INY-3004/5A, which in combination with other characteristics suggest use more as a campsite than as a lag quarry. Debitage. Debitage densities are quite low at three of the four Early Newberry lag quarry components (? 135/rn3, CU-1/4"), suggesting that comparatively little stone-tool reduction occurred at these locations (Table 33). Core-reduction flakes and shatter strongly dominate the debitage samples at these three components, accounting for 81.4% or more of the diagnostic debris (Table 38), with cortical flakes and simple interior percussion flakes constituting half or more of the core-manufacturing debris. At the same three components, the less-well represented biface manufacturing debitage is dominated by early stage removals. The debitage profile at INY-3004/5A is distinct from the other three Early Newberry lag quarries. Over three-fourths (76.4%) of the diagnostic percussion debris here is from biface manufacture, and three-quarters of this is middle stage removals. While the debitage profile indicates Stage 2 bifacial blanks were predominately worked, Stage 3 bifaces (n=34) are half again as numerous as Stage 2s (n=23). In sum, discarded preform fragments are overly represented at this atypical component. Table 38. Debitage Smmary Data for Early Newberry Period Components. Sample % Biface % Core % Biface Reduction Core Reduction Surf. 1/4" 1/8" Total % Diag. Reduct Reduct Shatter Early Middle Late Cort S.Int C.Int Other Lag Quarries 1824 Main A - 95 86 181 48.6 9.1 71.6 19.3 97.5 12.5 - 55.6 19.0 20.6 4.8 3004/5 A - 496 585 1081 20.8 76.4 18.7 4.9 15.7 75.0 9.3 50.0 33.3 2.4 14.3 4327 A 21 1 17 39 51.3 5.0 60.0 35.0 100.0 - - 25.0 25.0 16.7 33.3 4331 A - 100 9 109 84.4 18.5 78.2 3.2 53.0 41.2 5.9 58.3 29.2 9.7 2.8 Off-QualTy Contexts 1824 Main Y - - - - - - - - - - ..- - - 1984 B - 352 84 436 49.8 46.5 47.0 6.4 12.9 43.6 43.6 21.6 36.3 19.6 22.5 4329 K - - 1010 1013 19.1 85.6 8.2 6.2 19.3 40.4 40.4 43.8 37.5 18.8 0.0 NOTE: Surf. - surface; Diag. - diagnostic; Reduct. - reduction; Cort. - cortical; S.Int. - simple interior; C.Int. - complex interior. EARLY NEWBERRY PERIOD OFF-QUARRY ASSEMBLAGES The flaked stone assemblage from the three Early Newberry off-quarry components is strongly dominated by bifaces (76.6%), with substantially fewer flake tools (10.7%), and cores (8.2%; Table 33). One of these components contains a mixture of both Early Newberry and Early period material. Bifaces. Robust samples of bifaces were recovered from two of the tre Early Newberiy period off-quarry components (Table 33). Stage 2 specimens dominate their cumulative assemblage (52.2%), followed by a substantial amount of Stage 3s (30.7%). The number of unclassifiable percussion worked biface fragments (i.e., small margin fragments) is, however, substantial, at 12.9%, indicating the highly fragmented condition of these bifaces. As elsewhere, size decreases as biface reduction stage increases (Table 35). The Stage 1 and 2 bifaces at the Early Newbenry off-quany components tend to be slightly smaller than those at the quarries, but the Stage 3 preforms are longer, wider, and thinner than those from Early Newberry period quarries, averaging 64 x 46 x 9 mm (27.4 g). Comparatively few bifaces (17.7%) were discarded whole at non-quarry components, and most of these are Stage is and 2s (Table 36). Manufacturing breaks are most frequently the reason for rejection (56.59%), followed by 107 structural flaws (18.3%, Table 36). Triangular (27.3%) and wide-shouldered (20.2%) shapes are prevalent on the whole items, though rectangular and irregular shapes are well-represented (18.2% each), indicating little preference was given to particular morphologies. Projectile Points and Biface/Points. All three projectile points in Early Newberry off-quarry components derive from INY-4329K, and are Thin Elko points: a whole one of cryptocrystalline material (Plate 5i), a whole one of 3oshaa Ridge glass (Plate 5k), nd a poximial fragment made of West Cactus obsidian (Plate 5j). Both biface/points are from INY-1984B, an area identified as containing a mixture of both Early and Early Newberry period materials, and have been described above with the Early off-quarry assemblage. Cores. A modest number of cores was recovered from two of the three Early Newberry off-quarry components, though all core forms are represented (Table 33). Surprisingly, cobble/chunk tests are more numerous than any other core type; all of these, however, derive from INY4329K, which is contiguous to a lag quarry. Regardless of type, the cores from these contexts are about half the weight of their quarry counterparts (Table 34); yet about 70% are whole. Approximately one-third of the whole cores are blocky rectangular shapes, fewer are triangular, and wide-shouldered forms are rare. Unifaces. The six unifaces recovered from two Early Newberry off-quarry components differ little from those found at the lag quarries, and presumably served the same function. They average 58 x 46 x 19 mm (42.2 g, Table 37), and two-thirds are whole. Simple Flake Tools. Flake tools occurred at each of the three Early Newberry off-quarry components, cumulatively constituting 10.7% of their flaked stone assemblage (Table 33). Some 69.2% are whole, and typically one or two edges are modified per tool. They are large flakes (averaging 59 x 41 x 12 mm, 31.1 g), with fairly extensive portions of their edges modified. Convex (42.3%), straight (26.9%), and s-shaped (23.1%) used edges are all well represented, and use-wear commonly covers a 35-45 mm stretch of an edge (Table 26). Debitage. Only two of the three Early Newberry off-quarry components produced subsurface debitage samples, and their characteristics are quite different from one another. Debitage occurs in very low densities at INY-1984B, recorded at 138 pieces/m3, and about half results from biface manufacture, with half being core reduction debris or shatter. Among the biface reduction debitage, similar proportions of middle and late stage thinning flakes were recovered, with little indication of early stage removals (Table 38). Among the core reduction debitage, high proportions of all four major flake categories are represented: cortical, simple interior, complex interior, and rectangular interior flakes - indicative of a variety of core forms, some reduced to fairly advanced states. This site is immediately adjacent to a lag deposit, and the incidence of core-reduction debris at INY-1984B reflects quarrying of this nearby deposit. Component INY-4329K yielded the highest debitage density of any off-quarry component, regardless of period, at 4129 pieces/m3. Typical of off-quarry components, a very high proportion of the debitage (85.6%) is from biface manufacture. Among the biface reduction debris, similar high proportions of middle and late stage thinning flakes are represented (40.4% each), with less early stage reduction (19.3%). The small quantity of core- reduction debitage consists of high proportions of cortical and simple interior flakes, indicative of initial reduction. EARLY NEWBERRY PERIOD DISCUSSION Increasing disinterest in lag quarries and greater differentiation between site types denote subtle shifts in the land-use patterns from earlier periods. Fewer lag quarries exist than in earlier times, and three of the four studied have diminished flaked stone assemblages, containing fairly equal quantities of cores and bifaces. Their debitage profiles remain consistent with lag deposit quarrying, documenting an emphasis on iniial reduction and core manufacture, and less biface manufacture, most of which is early stage reduction. The lag quarry with the richest assemblage, however, displays characteristics typical of earlier off-quarry occupations - i.e., quarrying activities are poorly represented. Bifaces overwhelmingly outnumber cores at this component, late staged bifaces are the single most numerous artifact type, and flake tools are well represented. The associated debitage reaffirms the prevalence of biface manufacture and low quantity of initial reduction at INY-3004/5A. Considering these four 108 components together, fewer cores were manufactured at Early Newberry lag quarries compared to earlier lag quaries; nonetheless, they remain highly variable in form (though bifacial ones are less common) and comparable in size. Off-quarry components also show some differentiation between them. Their assemblages are similar, being dominated by bifaces (mostly Stage 2s and very few Stage ls), with cores slightly less common than flake tools. However, debitage profiles are dissimilar to one another: at one, core and biface manufacture are nearly equally represented; at the other, middle and late stage biface manufacture predominates. Aside from these slight shifts, flaked stone patterns for the Early Newberry are similar to earlier periods: initial reduction tends to occur at lag quarries, shifting to an emphasis on biface manufacture at off-quarry contexts; the size of artifact types diminishes in shifting from the quarry to off-quarry contexts, and no particular morphology dominates core or biface forms. Further, there is no appreciable change in the size of artifact types from one period to the next, indicating that comparable lag deposits continued to be used to produce the same kinds of artifacts. MIDDLE NEWBERRY PERIOD OFF-QUARRY ASSEMBLAGES Only three Middle Newberry components were identified. Each corresponds to a different off-quarry portion of one site, INY-1824 Main. The flaked stone assemblage at each is dominated by bifaces, collectively accounting for 87.1% of all flaked stone tools (Table 39). A few flake tools occurred at each, and they account for 8.6% of the composite asseblage (Table 39). Only one or two cores, unifaces, and biface/points were recovered. Bifaces. Each of the hee components yielded a robust biface sample, but the profile of that sample differs between them. Stage 2s are dominant at one, Stage 3s are dominant at another, and nearly equal amounts of Stage 2s and 3s were recovered from the third. However, fagments of indeterminate stage account for 22.3%, indicative of the bifaces' highly agmented condition. Only three of the 122 bifaces are whole (precluding meaningful size- estfimates [Table 40]), the remainder consisting mostly of margin (57%) and end fagments (20%). Two of the whole ones are rectangular shaped and the third is wide-shouldered. Compared to earlier or later periods, the proportion broken and discarded due to hulman error is high, and the proportion discarded due to structural flaws is low (Table 41). Projectile Points and Biface/Points. The single biface/point, from INY-1824 J, is a mid-section weighing 3.2 g, with a maximal width of 28 mm. Overall size indicates it likely is part of a dart-sized point. Cores. Both cores in the Middle Newberry off-quarry assemblage are from the same component, and both are whole. One is unidirectionally flaked and measures 81 x 55 x 37 mm (108.6 g). The other is a cobble/chunk test measuring 70 x 59 x 21 mm (84.7 g). Unifaces. Unifaces are as uncommon as cores, and both are from the same component (Table 39). One is a margin fragment; the other is whole, measuring 85 x 47 x 28 mm (89.8 g). Simple Flake Tools. A few flake tools were found in each of the three components (Table 39). Two-thirds of the flake tools are whole, and edge modification occurs primarily on straight or convex-shaped edges (Table 26). By comparison to flake tools in off-quarry components dating to other periods, they tend to be small, but are otherwise undistinguished. Debitage. Debitage densities range widely between these three loci, attaining minimal and maximal values of 283 to 1022 pieces/m3 (CU-1/4", Table 39). Regardless of the varying intensity of stone working between areas, their debitage profiles are remarkably similar, with biface reduction accounting for 96-97% of the diagnostic percussion debris (Table 42). Middle stage thinning tends to be the best represented activity, accounting for 44-51 % of the diagnstic biface percussion debris, with late stage inning debris only slightly less abundant, contributing a remaining 30-47%. Overall debitage profiles are quite consistent with artfact assemblages, reflecting intermediate and late stage percussion biface working and little else. 109 Table 39. Assemblage Inventories in Middle Newberry Period Components. 1824 Main J 1824 Main K 1824 Main M OQ OQ OQ Total Flaked Stone Bifaces, Uniface-B's and Flake Blanks Stage 1 2 - - 2 Stage 2 29 15 4 48 Stage 3 21 14 9 44 Indetennrminate 9 9 9 27 Not Analyzed - - 1 1 Subtotal 61 38 23 122 Projectile Points and Biface/Points 1 - - 1 Cores Unidirectional - 1 Other - 1 Subtotal 2 - 2 Flake Tools 3 5 4 12 Unifaces 2 - - 2 Debitage (1/4" nm3) 1022 283 616 Ground and Battered Stone Handstones Unifacial - 1 - 1 Subtotal - 1 - 1 Other Materials Features - -1 11 Seeds - - 55 55 NOTE: OQ - off-quarry. MIDDLE NEWBERRY PERIOD DISCUSSION The diminishing interest in lag quarry deposits that started in the Early Newberry period appears to have continued in the Middle Newberry period. Of the numerous quarries investigated, none was identified as a Middle Newberry period accumulation. Instead, several discrete flaked stone accumulations within a single off-quarry site document flaked stone reduction activities during this interval. Each of these components is quite similar to one another. Their assemblages are domnated by Stage 2 and/or 3 bifaces, of which nearly all broke during manufacture. Associated debitage is consistent with this near-exclusive focus on bifacial blank and prefom manufacture. LATE NEWBERRY PERIOD PRIMARY QUARRY ASSEMBLAGES A major shift in toolstone procurement practices occued in this period, with pimary outcrops serving as quarries instead of lag deposits. Of the numerous lag quarries investigated, none was identified as Late Newberry 110 Table 40. Summary Metics of Bifaces, Uniface-B's and Flake Blanks in Middle Newberry Period Components. Bifaces n mean s.d. max. min. 2 2 73 3 75 70 46 1 73 - - - 23 53 18 112 27 38 12 5 23 4 1 84.0 - - - 46 45 14 90 25 38 1 67 - - 26 45 10 64 22 37 8 2 14 5 Stage 1 edge angle Stage 2 length width thickness weight edge angle Sta~e 3 length width thickness weight edge angle 7 55 25 Off-Quarry Contexts Uniface-B's n mean s.d. max. min. Flake Blanks n mean s.d. max. m. 1 68 - - - 1 53 - 1 17 - - - 1 24 - 1 40 - - - 1 55 - 5 1 45 - - - 5 10 3 14 7 5 33 5 40 25 1 1 55 - 1 30 - 1 14 - 1 21.3 - 1 60 - Other 25 Not analyzed 1 Grand Total 112 2 8 2 NOTE: length, width, and thickness in mm.; weight in grams. Table 41. Summary Attributes of Bifaces in Middle Newberry Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeterminate All N Total Off-Quarry Contexts 2.1 4.5 2.5 3 121 Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Irregular Total % Total N Off-Quarry Contexts 33.3 66.7 100.0 Cause of Rejection Human Structural Cross- Poor Error Outrepasse Flaw Section Planar Shape Unknown Total % Total N Off-Quarry Contexts 79.3 6.6 12.4 0.8 0.8 - 100.0 121 Table 42. Debitage Summary Data for Middle Newberry Period Components. Sample % Biface % Core % Biface Reduction Core Reduction Surf. 1/4" 1/8" Total % Diag. Reduct. Reduct. Shatter Early Middle Late Cort. S.Int. C.Int. Other Off-Quarry Contexts 1824 Main J - 433 407 840 28.2 95.8 0.8 1824 Main K - 149 - 149 34.2 96.0 3.9 1824 Main M - 796 670 1466 26.9 97.0 2.0 3.4 1.0 25.1 44.5 30.4 - - 50.0 50.0 8.1 51.0 40.8 50.0 50.0 - - 7.3 45.3 47.4 12.5 12.5 37.5 37.5 NOTE: Surf. - surface; Diag. - diagnostic; Reduct. - reduction; Cort. - cortical; S.Int. - simple interior; C.Int. - complex interior. 111 1 22.0 38 36 3 1 1 in age; instead, two primary outcrops of Joshua Ridge glass were identified, both heavily exploited during the Late Newberny period. Cores dominate their combined assemblage (63.5%), followed by bifaces (34.0%), with two flake tools and two unifaces completing the flaked stone assemblage (Table 43). Cores. Cores dominate the flaked stone assemblage at both Late Newberry primary quarries. Non- pattemed and bifacially flaked forms are most abundant, accounting for 37.9% and 27.6%, respectively, with unidirectional and bidirectional forms contributing an additional 13.8% and 11.5 % (Table 43). Roughly 60% of the cores are whole. Cobble/chunk tests are, not surprisingly, the largest of the forms, averaging 330.2 g, followed by non-pattered ones, averaging 238.9 g (Table 44). Prevalent shapes include blocky rectangular, triangular and wide-shouldered forms, which occur in near equal frequencies. Bifaces. Bifaces occur at both primary quarries, but are particularly well represented at INY-2826C (Table 43). Most are Stage 2s (58.5%) followed by Stage ls (26.4%). As indicated in Table 45, bifaces from these primary quarries are large, and size diminishes as reduction stage increases: Stage ls average 101 x 65 x 25 mm (136.9 g); Stage 2s 96 x 56 x 21 mm (110.1 g). As a comparison of Tables 44 and 45 shows, Stage is are only slightly lighter weight than bifacial cores, yet dimensional relationships are similar, indicating that the former developed out of the latter: both bifacial cores and Stage 1 bifaces are about half again as long as they are wide, and width is about two and a half times greater than thickness. Only about one-fourth of the bifaces discarded at the Late Newberry primary quarries are whole (26.4%); and triangular and wide-shouldered forms predominate at 50.0% and 35.7%, respectively (Table 46). Human error during manufacture accounts for the discard of 62.3%, followed distanty by hose with irregular cross-sections (13.2%, Table 46). Structral flaws account for only 11.3b% of the discards, indicative of the high quality of obsidian at these outcrops, compared to lag deposits. Unifaces and Flake Tools. Two unifaces were recovered at one primary quarry. Both are whole, but one is small at 41.3 g, while the other is comparable in size to the unidirectional cores, at 145.4 g (Table 47). One simple flake tool was found at each of the components; and they, too are whole. Each displays damage on a single margin: a notched or spokeshave-like edge on the heavier item weighing 147.7 g, and a convex-shaped edge on the other weighing 84.3 g. The few unifaces likely are aberrant unidirectional cores rather than a functionally distinct tool; the few flake tools, judging from their large size, probably are damaged flakes rather than tools. Debitage. A high quantity of debitage occurred at one Late Newberry primary quarry (570/in3), and an extraordinarily high quantity (4832/im3) occurred at the other, indicating large volumes of stone were reduced. The debitage profiles are remarkably similar for the two components (Table 48), suggesting uniformity in the manner with which the quarries were exploited: about 75% of the diagnostic percussion debris is core reduction flakes or angular shatter, and 25% is from biface reduction. Of the core-reduction debris, cortical flakes are most abundant (about 50-60%), followed by high proportions of simple interior flakes (23-31 %). The biface reduction debris consists of about 48% early, 35% middle, and 17% late stage thinning flakes. The debitage profiles are generally consistent with the co-associated artifact assemblages: initial reduction activities were dominant, and the resulting cores remained large; fewer bifaces were made than cores, and Stage 1 and 2 forms predominated. LATE NEWBERRY PERIOD OFT-QUARRY ASSEMBLAGES Late Newberry off-quarry components are prolific in the project area, and rarely occur at localities used earlier or later. Twenty of the 21 investigated contain only Late Newberry period materials; the exception, INY- 3012C, contains a mixture of both Haiwee and Late Newberry period materials. Bifaces are plentiful in virtually each context, and account for 81.5% of the cumulative flaked stone assemblage from all such areas (Table 43). Flake tools make a distant secondary contribution, at 10.0% of the assemblage. No other flaked stone artifact type contributes more than 4.2% to the cumulative assemblage. 112 z to %09 r4 d: g C4 m a x o Lw g oe 4 m le 9 Ag A -9 A C-4 A CN Ag m 9 o t- . . t , I , I * e-( en. I - - - . . . * C4 c4 , , v 0 n I 113 113 cn -W C4 C4 C4 t- cn IW t- 00 C4 C4 (4 c'n C4 'V 00 C4 C4 en . W) 0 . . c I . I V I . I * . I . C4 . I M - . - , . *4 C4 u a) ca) .5 0 a) 0 -C's F4 , M en * * * 00 * O" - ' * WI * a *g g 0 3 0 A. re Zo CO cq 11- - fn 0 t- C4 1 en 9 2 1 - tn C4 . . %O %C I . 0 0 C4 . . fn e4 . fn I . - . I fn . . C4 enI - fn -. o i . (;% s . %C I I . I I I &.I z . ec ' 0 . en I mt * 0, e n 0^ . - 32% . C: . 02 * . . .. . . . .** * * . . . . . . . . - . - . . . . . . . . . . . . . . So...q CA z -~a *0 - * * * * * * * * * 0 0 C4 en * , * , * , * * ' , , , C4 C I a) , .. , * * *% *' ' * * O * * . . * . N~ , . . . P.~ cc .o C/3 Cu ~~~~* -0 z . Mtn Ss C4 i - I - C4 . %O . C4 An a fn . 9 . - a 00 C%14 I C14 C4 1 I a I I 0 . ? . l . . . . ' 4 . I . c - . - 4 . I C4 Ch - . . VN 0 00 e I f I y Mg 21 I :s ;; t- a . . .4 I - . fn W . M"QVWC4& C4" 1 1 cn--c4 - Table 44. Summary Metrics of Cores in Late Newberry Period Components. Primar Ouarries Off-Quarry Contexts n mean s.d. max. min. n mean s.d. max. min. Bifacial 24 7 length 15 95 16 127 73 4 98 19 119 66 width 20 67 14 108 51 7 57 13 81 38 thickness 20 29 7 42 18 7 37 7 45 28 weight 15 166.6 61.3 276.6 66.2 4 232.2 132.9 445.2 79.7 edge angle 24 79 8 95 60 7 74 7 80 60 Bidirectional 10 10 length 5 97 18 127 76 6 83 21 120 59 width 9 66 15 96 43 7 50 12 75 37 thickness 9 28 8 44 15 7 32 14 60 18 weight 4 152.8 60.8 253.2 93.4 5 134.0 107.6 327.7 35.1 edge angle 10 80 13 110 55 10 71 7 85 60 Non-patterned 33 11 length 30 102 18 144 71 8 83 20 119 54 width 33 71 16 128 48 10 62 21 93 21 thickness 33 36 12 84 19 11 33 10 48 19 weight 27 238.9 102.8 509.7 73.6 8 215.8 171.3 553.3 62.2 edge angle 33 78 14 110 55 11 77 15 100 50 Unidirectional 12 3 length 10 107 19 146 85 3 51 6 57 43 width 11 70 15 109 53 3 49 3 51 45 thickness 12 30 6 41 22 3 31 10 43 20 weight 9 190.2 54.7 280.0 115.4 3 78.9 27.9 114.3 46.3 edge angle 12 75 12 100 60 3 67 12 75 50 Cobble and Chunk Test 8 2 length 7 84 22 121 63 2 76 8 84 68 width 8 64 19 99 31 2 65 2 67 63 thickness 8 48 21 93 31 2 48 5 52 43 weight 6 330.2 220.9 640.2 87.9 2 272.1 89.2 361.3 182.9 edge angle 8 78 19 105 55 2 93 3 95 90 Not analyzed 12 Grand Total 99 33 NOTE: length, width, thickness in mm.; weight in grams. Bifaces. The impressive quantity of bifaces recovered from Late Newberry off-quarry components make them the "type" artifact for this period. Bifaces do te the flaked stone assemblage at 19 of the 21 components, and are absent at only two (Table 43). Stage 2s are most numerous at the majority of the components, constituting 52.3% of the bifaces, overall. An additional one-third (34.4%) are Stage 3s, while 10.9% are too fragmented for classification, and Stage ls account for merely 1.6%. Size and edge angle decrease as reduction stage increases, with Stage ls averaging 93 x 52 x 21 mm (98.3 g), Stage 2s 74 x 47 x 13 mm (56.0 g), and Stage 3s 65 x 41 x 9 mm (17.8 g); and they are much smaller than their counterparts from contemporaneous primary quarries (Table 45). Only 9.0% of these bifaces were discarded whole (Table 46), but among these, triangular forms dominate at 41.8%, followed by wide-shouldered ones (21.8%), though a variety of other shapes are well represented. Two-thirds (66.5%) of the bifaces were discarded due to human error during manufacture, augmented by 10.6% outrepasse fractures, with structural flaws accounting for 14.9% of the remainder. 114 Table 45. Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Late Newbeny Period Components. Primary Quarries Bifaces n mean s.d. max mi. 13 5 101 21 135 73 13 65 7 74 47 13 25 5 35 17 5 136.9 61.4 242.3 64.7 13 59 10 75 45 30 8 96 9 118 86 24 56 10 79 41 30 21 6 37 11 8 110.1 33.5 187.6 62.0 30 53 12 75 35 State 1 lengdth width thickness weight edge angle Stage 2 length width thickness weight edge angle Stane 3 width thickness edge angle 6 54 39 3 17 10 4 40 31 Uniface-B's n mean s.d. max m. 1 1 60 1 21 1 60 1 133 1 79 1 26 1 230.0 1 35 1 1 50 1 11 1 40 Flake Blanks n mean s.d. max. min. Other CGrand Total 49 Off-Quarry Contexts Bffaces n mean s.d. max min 9 2 93 26 118 67 4 52 10 63 36 8 21 4 27 14 2 98.3 60.7 159.0 37.6 8 59 16 90 40 298 24 74 17 103 41 174 47 14 98 17 273 13 4 26 5 23 56.0 25.4 119.5 12.6 298 46 11 80 30 182 8 65 15 92 41 148 41 11 84 18 179 9 3 18 4 9 17.8 5.5 28.4 9.1 180 37 8 60 25 5 2 65 7 71 58 5 38 8 53 30 5 7 3 11 3 2 29.5 5.3 34.8 24.3 5 31 5 40 25 60 28 82 Unifae-B's n mean s.d. max min. 1 1 134 - - - 1 27 - - - 1 60 - - 18 3 89 6 98 84 14 56 18 92 28 18 16 5 30 90 3 65.4 42.5 123.1 21.9 18 46 8 60 30 12 2 54 - 54 54 11 46 16 77 23 12 11 5 20 4 2 26.9 9.7 36.6 17.2 12 37 7 50 25 5 36 Flake Blanks n mean s.d. max. min. 5 5 60 13 78 41 5 48 17 70 26 5 18 5 24 9 5 50.6 33.3 101.2 9.1 5 49 7 60 40 17 7 52 13 72 35 17 36 14 70 21 17 9 4 20 5 8 12.3 8.9 34.2 5.0 17 33 5 45 25 2 24 NOTE: length, width, thickness in mm.; weight in grams. 115 6 5 46 6 12 6 37 4 Stage 1 lensth width thickness weight edge angle Staae 2 leth width thickness ~it weight edge angle Stage 3 length width thickness weight edge angle State 4 length width thickness weight edge angle Other Not analyzed Grand Total t A A 51 Table 46. Summary Attributes of Bifaces in Late Newberry Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeteminate All N Total Primary Quarries Off-Quarry Contexts Primary Quarries Off-Quarry Contexts 35.7 22.2 29.0 9.7 9.1 40.0 - 26.4 14 1.5 9.0 55 53 611 Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Irregular Total% Total N 50.0 35.7 14.3 - - - 100.0 14 41.8 21.8 14.5 12.7 5.5 3.6 100.0 55 Cause of Rejection Human Structural Cross- Poor Error Outrepasse Flaw Section Planar Shape Unknown Total % Total N Primary Quarries Off-Quarry Contexts 62.3 66.5 3.8 10.6 11.3 14.9 13.2 2.6 0.8 9.4 100.0 4.6 100.0 Projectfle Points and Biface/Points. Eight of the 21 components contained a few projectile points and/or biface/point fagmt (Table 43). Of the 11 points, six are Elkos (three Thin and three Thick variants), while the others include a Rose Spring, Gypsum, Little Lake, Great Basin Stemmed, and leaf-shaped forms (Table 49). The Gypsum and Little Lake points are both made of cryptocxystallin materials; e remaer are West Sugarloaf (n=7) or Sugarloaf (n=2) obsidian. Only two points are whole: one Elko and the Gypsum. Three of the six biface/point fragments are of sufficient size to infer they derive from dart-sized points, one other appears to be the distal tip of a drill, and the remning two are too fragmented to determine if they derived from dart or arrow-sized implements (Table 49). Unifaces. Only a few unifaces occurred at six of the Late Newberry off-quarry components (Table 43). Twelve of the 16 unifaces are fragmented; the few whole ones average 91.9 g, making them nearly twice the weight of associated Stage 2 and 3 bifaces (compare Tables 45 and 47). The highly fragmented condition of the unifaces, and their low frequency (2.0 % of the cumulative flaked stone assemblage) suggest that they represent uncommon, cast-off functional tools. Table 47. Summary Metrics of Unifaces in Late Newberry Period Components. Primary Quarries n mean s.d. max. min. 2 74 52 34 93.3 27 10 6 52.1 100 61 40 145.4 47 42 28 41.3 2 2 2 2 Off-Quarry Contexts n mean- s.d. max. 16 4 5 10 4 73 47 15 91.9 14 10 10 54.4 90 60 36 171.3 54 31 5 33.0 NOTE: length, width, thickness in mm.; weight in grams. 116 53 611 min. length width thickness weight Table 48. Debitage Summary Data for Late Newberry Period Components. Pimary Quarries 2826 C 4378 A Off-Quarry Contexts 1816 A 1824 Main D 1824 Main F 1824 Main Z 1824-B all 1824-X all 1906 B 1907 A 1984 A 2103 A 2103 B 2826 B 3012 C 3012 D 3012 E 3456 A 4243 A 4243 C 4244 all 4252 all 4267 N Sample Surf. 1/4" 1/8" Total - - 1966 1966 - 784 315 1099 - 614 521 1135 - 666 527 1193 - 268 859 1127 - 206 22 228 - 531 356 887 - 605 848 1453 - -- 1211 1211 - 684 133 817 - 525 743 1268 - 765 322 1087 - 215 43 258 - 276 -- 276 - 193 - 193 - 328 -- 328 - - 1059 1059 - 175 70 245 - 78 13 91 - - 2341 2341 - 389 408 797 45 -- - 45 % Biface % Core % % Diag. Reduct. Reduct. Shatter Biface Reduction Core Reduction Early Middle Late Cort S.Int. C.Int. Other 24.7 26.2 54.2 19.6 47.2 34.6 18.1 51.0 31.2 12.9 4.9 38.8 23.6 48.4 27.9 48.5 34.7 16.8 63.3 22.7 10.6 3.4 32.0 25.8 28.1 33.3 29.5 26.3 21.8 40.9 26.5 33.4 41.9 37.7 28.5 45.1 14.6 17.6 20.9 81.6 25.1 62.2 92.0 89.0 91.8 77.6 87.0 92.4 97.3 89.2 95.2 94.8 15.7 83.6 96.4 99.3 88.4 81.2 78.9 93.2 95.5 96.4 6.6 7.8 6.6 18.4 10.7 5.0 0.7 8.7 2.1 4.1 73.1 12.5 0.7 4.5 12.5 15.8 6.8 2.5 3.6 1.4 3.2 1.6 3.9 2.3 2.6 1.9 2.1 2.7 1.1 11.1 3.8 3.6 7.1 6.2 5.3 2.0 7.2 70.4 22.5 33.3 33.3 12.5 20.8 12.4 71.5 16.1 25.0 33.3 29.2 12.5 15.1 52.9 32.0 28.6 33.3 23.8 14.3 18.7 40.7 40.7 50.0 701 14.3 28.6 7.5 39.0 53.5 21.4 46.4 10.7 21.4 16.7 43.9 39.4 31.6 31.6 10.5 26.3 3.5 27.2 69.3 - - 50.0 50.0 6.3 48.7 45.0 6.8 44.8 13.8 34.5 10.0 42.8 47.2 - 42.9 14.3 42.9 16.5 48.3 35.2 43.5 26.1 17.4 13.0 41.1 58.8 - 63.3 21.5 11.4 3.8 - 32.2 67.8 30.8 23.1 38.5 7.7 5.7 73.6 20.8 - - - - 2.1 40.8 57.1 - 100.0 - - 10.2 48.2 41.6 14.3 14.3 14.3 57.1 15.4 61.5 23.1 66.7 33.7 -- - 6.7 20.0 73.3 33.3 33.3 -- 33.3 9.0 41.0 50.0 30.8 53.8 - 15.4 7.8 30.9 61.3 60.0 - -- 40.0 22.2 74.1 3.7 100.0 - - NOTE: Surf.- surface; Diag.- diagnostic; Reduct. - reduction; Cort. - cortical; S.Int. -simple interior; C.Int. - complex interior. Cores. A few cores occurred at about half of the Late Newberry off-quarry components (Table 43). In spite of the low number, all core forms are represented, with non-pattemed and bidirectional ones co-dominating (33.3% and 30.3%, respectively). Two-thirds of the cores (66.7%) were discarded whole - a prQportion slightly higher than occurred at the primary quarries. In general, they are only slightly smaller (i.e., lighter) than com le items at the primary quarries (Table 44), suggesting that the few at off-quarry localities were discarded in much the same form as they entered the site. Table 49. Projectile Points and Biface/Points fiom Late Newbenrry Off-Quaty Components. Component 1816 A 1824 MainF 1824-B 1984 A 2103 A 2103 B 3012 C 4244 Total Rose Spring Gypsum Elko Little Lake Great Basin Stemmed Leaf Shaped Dart-sized Drill Indeterminate Total I 3 1 1 1 1 6 1 1 1 -- -- 1 -- 1 1 -- -- -- 1 -- -- -- 2 6 1 -- -- -- 1 -- - ---- 1 -- 1 ....... -- 1 -- 1 1 1 -- -- 3 -- -- -- 2 2 4 1 1 2 17 117 Simple Flake Tools. A small quantity of flake tools occur in the majority (76.2%) of the Late Newberry off-quarry components (Table 43), and they are the second most-numerous artifact in the composite assemblage. Over half (60.8%) of the 79 implements are whole. Usually two or one modified edge exists per tool, and that edge has microchipping extending, on average, along 33 mm of a straight edge, or along 43 mm of a convex-shaped edge (Table 26). In comparison to those in earlier or later components, these flake tools are more extensively modified, as indicated by the average number of worked edges per tool. Drill. Only one drill was found within a component area, and that item is from INY-2103A. It is an 8.0 mm thick percussion flake that was bifacially pressure flaked along two parallel margins. Portions of the drill tip and proximal end have broken by snap fracture. Maximal dimensions of the fragment are 38 x 31 nmmn, with the bi-convex bit measuring 15 x 11 x 7 mm. Debitage. Debitage densities vary considerably from one component to the next, but generally are moderate to high, ranging from 100-500 pieces/m3 at half of the components, between 500-100/m3 at five, and in excess of 1000/m3 at three (Table 43). Biface reduction is the near exclusive activity represented at all but one of these components. The debitage profile at INY-2826B is atypical from all other Late Newberry off-quary components, with 84.3 % of the diagnostic percussion debitage identified as core production debris or shatter. This site, however, is within 30 m of a primary quarry, and the debitage is reflective of this. Excluding this one component, biface reduction debris accounts on average for 90.4% of the diagnostic percussion debris at the Late Newberry off-quarry components, and never less than 77.6% (Table 48). Debitage profiles at these components readily separate into two groups: those where early through late stage thinning activities are fairly well represented (n= 9), and those whee either middle or late stage inning debris predominates (n=10). At five components in this latter group, middle stage thinning debris constitutes anywhere between 61.5-74.1% of the bifacing debitage; while at the other five, late stage thinning flakes account for 57.1- 73.3% of the bifacing debitage. The repeated occurrnce of locations with high quantities of debitage from narrowly focused reduction activities is without precedent in earlier periods, and is indicative of a different, more highly structured, organizational strategy than was applied earlier. The reduction sequence was being segmented, and different segments were occurring at different locations. Debitage profiles clearly show this spatial differentiation in biface reduction activities; however, the biface samples do not, since Stage 2 bifaces are most common in nearly every context. This reconfirms the notion that debitage profiles are more reliable indicators of on-site reduction activities than artifact assemblage composition. LATE NEWBERRY PERIOD DISCUSSION By the Late Newberry period, quarrying has shifted to primary outcrops, where large volumes of obsidian were reduced, lag quanies having fallen completely into disuse. Items manufactured at the quarries are pfimarily non-patterned cores, bifacial cores, and triangular biface blanks, and items made at these outcrops are generally bigger than those from lag quarries. In contast to earlier lag quarries, flake tools and unifaces are extremely rare, further reflective of a narrowing in product manufacre and seemingly an exclusive preoccupation with quarrying. Concomitant with this shift to primary outcrops, the number of off-quarry, secondary reduction, components rises dramatically, further documenting the increased magnitude of stone tool manufacture. Reduction activities in off- quarry contexts are increasingly restricted to bifacial blank and preform manufacture, frequently undertaken at separate locations, indicating a more highly organized technological strategy than was practiced earlier. The plethora of these off-quarry biface m stations is reflected in Figure 17, accounting for the dramatic bulge in the Coso cumulative hydration curve between 2300-1275 B.P. HAIWEE PERIOD PRIMARY QUARRY ASSEMBLAGES Of the eight Haiwee period components, one occurs at a primary quarry, two at lag quarries, and five in off-quarry locations (Table 50). The flaked stone assemblage at the single primary quarry is dominated by cores (77.2%), followed distantly by bifaces (21.8%). A single flake tool completes the assemblage; no projectile points or unifaces were recovered. 118 Table 50. Assemblage Inventories in Haiwee Period Components. 4378 B 3004/5 I* 4322/H C Subtotal 3012 C* 4239 A 4240 4325 B 4329 O* Subtotal QP QL QL QL OQ OQ OQ OQ OQ OQ Flaked Stone Bifaces, Uniface-B's, and Flake Blanks Stage 1 5 - I - - - A 1 Stage2 17 A 1 1 6 7 3 - A 16 Stage 3 - A - - 5 48 - - A 53 Stage 4 - - - - - 7 - - - 7 Indeterminate - - - - 1 - - - A 1 Subtotal 22 A 1 1 12 63 3 - A 78 Projectile Points and Biface/Points - A - - 1 3 - - 1 5 Cores Non-Patterned 15 A - - 1 - - - A 1 Unidirectional 6 A 2 2 - - - - A - Bidirectional 6 A 1 1 - - - - A - Bifacial 2 A 1 1 - - - - A - Other 20 - - - - 1 - - A 1 Not Analyzed 29 - - - - - - - Subtotal 78 A 4 4 1 1 - - A 2 Flake Tools 1 A 6 6 3 6 2 - A 11 Unifaces - A - - - 6 1 - A 7 Debitage (?"/4m3) 932 A 2047 333 1698 273 57 A Ground and Battered Stone Millingstones Thin - 1 - 1 4 . 4 Slab - 9 - 9 2 - - - - 2 Block - 2 2 - - - - - - Indetennminate - - - - 1 - - - - 1 Subtotal - 12 - 12 7 - - - - 7 Handstones Bifacial - 1 - 1 - - - - - Unifacial - - - - 6 - - - 1 7 Subtotal - 1 - 1 6 - - - 1 7 Misc. Ground Stone - - - - 10 - - - - 10 Cobblo-CoreTools 1 - - - 5 - 1 - 4 10 Anvils - 1 - 1 - - - - - - Abraders - - - - 2 - - - - 2 Other Materials Features - 1 - 1 .2 - 2 Miscellaneous Stone - - - - - - - - 1 1 NOTE: * - Dual period assignments (see Table 10); QP - primary quarry; QL - lag quarry; OQ - off-quarry; A - items assigned to Early Component. Cores. Most cores at the Haiwee primary quarry are cobble/chunk tests and non-patterned, with comparatively few examples of other forms (Table 50). As average weight indicates, the cores tend to be large, often weighing between 'A to 1h kg (Table 51), averaging much larger than ones in pre-Haiwee period quarries. 119 Table 51. Summary Metrics of Cores in Haiwee Period Components. Bifacial length width thickness weight edge angle Bidirectional length width thickness weight edge angle Non-Patterned length width thickness weight edge angle Unidirectional length width thickness weight edge angle Lag Quarries Off-Ouarry Contexts n mean s.d. max. min. n mean s.d. max min. 1 1 111 - - 1 50 - - - 1 32 - - - 1 180.7 - - - 1 60 - - - 1 1 74 - - - 1 58 - - - 1 22 - - - 1 89.5 - - 1 75 - - - Primary Quarries n mean s.d. max. min. 2 1 112 - - - 2 67 15 82 52 2 50 15 55 25 1 383.6 - - - 2 95 5 100 90 6 5 94 19 123 63 6 74 11 86 57 6 41 14 72 32 4 216.9 169.3 460.6 22.5 6 84 15 105 65 15 15 113 26 190 86 15 77 12 107 59 14 42 8 55 29 14 355.2 264.6 1282.0 189.4 15 82 16 16 110 6 2 117 11 128 106 3 79 3 83 76 5 35 10 44 20 2 283.4 22.8 306.1 260.6 6 81 12 100 65 2 2 80 2 82 78 2 48 10 58 38 2 35 4 39 31 2 135.6 52.1 187.7 83.4 2 73 3 75 70 Cobble and Chunk Test 20 length 20 102 19 140 70 width 20 79 15 111 46 thickness 20 59 16 93 39 weight 20 491.3 312.8 1241.8 160.8 edge angle 20 91 14 110 65 Not analyzed Grand Total 29 78 NOTE: length, width, thickness in mmn.; weight in grams. Some 83.7% are whole, and planar shapes are generally blocky rectangular, rhomboidal, and triangular. Form and size indicate they generally received mini reduction; however, debitage presents a slightly different picture. Bifaces. Bifaces at the Haiwee period primar y are Stage 2s (77.3%), or Stage ls (22.7%). Bifaces, like the cores, are extremely large, with whole ones averaging nearly 200 g (Table 52), regardless of reduction stage, maldng them nearly three times as heavy as bifaces at earlier lag quarries, and more than half again the weight of those from Late Newberry primary quarries. About 40% of the bifaces are whole, being more prevalent among discardedStage ls than Stage 2s (Table 53). Half of the whole bifaces arewide-shouldered forms, and fewer are rounded-elongate, triangular, or irregular. Human error during m e is the most common reason for discard (31.8%) followed by structural flaws in the stone (22.7%). Simple Flake Tools. The single flake tool is 91 x 70 x 32 mm (149.3 g), and displays two unifacially microchipped straight edges. Debitage. Debitage occurs in high densities at this primary quarry, recorded at 932 pieces/m3. About 85 % of the diagnostic debitage is core reduction debris or shatter, with only 14.4% attributed to biface manufacture (Table 54). Core reduction debitage consists of 34.7% cortical flakes, 37.1% simple interior flakes and 23.5% 120 1 1 70 1 50 1 39 1 146.6 1 90 1 1 89 1 56 1 44 1 244.0 1 85 2 4 Table 52. Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Haiwee Period Components. Bifaces n mean s.d. max. min 3 2 92 9 101 83 3 68 6 72 59 3 36 2 39 34 2 198.9 50.7 249.7 148.2 3 48 6 55 40 15 6 107 18 129 84 11 66 12 79 43 15 26 7 40 15 5 195.7 82.1 314.8 93.9 15 51 15 75 25 Grand Total 18 Primary Quarries Uniface-B's n mean s.d. max min. 2 1 97 2 54 2 18 1 124.4 2 55 2 Flake Blanks n mean s.d. max. mi. 2 1 117 - - - 2 73 6 78 67 2 22 2 24 20 1 150.2 - - - 2 53 3 55 50 4 58 50 2 20 16 5 60 50 2 Bifaces n mean s.d. max mn. 1 1 17 - - - 1 35 - - - 16 1 70 - - 8 45 11 62 29 16 15 7 39 8 1 37.5 - - - 16 47 9 60 30 49 2 47 6 52 41 17 37 10 61 18 49 8 3 19 4 2 11.3 2.2 13.4 9.1 49 34 6 50 25 7 width 3 31 thicikness 7 7 edge angle 7 33 Other 1 Grand Total 74 Off-Quarry Contexts Uniface-B's n mean s.d. max. min. 1 1 50 1 12 1 40 Flake Blanks n mean s.d. max min. 3 3 48 3 51 45 3 30 2 33 28 3 6 1 6 5 3 8.5 1.3 10.4 7.6 3 27 2 30 25 5 37 26 2 10 5 4 40 30 1 3 NOTE: length, width, thickness in mm.; weight in grams. complex interior flakes. The relatively high proportion of complex interior flakes suggests greater concem for core shape than is indicated by the size and shape of the cores. The small amount of biface production debris represents all stages of manufacture, with middle stage thinning flakes most abundant (46.7%), and early (28.9%) slightly outnmbering late stage thinning flakes (24.4%). This, too, is at odds with the artifact assemblage, since no Stage 3 bifaces were recovered. HAIWEE PERIOD LAG QUARRY ASSEMBLAGES Two Haiwee period components were identified at lag quarry deposits (Table 50), but hydration data indicate that the flaked stone assemblage at only one (INY-4322/H-C) actually dates to the Haiwee period. This componnt is a small (5 m in diameter) chipping station within a lag deposit. Only flaked stone occurred here, and the assemblage was limited to one biface, four cores, six flake tools, and a high density debitage concentration. 121 Stake 1 length width thickness weight edge angle Stae 2 length width thickness weight edge angle LiameI thickness edge angle Staue 2 length width thickness weight edge angle Staze 3 length width thickness weight edge angle Stare 4 Table 53. Summary Attributes of Bifaces in Haiwee Period Components. Percent Whole Stage 1 Stage 2 Stage 3 Stage 4 Indeteminate All N Total Primary Quarries Lag Quarries Off-Quarry Contexts Primary Quarries Lag Quarries Off-Quarry Contexts 60.0 35.3 6.3 40.9 9 7.7 6 9.4 22 1 78 Shape of Whole Bifaces Wide Rounded Triangular Shouldered Elongate Rectangular Circular Irregular Total% Total N 12.5 50.0 25.0 - - 12.5 100.0 8 33.3 33.3 33.3 - - - 100.0 6 Cause of Rejection Human Structural Error Outrepasse Flaw Cross- Poor Section Planar Shape Unknown Total % Total N Primary Quarries Lag Quarries Off-Quarry Contexts 31.8 100.0 59.0 9.1 15.4 22.7 16.7 13.6 1.3 9.1 13.6 100.0 - 100.0 7.7 100.0 Artifacts. The biface is a Stage 2 end fragment. The four cores include two unidirectional and single bifacial and bidirectional specimens. Each is whole, and they tend to be small, individually weighing between 83.4- 187.7 g (Table 51). Two are wide-shouldered in plan-view, one is rhomboidal, and one blocky rectangular. Four of six flake tools are whole and quite large, averaging 63 x 42 x 15 mm (55.6 g). Of the modified edges, six are convex-shaped and two straight. Modification extends an average of 47 mm along the convex edges, but only 19 mm along straight edges. Table 54. Debitage Summary Data for Haiwee Period Components. Primay Quarries 4378 B Lag Quarries 4322/H C Off-Quarry Contexts 3012 C 4239 A 4240 all 4325 B Sample % Biface % Core % Surf. 1/4" 1/8" Total % Diag. Reduct Reduct. Shatter - 744 283 1027 30.5 - 614 305 919 32.1 276 - - 983 887 547 - 139 276 983 1434 139 37.7 20.7 26.0 19.4 Biface Reduction Early Middle Late Core Reduction Cort S.Int C.Int Other 14.4 54.3 31.3 28.9 46.7 24.4 34.7 37.1 23.5 4.7 22.8 56.2 21.0 43.3 44.8 11.9 25.3 33.1 36.1 5.4 83.6 98.0 87.0 92.5 12.5 2.0 10.0 7.4 3.8 3.2 - 32.2 19.3 40.4 22.2 43.3 12.0 40.0 67.8 40.0 34.4 48.0 30.8 23.1 38.5 43.8 37.5 18.8 43.2 48.6 8.1 - 50.0 - 7.7 50.0 NOTE: Surf. - surface; Diag. - diagnostic; Reduct. - reduction; Cort. - cortical; S.Int. - simple interior; C.Int. - complex interior. Debitage. Debitage density is high (2047/m3), and most diagnostic pieces are core production debris (56.2%) or shatter (21.0%). Only 22.8% is from biface production (Table 54). Of the core production debitage, 122 22 1 78 - complex and simple interior flakes occur with equally high frequencies (36.1% and 33.1%, respectively), outnumbering cortical debris (25.3%). The debitage profile suggests cores were the prevalent artifact made here, and that they were not haphazardly reduced: this is consistent with the artifact assemblage. Early and middle stage thinning flakes co-dominate the biface reduction debitage, indicative of Stage 2, bifacial blank, manufacture. HAiWEE PERIOD OFF-QUARRY ASSEMBLAGES Five Haiwee period components occurred in off-quarry contexts (Table 50). Three of these contain exclusively Haiwee period materials; Late Newberry and Haiwee period materials are commingled at INY-3012C; the flaked stone assemblage at INY-43290 dates predominately to the Early period, with one projectile point and the ground and battered stone assemblage ascribed to the Haiwee period. The cumulative Haiwee period flaked stone assemblage from these components is dominated by bifaces (75.7%), followed distantly by flake tools (10.7%). Unifaces and projectile points occur in moderate frequencies, at 6.8% and 4.9%, respectively. Cores are rare. Bifaces. Bifaces occur at thee of the components, and are particularly abundant at INY-4239A (Table 50). The vast majority of the bifaces are Stage 3s (68.8%) followed by Stage 2s (20.8%, Table 50); this is a marked change from earlier off-quarry assemblages, where Stage 3s typically constitute only one-third or less of the bifaces, and Stage 2s predominate. As a further distinction, pressure-worked bifaces (Stage 4s) are well-represented, accounting for 9.1 % of the biface sample; in earlier off-quarry assemblages, they account for less than 1 %. Very few (7.7%) of the discarded bifaces are whole, but they appear comparable in size to those in earlier off-quarry components, with Stage 2s averaging 45 mm wide and 15 mm thick, and Stage 3s 37 mm wide and 8 mm thick (Table 52). Three shapes are equally represented in the small sample of whole bifaces: triangular, wide-shouldered, and rounded-elongate. Human error during manufacture is identified as the reason for discard of 59.0% of the bifaces, supplemented by an additional 15.4% with outrepasse fractures. Structural flaws are evident on 16.7% (Table 53). Projectile Points and Biface/Points. Three Haiwee period off-quarry components contained one or a few projectile points or non-diagnostic fragments of points (Table 50). The four points include three Saratoga Spring (Plates lo, lp, lq), and one Rose Spring (Plate 2b). Each is made of Coso glass, and rim measures range between 3.0-4.9 1u. The one biface/point is an unfinished basal fagment of a Saratoga Spring. The point Wpes are consistent with the Haiwee period, as are the hydration rim values obtained on them. Cores. One core was recovered from each of two Haiwee period off-quarry components (Table 50). Each is whole: one is non-patterned, wide-shouldered, and weighs 146.6 g; the other is a circular cobble/chunk test, weighing 244.6 g. Unifaces. Unifaces are present at two of the components, and six of the seven were found at INY-4239A (Table 50). Each is a small margin fragment, and most weigh less than 5.0 g (Table 55). It is conceivable that some are actually unidirectional core or uniface-B fragments, rather than true unifaces; but their fragmented condition precludes such classification. Table 55. Sunnmmary Metrics of Unifaces in Haiwee Period Components. Off-Quanrv Contexts n mean s.d. max. min. 7 length* 7 34 13 53 18 width* 7 19 6 30 11 thickness* 6 8 3 12 5 weight* 7 4.5 4.2 13.8 1.3 NOTE: length, width and thickness in mm.; weight in grams; * - all measurements incomplete. 123 Simple Flake Tools. Several flake tools occurred at each of three Haiwee period off-quarry components (Table 50). Six are whole; five are fagmented. The whole ones are moderate in size, averaging 54 x 39 x 7 mm (12.3 g; Table 26). Typically one or two worked edges are present, mostly commonly straight (54.5%) or convex- shaped (27.3%). The few remaining have beaked projections 26-33 mm long. Debitage. Debitage densities vary greatly from one component to the next, ranging from 57 to 1698/m3 (Table 50). Though the intensity of reduction varies, the nature of the reduction activities is similar between components. Some 90.3 % of the diagnostic percussion debris, on average, is attributed to biface production, with minimal amounts of shatter or core-reduction debris (Table 54). Biface reduction at two of the components is narrowly focused on late stage thinning, while all reduction stages are fairly well represented at the other two loci. HAIWEE PERIOD DISCUSSION Quanzying remains concentrated at prmary outcrops during this period, however, the focus of manufacture shifts to producing relatively more, and larger, non-patterned cores than in the preceding Late Newberry period, and comparatively less, but larger, bifaces. Lag deposits completely ignored in the Late Newberry period are of some small interest in the Haiwee. There appear, however, to be very few of these Haiwee period lag deposits. The frequency of secondary reduction components, so common m the Late Newberny period, decreases significantly in the Haiwee period. At these fewer off-quarry components, like the primary quarries, biface reduction continued to more advanced stages, with preforms substantiy better represented than in prior intervals, and pressure-worked bifaces make their most significant contribution ever. As in earlier times, there is little evidence of further core reduction at off-quarry locations, indicating many cores were taken from quarries to distant sites. MARANA PERIOD LAG QUARRY ASSEMBLAGES A total of 26 Marana period componentswas identified at 13 sites (Table 56). Only two of these were at lag quarries, and only one (INY-3300 D) has flaked stone associated. The flaked stone at this component represents a single reduction event where a few cobbles were processed. All materials occurred within a 3 x 3 m area, and the assemblage is limited to a whole Stage 2 biface (92 x 42 x 29 mm, 92.3 g), two non-pattemed and two unidirectional cores, 15 pieces of percussion debitage, and the anvil on which the obsidian cobbles were worked. The cores are whole and fairly large, ranging from 121.1-370.2 g (Table 57). Blocky-rectangular, triangular, and circular shapes are all represented. Nine pieces of debitage are from core reduction (six cortical and three simple interior flakes); the six remaining flakes are from biface reduction. MARANA PERIOD OFF-QUARRY ASSEMBLAGES Twenty-four Marana period off-quarry components were identified (Table 56). Many of these were identified as late deposits based on radiocarbon dates obtained from features; hydration data indicated that the flaked stone assemblage at eight of the 24 components was not genuinely associated with late period use. Consequently, the flaked stone from these areas has been excluded from the Marana period composite assemblage. The flaked stone assemblage from the 16 remaining areas is incredibly meager: 14 bifaces, seven cores, five flake tools, four projectile points, and one biface/point fragment (Table 56). Debitage is similarly sparse, never exceeding 100 pieces/m3 (CU-1/4") at the few locations with subsurface deposits. Extremely limited amounts of flaked stone is the most noteworthy characteristic that these Marana period off-quarry components share. Bifaces. A few bifaces occurred at only seven Marana components with flaked stone. Of the 14, half are Stage 2s, half are Stage 3s (Table 58). The two whole Stage 2s are dissimilar: one is 76 x 71 x 24 mm (103.8 g), the other 46 x 36 x 9 mm (14.8 g). Three Stage 3s are whole, each a flake blank variant, and they average 45 x 28 x 7 mm (8.4 g). Of the whole ones, two each are triangular and wide-shouldered, one is circular. Human error during manufacture accounts for the discard of 50.0%; the whole ones display no apparent flaws; (Table 59). 124 (44 co 4 le - V'.bO - vl A R; 0 R g Q 0~ C3 g .mo o0, (m .m C3 .g I 'IC .4 . .< < < I I . 0 1 . I . . ,< ",.0 < .4 five seeds. 136 .? . . . . . . . . . o.. . ...:*'. .' ..... . , ' 0 SURFACE ROCK * SUBSURFACE ROCK (0 - 10 cm) 1- 0 1 METER PLAN PROFILE 6 A' FEATURE 2 LEGEND C) SURFACE ROCK- * SUBSURFACE ROCK 1637 - MILLINGSTONE Figure 21. Formnalized Rock-lined Hearths at INY4329, Locus 1. 137 PLAN FEATURE 1 LEGEND N175/W157 N175/WI57' 0 0 1 METER I I i I N72/E276 ' 0 2 FEATURE 3 LEGEND Q SURFACE ROCK i OBSIDIAN FLAKE CHARCOAL POCKET * BIFACE METERS PLAN 0% 1.0 cm deep). Rhyolite millingstones display a much higher incidence of slightly concave (52.1%) and deeply concave worn surfaces (18.6%), with flat/irregular surfaces reduced to secondary frequency (29.3%). For both materials, the depth of the ground surface correlates with artifact thickness. Of the granitic millingstones, blocks have the highest frequency of slightly or deeply concave surfaces (32.5%), followed by slabs (23.5%), then thin slabs (13.6%). This same trend exists among the rhyolitic millingstones: blocks have the highest frequency of deeply worn surfaces (33.3%), then slabs (20.6%), and, finally, thin slabs (7.3%). Wear type also correlates with material type and thickness (Table 67): simple polish tends to co-occur with flat/irregular wear, while deeper surfaces have a higher frequency of pecldking with polish; polish in combination with pecking and striations is most prevalent on concave surfaces, particularly on rhyolite millingstones. In short, an equal amount of use is likely to produce both a deeper mnilling surface and more extensive wear patterns on a rhyolitic millingstone than on a granitic one. The point here is that different wear patterns on granite and rhyolite millingstones are not necessarily the result of different functions, but of inherent properties of the stone. 141 I * EARLY o --- A LIlTLE LAKE O 5WMETERS Map 9. Distribution of Early and Little Lake Period Components. 142 111 -1-1 * EARLY NEWBERRY A MIDDLE NEWBERRY 0 1300 FEET 0 500 METERS Map 10. Distribution of Early Newberry and Middle Newberry Period Components. 143 SUG MOT ,St- 44327 520 b C-NY4 K:N VI < / /d * LATE NEWBERRY o 1300 FEET 0 500 METERS Map 11. Distribution of Late Newberry Period Components. 144 D(5J D I 'e:C L 1k1U; mwl r 1 'c ] ii * MARANA ,0 A HAIWEE 0 SMEORS Map 12. Distribution of Haiwee and Marana Period Components. 145 73 t It, I QL n c Table 66. Millingstone Surface Configuration by Form. Thin Slab Block Total Granite Flat/Irregular 19 42 27 88 Slightly Concave 1 10 11 22 Deeply Concave 2 3 2 7 Total 22 55 40 117 Rhyolite Flat/Irregular 24 35 4 63 Slightly Concave 27 73 12 112 Deeply Concave 4 28 8 40 Total 55 136 24 215 Bifacial millingstones are rare irrespective of material (Table 68), accounting for 9.5 % of the granitic and 15.8% of the rhyolitic ones, and their frequency changes little according to artifact form. Among bifacial rhyolitic millingstones, primary sides (the one side with the heaviest wear) show a stronger tendency to be more deeply worn (31.4% are deeply concave, 60.0% are slightly concave; Table 69) than is the case on the unifacial ones (15.7% deeply concave, 50.5% slightly concave), while the depth of wear on the secondary sides parallels that of the unifacial tools. Granite millingstones, in contrast, show equal frequencies of concave surfaces on bifacial (primary side) and unifacial forms, with flat/irregular surfaces prevalent on unifacial millingstones, as well as both sides of bifacial ones. These trends fiuther suggest that concave surfaces developed from use rather than purposeful construction; otherwise, the apparent "discard" of a concave milling surface by using the opposite face would not have been as common as it was. Table 67. Millingstone Surface Configuration by Wear Type. Polish Polish, Pecking Polish, Pecking, Striations Total Granite Flat/Irregular 89 13 4 106 Slightly Concave 10 14 - 24 Deeply Concave 3 4 - 7 Total 102 31 4 137 Rhyolite Flat/Irregular 68 10 3 81 Slightly Concave 89 46 5 140 Deeply Concave 19 25 2 46 Total 176 81 10 267 Other millingstone assemblage traits document diachronic trends suggestive of changg patterns of mobility and increasing intensification of plant resources. Basgall et al. (1988) have found that millingstone portability indices 146 Table 68. Millingstone Form by Bifacial and Unifacial Wear. Thin Slab Block Total Granite Biface 3 5 3 11 Uniface 18 50 37 105 Total 21 55 40 116 Rhyolite Biface 8 24 2 34 Uniface 47 112 22 181 Total 55 136 24 215 seem to monitor differing degrees of residential mobility (see Chapter 2). In their Fort Irwin study, for example, thin slab millingstones exhibiting heavy wear were often associated with expansive settlement systems, probably indicating that the thin slabs were used and carried for extended periods of time. In partial contrast to the trends documented by Basgall et al. (1988), at Coso thin slabs are scarce, and they tend to have flat surfaces. As indicated in Table 70, the use of thin slabs (ess than 6.0 cm thick) decreases dramatically through time, and blocks (greater than 13.0 cm thick) are in greatest use during the Marana period. Thin slabs account for 33.3% of the Early millingstones, rise to their highest at 52.9% in the Early/Middle Newberry period, and decline in subsequent periods to 39.1% in the Late Newberry, 27.8% in the Haiwee, and 17.6% in the Marana period. Block millingstones outnumber thin slabs only in the Marana period, but it is the thicker millingstones that have heavier, i.e. relatively deeper, wear. Given the increasing prevalence of thick millingstones late in time, and the tendency for deeper wear surfaces to occur on thicker millingstones, it follows that wear depth should also show an increase over time, and data bear this out (Table 71). Among the rhyolite sample, the frequency of pre-Marana millingstones with slightly or deeply concave surfaces figures to 58.1% (ranging from 41.6% to 75.0% depending on period) compared to 71.6% of the Marana rhyolite millingstones. Granitic millingstones adhere to this pattern, but due to material hardness, the difference is less evident - 18.2% of the pre-Marana ones are slightly or deeply concave, compared to 25.4% of the Marana ones. Table 69. Millingstone Surface Configuration by Bifacial and Unifacial Wear. Flat/Irregular Slightly Concave Deeply Concave Total Granite Uniface 83 22 6 111 Biface Primary 10 2 1 13 Biface Secondary 13 - - 13 Total 106 24 7 137 Rhyolite Uniface 67 100 31 198 Biface Primary 3 21 11 35 Biface Secondary 11 19 4 34 Total 81 140 46 267 147 Table 70. Millingstone Attributes Across Time. Early/Middle Early Little Lake Newberry 4 2 1 7 1 7 8 5 9 1 15 Granite Thin Slab Block Total Rhyolite Thin Slab Block 5 2 7 4 6 10 9 8 17 5 5 Late Newberry 2 3 1 6 7 4 2 13 1 2 3 1 1 9 9 5 23 Haiwee Marana Total 1 1 4 11 2 4 33 21 58 17 46 7 16 40 23 79 33 79 11 17 70 123 - 2 2 - 6 7 - - 2 - 8 11 - 2 2 - 2 3 - 2 2 - 6 7 5 25 53 11 87 129 2 30 38 18 142 220 NOTE: Indeterminate types not included. Two possible explanations for the divergent Fort Irwin and Coso patterns readily come to mind. First, the two ends of the spectrum could represent separate functions. For example, thin-flat rnillingstones could have been carried to distant seed patches for light hulling (producing little wear), while thick concave forms could have been placed near residential bases where more intensive pounding and grinding took place. Because there appears to have been a progressive change in the mix of millingstone forms used over time at Coso, it appears more likely that differences in wear might have simply resulted from longer or shorter durations of use. Rather than pecking a surface to purposefully create a concavity, a millingstone used for an extended period of time would have required pecking to sharpen a surface smoothed over by heavy use. If this were the case, thinner objects may have been broken relatively early in their use-life, often prior to developing roughed-up concave surfaces. Partial support for this notion is gained through an assessment of fragment type according to artifact form (Table 72). Some 75.0% of the rhyolite blocks are whole or nearly whole, compared to 19.1% of the slabs and 5.5% of the thin slabs. This general trend applies to the granitic sample, though the tendency for overall completeness is higher due to granite's superior strength compared to rhyolite: 87.5% of the granite blocks are whole or nearly whole, compared to 76.4% slabs, but only 27.3% of the granitic thin slabs. 148 5 5 Total Dike Thin Slab Block Total Basalt Thin Slab Block All Total Thin Slab Block Total Table 71. Millingstone Surface Configuration Across Time. Early/Middle Early Little Lake Newberry Late Newberry Haiwee Marana Total Granite Flat/Irregular 6 - 7 4 1 47 65 Slightly Concave 1 - 1 1 - 14 17 Deeply Concave - - - 1 - 2 3 Total 7 - 8 6 1 63 85 Rhyolite Flat/Irregular 4 2 7 5 8 25 51 Slightly Concave 2 6 1 6 6 50 71 Deeply Concave 4 - 4 3 4 13 28 Total 10 8 12 14 18 88 150 All Flat/Irregular 10 2 14 9 9 72 116 Slightly Concave 3 6 2 7 6 64 88 Deeply Concave 4 - 4 4 4 15 31 Total 17 8 20 20 19 151 235 NOTE: Other materials not included. Although few Coso millingstones show clear evidence of long-term curation, there seems to be a tendency for pre-Marana peoples to select objects suitable for at least short-term transport. Such a strategy is consistent with an opportuistc approach to seed gatherng whereby portable millingstones easily carried but prone to breakage were moved to ripening seed patches, and used with little concern for reuse in the future. The Marana pattern, in contrast, is characterized by a more focused approach to seed gathering, reflected by numerous clusters of milling equipment distributed across the landscape. In several instances, block millingstones were moved to known areas of high productivity, as if individuals were fully anticipating return visits to the same location. Despite the greater short-term effort required to transport block millingstones, it was apparently considered a worthwhile endeavor by people heavily dependent on seeds. Table 72. Millingstone Form by Condition. Thin Slab Block Total Granite Whole 4 31 30 65 Near Whole 2 11 5 18 Margin 7 12 4 23 Interior 9 1 1 11 Total 22 55 40 117 Rhyolite Whole 1 15 11 27 Near Whole 2 11 7 20 Margin 28 76 5 109 Interior 24 34 1 59 Total 55 136 24 215 149 Handstones Like the millingstones, rhyolite and granite are the prevalent materials for handstones, but basalt makes an increased contribution (Table 73). The frequency of bifacial wear appears dependent on material type, showing a consistent increase from rhyolite (13.4%) to granite (26.3%), and finally basalt (54.2%). Chronological change in these relationships is difficult to measure due to the small number of items from any particular time period (Table 74). Compansons between Marana and pre-Marana assemblages, however, show relatively even frequencies of bifacial items (21.2% and 22.9%, combined material types), as well as the same rough mix of the three material types. Table 73. Handstone Wear by Material Type. Granite Rhyolite Basalt Total Flat Polish 22 29 6 57 Polish, pecking, striations 2 4 - 6 Total 24 33 6 63 Slightly Convex Polish 30 44 10 84 Polish, pecking, striations 6 18 7 31 Total 36 62 17 115 Convex Polish 7 4 2 13 Polish, pecking, striations 1 4 2 7 Total 8 8 4 20 Rounded Polish 4 5 4 13 Polish, pecking, striations - - 2 2 Total 4 5 6 15 Uniface 42 84 11 137 Biface 15 13 13 41 Total 57 97 24 178 Most rhyolite and granitic handstones have slightly convex and flat grinding surfaces, followed by fewer with convex or rounded faces (Table 73). Basalt handstones follow the same pattern but have a higher frequency of rounded surfaces. Simple polish is the dominant wear type, followed by some combination of polish, pecking, and/or striations (nmltiple wear types were collapsed into a single category to increase sample size). Wear other than just polish occurs less frequently on granitic than on other types of handstones, probably due to its hardness. Secondary wear is much more prevalent on basalt and granitic handstones than rhyolitic ones, with edge grinding and battering observed more frequently than edge flaking (Table 75). Basalt and rhyolitic handstones rarely were formally shaped, while this phenomenon is slightly more common on granitic ones. Finally, whole or nearly whole handstones are common regardless of material type. 150 Table 74. Frequency of Unifacial and Bifacial Handstones Across Time. Early/Middle Early Little Lake Newberry 2 1 3 1 1 2 1 1 2 2 Late Newberry 5 2 7 9 2 11 Granite Uniface Biface Total Rhyolite Uniface Biface Total Basalt Uniface Biface Total Haiwee Marana Total 1 1 6 6 1 1 1 21 7 28 27 6 33 31 11 42 44 8 52 4 1 5 4 3 7 2 1 3 1 1 2 3 3 14 5 19 7 1 8 66 101 NOTE: Other materials not included. Table 75. Handstone Shaping, Secondary Wear, and Condition by Material Type. Granite Shaping Shaped Unshaped Total Secondary Wear Edge Batter Edge Flaking Edge Grinding None Total Fragment Type Whole Near Whole Margin End Total 8 42 50 5 5 9 36 55 33 12 8 4 57 Rhyolite 3 89 92 8 84 92 57 15 19 2 93 Basalt 2 20 22 4 4 14 22 16 1 3 2 22 NOTE: Indeterminate objects and other materials not included. 151 1 All Uniface Biface Total 52 14 79 22 Total 13 151 164 9 5 21 134 169 106 28 30 8 172 The correlation between surface configuration and wear type remains the most important pattern documented. Flat handstones almost always have simple polish, perhaps used primarily on flat millingstones. When working surfaces become more convex, also corresponding to the millingstone data, more rigorous grinding is reflected by the presence of striations and pecking (Table 73). This trend is most evident on basalt and rhyolite items. Given that millingstone wear depth increased during the Marana period, one might expect a corresponding increase in the frequency of convex handstones, but this is true for only some material types (Table 76). Marana granitic handstones show more convex surfaces than pre-Marana ones, (67.7% to 58.8%), but the opposite is true for rhyolite handstones (55.3% to 68.2%). These data, when combined with observed similarities in material selection and bifacial use over time (see Table 74), suggest little change in handstone function over time. Notwithstanding differences in overall abundance, and sampling problems associated with pre-Marana assemblages, it appears that basalt and granitic handstones were used (and presumably curated) for longer periods of time than rhyolitic ones, no doubt indicating a greater preference for these material types (see Bostwick and Burton 1993). The former commonly exhibit bifacial wear, secondary damage and, in the case of granite, formal shaping. Rhyolite handstones rarely showed these attributes, the material is more friable but more easily shaped, characteristics favoring more expedient, casual use. Table 76. Handstone Surface Configuration Across Time. Early/Middle Early Little Lake Newberry Late Newberry Haiwee Marana Total Granite Flat 1 1 1 4 - 10 17 Slight Convex - - - 4 1 19 24 Convex 3 2 - - - 2 7 Total 4 3 1 8 1 31 48 Rhyolite Flat - - - 5 2 17 24 Slight Convex - - 2 9 4 19 34 Convex - - - - - 2 2 Total - - 2 14 6 38 60 Basalt Flat - - - 1 - - 1 Slight Convex - - - 1 1 5 7 Convex - - - - - - - Total - - - 2 1 5 8 NOTE: Rounded and indeterminate specimens not included. Non-Obsidian Cobble-Cores Cobble-cores were recovered in most contexts, but the majority came from Early and Little Lake period lag quarries, and Marana period milling stations. Given the lack of plant processing in the former contexts, and the lack of stone working in the latter, a good deal of functional variability is expected to exist in this artifact class. These implements likely were used as bmmerstones in flaked stone manufacture at the early lag quarries, and for processing vegetal material and fashioning and/or maininig milling equipment during the late period. These divergent uses should be apparent by differing proportions of the types of rock selected, and by the types of use- related wear present. 152 The mix of materials used for cobble-cores differs markedly from the milling equipment, and evidently corresponded with the need for tools more resistant to fracture. Granite (47.1 %) and dike rock (40.7%) occur in high quantities, followed by basalt (5.8%), with few of rhyolite or assorted other materials (Table 77). Basalt, though not abundant, must have been a favored material, given that such pieces were probably carried into the study area, while other material types could be locally obtained. Tracking the distribution of material types across time reveals the persistence of granite and dike rock in all but Late Newberry contexts. Granite and dike rock occur in similar high proportions in the Early and Little Lake components combined, at 46.8% and 44.3% respectively; while granite cobble-ores become more abundant than dike rock ones in Marana contexts (53.8% and 38.5%, respectively). It is also noteworthy that basalt is largely restricted to stone worling contexts, while granite and dike rock are found in all settings, suggesting various uses throughout time. Table 77. Non-Obsidian Cobble-Core Material Types Across Time. Early/Middle Early Little Lake Newberry Late Newberry Haiwee Marana Total Granite 30 7 9 7 7 21 81 Rhyolite - - - 4 1 2 7 Basalt 4 - 1 4 - 1 10 Dike Rock 16 19 11 6 3 15 70 Other - 3 - - 1 - 4 Total 50 29 21 21 12 39 172 The changing functional role of these tools can be monitored in part by intensity (i.e., heavy, equivocal, no wear), and types of use-wear over time (Table 78). Early period deposits show a disproportionately high frequency of items with little or no observable wear (74.0%), while Marana period contexts have a much lower frequency of these items (29.7%). Intervening time periods show variable relationships, some of which are dependent on material, but about 40% of dike rock and granite cobble-cores in the Little Lake through Haiwee periods have heavy wear. Irrespective of time, basalt cobble-cores rarely have equivocal or no wear (20.0%), reflecting a higher degree of tool curation. Analysis of the implements with heavy use-related wear indicates that the vast majority of granitic tools in pre-Haiwee period deposits exhibit battering (83.3%, Table 79). Only 53.8% of the combined Haiwee-Marana sample have this type of damage, most of the balance displaying grinding. The same pattern holds for dike rock cobble-cores, with battering dominant until Late Newberry times, at which point, edge flaking and grinding gain importance. These data indicate that non-obsidian cobble-cores were composed of materials purposefully chosen for durability. Those in relatively early settings are usually at lag quarries and appear to have been used as hammers for working the obsidian. Beginning in Late Newberry times, but reaching peak proportions d g the Marana period, non-obsidian cobble-cores show more variable patterns of use: heavy wear is found on a greater number of objects (reflecting more intensive use?), and the incidence of grinding and edge flaking document activities not associated with the reduction of obsidian, but with vegetal processing. FLORAL REMAINS Of the 12 coponents with identifiable floral remins (see Tables 20, 39, and 56), only six produced five or more seeds (Table 80), one from a Middle Newberry feature and five from Marana period features. Of the five economic plant complexes defined for eastern California (greens and shoots, roots and tubers, seeds, fruits and 153 Table 78. Non-Obsidian Cobble-Core Wear Intensity Across Time. Early/Middle Early Little Lake Newberry Late Newberry Haiwee Marana Total Granite Heavy 7 3 3 5 1 12 31 Equivocal 5 3 1 - 1 2 12 None 18 1 5 2 5 7 38 Total 30 7 9 7 7 21 81 Dike Rock Heavy 5 9 2 3 1 13 33 Equivocal - 4 3 1 - - 8 None 11 6 6 2 2 2 29 Total 16 19 11 6 3 15 70 Basalt Heavy 1 - - 4 - 1 6 Equivocal - - - - - - - None 3 - 1 - - - 4 Total 4 - 1 4 - 1 10 berries, and tree crops), only small seeds and one fruit/berry taxon (desert tomato) were recovered. Notwithstanding problems associated with archaeological visibility (e.g., greens and shoots not preserving in the archaeological record), these findings are entirely consistent with the environmental conditions of the Volcanic Field. Table 79. Non-Obsidian Cobble-Core Wear Type Across Time. Early/Middle Early Little Lake Newberry Late Newberry Haiwee Marana Total Granite Battering 6 3 3 3 - 7 22 Edge Flaking - - - 1 - - 1 Grinding - - - 1 1 4 6 Other 1 - - - - 1 2 Total 7 3 3 5 1 12 31 Dike Rock Battering 4 8 2 - - 8 22 Edge Flaking - 1 - 2 - 3 6 Grinding 1 - - 1 1 2 5 Total 5 9 2 3 1 13 33 Seasonal availability of most of these foods ranges from spring through summer, while some extend into the fall (Figure 23). Season-specific estimates can be offered from some individual features by considering the 154 Table 80. Plant Macrofossils from Select Features. Middle Newberrv Marana 1824 Main M 4243 B 1816 C 1906 F 1906 H 1906 H F-1 F-1/2 F-2 F-9 F- 11I F-10 Total Grasses Oryzopsis sp. (Ricegrass) 9 >5500 22 8 - - >5539 Stipa sp. (Needlegrass) - 1 - 1 Poaceae (Grass Family) 2 2 2 - 5 - 11 Herbs Boraginaceae (Borage Family) 8 - - 2 9 1 20 Chenopodium sp. (Goosefoot) - 20 - 4 - - 24 Descurainia sp. (Tansy mustard) 2 2 - - 10 - 14 Mentzelia sp. (Blazing star) 9 7 2 24 80 2 124 Salvia columbariae (Chia) - - 3 27 - - 30 Shrubs Malvaceae (Mallow Family) - 1 3 - - - 4 Berries Lycium sp. (Desert tomato) 11 - 6 - - 17 Unknown 14 7 6 2 - 3 32 Total 55 >5540 44 67 104 6 >5816 combination of taxa recovered. Two Marana period features (Feature 1/2 at INY4243B and Feature 9 at INY- 1906F), for example, had a co-occurrence of ricegrass, goosefoot, and blazing star seeds, but no desert tomato. S r use is suggested, assuming they were collected together or at the same time. Feature 1 at INY-1824 Main- M and Feature 2 at INY-1816C, in contrast, both contain desert tomato in addition to ricegrass and other small herbaceous seeds (not including goosefoot), probably indicating late spring/early summer use. The other features, lacking either desert tomato or goosefoot, probably represent spring/summer collection. All seeds recovered are mentioned in local ethnobotanical studies, particularly ricegrass, blazing star, chia, tansy mustard, and desert tomato. As discussed in Chapter 4, ricegrass seeds were either detached from the plants with seed beaters and baskets at individual seed patches, or separated after the entire plants had been cut and transported to a base cam. Following collection, most grasses were winnowed and parched, and then ground into flour and eaten with no further preparation, or used in stews, gruels, and breads (Delacorte 1990; Zigmond 1981). Blazing star, clhia, and tansy mustard were collected with a seed beater and basket, and after parching and grinding, chia and tansy mustard were mixed with water into a beverage, while blazing star was tumed into a paste like peanut butter (Zigmond 1981). Desert tomato fruit was eaten raw, squeezed for juice, or pulverized into a paste. The plant remains provide additional support for the idea that local seed exploitation increased in the late period, but little can be said about the nature of pre-Marana seed exploitation patterns, given the limited data available. Judging from the mix of resources recovered from the one Middle Newberry sample, it appears that at least some plants used during the Marana period were also used earlier. The difference between pre-Marana and Marana period strategies, however, is probably one of opportunistic gathering earlier in fime (i.e., if seeds were ripe when obsidian deposits were being exploited, they were harvested) as opposed to the Marana period approach where local crops were carefully monitored and maximally exploited. The former approach is consistent with the Middle Newbery asseblage which shows a high frequency of desert tomato which is available early in the year (cooler) and requires little processing. Given the single sample at our disposal, such an interpretation is, however, largely conjecture. 155 - - - - - - - - - - - - - - - - -- - - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - - E 0 E z ----------------------------------------------------- ---------------------------- ------------------------- E ,----------------- ------ ---------------------- -----------------1-1 ------ - ---------- mmm.m m m m m m m m m mm , m ml a aa) .----------------------------------------------------- ------------------------------------------------------S 0.. mmmmmmm- - - -m mmmm- - - -m m co0 co a) '0 mmmm mI <5 _ Au 156 FAUNAL REMAINS Faunal remains, like seeds, are rare in all but a few Marana period deposits (Table 81), making it difficult to compare the nature of animal exploitation patterns over time. The pre-Marana collection consists of two jack rabbit bones (neither burned) and eight unidentifiable small mamm fragments (none burned). The Marana assemblage comes from features and includes two jack rabbit bones (both burnt) and 106 unidentifiable small mammanl fagments, the majority burnt. Large mamn rema ns are restricted to a single unidentifiable fragment at INY-4267. Table 81. Faunal Remains from Components. Lepus Small Mammal Large Mam l Intrusive/ Burnt Unburnt Burnt Unburnt Burnt Unburnt Indeterminate Total Early 1984B* - 1 - - - - - 1 3299A* - - - - - - 3 3 Total - 1 - - - - 3 4 Little Lake 3299A* - - - - - - 3 3 4330A - - - 3 - - - 3 Total - - - 3 - - 3 6 Newberry 1906B - 1 - 1 - - - 2 1907A - - - 1 - - - 1 1984A - - - 1 - - - 1 1984B* - 1 - - - - - 1 4244A - - - 2 - - - 2 4329K - - - - - - 2 2 Total - 2 - 5 - - 2 9 Marana 1816C - - - - - - 53 53 1906F* - - 36 4 - - 8a 48 1906H 2 - 37 14 - - 105 158 4267A - - 1 2 1 - 10 14 4329A - - - - - - 4 4 4329D - - 7 - - - - 7 4329N - - 3 1 - - 1 5 4329W - - - 1 - - - 1 Total 2 - 84 22 1 - 181 290 NOTE: Intuive/indeterminant includes bones that are obviously modem and those that could not be classified as large or small mammal. * - dual period assignment; I - one sample missing (8 pieces). To detemine the influence of soil acidity on faunal remains, pH tests were conducted on a series of soil samples from INY-1906H, where bone was abundant, and INY-4329E, where it was not. Two samles from the former yielded rading of 6.2-6.4 (slightly acidic) and 6.8 (very slightly acidic), while three samples from the other 157 location produced 6.8 (very slightly acidic), 6.8-7.0 (very slightly acidic to very slightly aLkaline), and 7.0-7.2 (very slightly aLkaline). The overall low level of acidity observed, combined with the correlation between slightly higher levels of acidity and the abundant presence of bone, suggests the scarcity of faunal remains is not a function of poor preservation. We conclude, therefore that game animals such as pronghorn and bighorn sheep were rarely hunted and consumed during forays into the Volcanic Field. Rather, it seems that local subsistence resources of all kinds were of secondary importance until the Marana period, at which time exploitation of small seeds and, to a lesser extent, rabbits and smaller game, became the primary reason for visiting the area. DISCUSSION The review of subsistence-settlement pattern studies from areas surrounding the Coso Volcanic Field presented in Chapter 2 indicates that adaptive strategies were largely consistent across much of eastern California during the early Holocene. Most Early sites investigated produce assemblages indicative of high degrees of residential mobility, minimal use of seed resources, and a probable emphasis on hunting large and small game animals. With the possible exception of the Stahl site (see also Delacorte 1988), Pinto/Little Lake period adaptations appear to be quite similar with respect to mobility and hunting strategies, but differ from the Early period by showing an increased use of milling equipment, a development thought by many to reflect increased dependence on seed resources in response to declining mid-Holocene environmental conditions. Witiin Owens Valley, Gypsum/Newberry settlement systems remained mobile, but appear to have been less expansive and more regularized than in earlier times. Unlike the Mojave Desert, where numerous short-term occupations were used to take advantage of a variety of resource zones, Owens Valley populations established fewer seasonal residential bases (probably spending summers near Long Valley and winters in southem Owens and Rose valleys), from which logistically organized groups would set out to various resource patches to obtain materials necessary to support the residential base. By Haiwee times, the Owens and Rose valley areas were characterized by increased sedentism, territoriality, and sociopolitical complexity - phenomena made possible by the more intensive exploitation of progressively smaller pieces of land, as well as a more diverse use of areas previously used almost exclusively for hunting (e.g., Sierra Nevada, White Mountains). Although the general pattem appears to have continued into the Marana period in Owens Valley, permanent or semi-permanent residential bases rarely developed in surrounding areas. Instead, most groups required the use of numerous short-term residential base camps to exploit the various resource zones within their respective territories. Findings from the current study, although representing only a small part of the overall land-use systems in operation during the prehistoric past, are largely consistent with the various adaptive strategies outlined above. Early assemblages, lacldking milling equipment and other subsistence-related material, seem to represent only quick visits to the Volcanic Field. The Little Lake data, like in most other areas, reflect a pattern not dissimilar to the Early period. Although quarries remain the dominant manifestation, off-quarry biface production loci increase in relative proporton, perhaps indicating some kind of subsistence-settlement shift not clearly delineated by the data currently in hand. The Newbery period, previously viewed as a single entity, appears to have had a great deal of internal variability. Early and Middle Newberry period materials within the Volcanic Field seem to represent a transitional phase between the early emphasis on short-term quarry use and the Late Newberry focus on the production of bifaces. Most Early Newberry period components are associated with lag quarries in the northern half of the study area. All three Middle Newberry components lie to the south in off-quarry contexts where biface manufacture and limited subsistence pursuits took place. In addition to the increased production of bifaces, many undoubtedly for exchange (see the following chapter), Late Newberry quarrying shifts to the use of primary deposits. Given the subsistence-settlement models developed for the Owens Valley/Rose Valley area, these patterns may reflect logistically organized groups visiting the Volcanic Field from base camps in Rose Valley, similar in organizational structure to those who traveled to the Sierra Nevada from the Lubkin Creek site (INY-30) to obtain pinyon, marmots, and mountain sheep. Several of the biface production sites are highly specialized and are therefore consistent with such a scenario. Others also include a small amount of subsistence-related maerials and may reflect oppornistic pursuits by the stone workers, or more diversified social groups originating from desert areas to the east and south. 158 Evidence for Haiwee period use of the Volcanic Field is meager at best, largely due to a sudden decrease in the number of biface production sites. In contrast to the Late Newberry pattern, it appears that only a few high quality seams of obsidian were exploited, with secondary reduction occurring at nearby residential areas like Coso Junction Ranch (INY-2284) and the Rose Spring site (INY-372). By the Marana period, the Volcanic Field was intensively used for its small seed resources. As originally hypothesized by Bettinger and Baumhoff (1982) and Bettinger (1991), this latter development clearly demonstrates not only a more intensive use of small seeds late in time, but also a late period expansion into areas under-used by earlier subsistence-settlement systems. 159 CHAPTER 8: COSO OBSIDIAN EXCHANGE PATTERNS The goals of this chapter are to develop a sense of the relative and perhaps absolute amount of Coso glass that found its way into prehistoric sites in surrounding areas, in order to better appreciate the breadth and depth of the Coso glass production and exchange system; to document Coso glass consumption patterns outside the procurement zone, assessing their influence on production activities at the source; and, assming that exchange networks are responsible in part for the prehistoric distribution of Coso glass, to identify the prevalent directions of movement so that mechanisms of exchange and interaction can be isolated. To assess the time-space distribution of Coso obsidian necessitated reviewing hydration and sourcing information from numerous archaeological projects laying within a fairly expansive area. Special effort was expended in tying to trace the prevalence of Coso glass outward, roughly in cardinal directions, from the source. Given the varying intensity and types of archaeological investigations throughout southern California, it was in some instances difficult to differentiate areas where data were absent, from areas where Coso glass does not occur in prehistoric contexts. Notwithstanding the difficulties inherent in trying to draw fairly standardized observations from research presented by different scholars with different interests, a fairly reliable picture of the quantity and form of Coso obsidian exists for many outlying areas. Two large-scale projects provide evidence for the quantity of Coso obsidian prehistorically carried into lands laying to the east (Map 2): excavations at some 60 sites in the Mojave B Range of China Lake Naval Air Weapons Station (Bouey and Mikkelsen 1989), and the Fort Irwin Archaeological Project (excavations at over 100 locations by Wirth and Associates, followed by Far Westem Anthropological Research Group, Inc.). Excavations at the Rose Spring site (INY-372, Locus 1 [Lnning 1963; Yohe 19921) and the Coso Junction Ranch site (INY-2284 [Allen 1986]) provide evidence of the intensity of stone-working activities at two major habitation sites just west of the Volcanic Field, occupied during the Newberry and Haiwee periods. In tracing the distribution of Coso obsidian to the north, data from the Contel Project (Delacorte and McGuire 1993) are useful, supplemented by information from the Lubkin Creek site (INY-30 [Basgall and McGuire 1988]). The former project involved excavations at about 20 sites within a north-south corridor between Little Lake (at the south end) and Big Pine (at the north end). Investigations at the Lubkin Creek site, south of Lone Pine, provide evidence of Coso glass consumption patterns at a residential base some 70 km to the north, occupied during the late Newberry and Marana periods. Looking to the west, data from several projects on the Kern Plateau are combined to paint a picture of the temporal distribution of Coso glass in the southern Sierra uplands, 40 km from the Volcanic Field. Following distributions onto the western slope of the southern Sierra, data from Lake Isabella and the Tule River Indian Reservation are next considered. Too few archaeological data are available from the southern San Joaquin Valley to build a composite hydration curve, but work completed by Wedel (1941), Fredrickson and Grossman (1977), and Hartzell (1992) is important for our assessment, as it indicates that obsidian is perhaps less than one percent of the chipped stone in sites around Buena Vista Lake. Next, we move southwest into the Los Angeles basin and southem California coastal areas (Ventura, Los Angeles, and Orange counties). Hughes and True (1985) have shown that Coso glass is present in eteme southem Califo (San Diego, San Bernardino, and Riverside counties) in only trace amounts, and so we make no effort to compile item-specific data from this area. Their study, however, points to two equally noteworthy patterns: Coso glass is more numerous in coastal than in inland sites, suggesting a coastal distribution route; and it is extremely rare in late-period sites, most pieces are found instead in earlier deposits. The scarcity of Coso glass in inland San Diego County sites is reaffirmed by Laylander and Christianson (1988) and McDonald (1992). Using McDonald's (1992) research at Indian Hills Rockshelter in Anza Borego as an example, the deposit there extends back to about 4000 B.P., but only one of the 239 pieces of obsidian subject to XRF-analysis was Coso. Coming full circle, our review is completed with a synopsis of Sutton's (1988a, 1988b, and 1991) observations pertaining to the distribution of Coso obsidian in Antelope and Fremont valleys, encompassing areas up to 150 km south of the Volcanic Field. For each of these localities, a hydration profile was developed from readings from varying numbers of sites (Table 82). We recognize that many factors could readily influence a composite curve such that it poorly correlates with what the actual relative frequency of Coso obsidian was at varying points in the past for any one area; nonetheless, for lack of standardizd data ameable to rigorously traclng and quantifying the amount of Coso glass distibuted over a huge area through time, we ask that the reader grant us this admittedly and necessarily subjective review. 161 ? N o^c o 0 t- - ( o o t- - - ON t -o t - N O t - I .. . o tN NI- q , t-t 10 .i e W -. . -' ,4 - --e -in o c. " N - - mmt --- - - - - ~ ,? ~ , , -.. C,l OOI 00 l ~t e~ O1 ~ 00 1C t-'" ' ' ' e e -- - - a a-a I I I I I -4 C) o 4) 04 o o "M C.) -o o 0 4- 0 I- AE o 0 C) 40 UL) 0 00 94) ,0 CU A 4)- (AXt 'a 0 u > S - (A 3u Zl-< e4S CA en V3 z co!! rl~ . - Ca' 0 .E S *J - . nA I a -O I S - ? ' tI I N t- In I t I *- a - I I I I a 4 -4 W- a a a - - a a -n -^^vt^- - -i- c' s> a I I I . O m 0 CfO O 04 - -Cfsc' Xf e >-- - .- -a-4c'>t < seIe4} s s- - a a a a a - a a a a I - a - a a a a a - a a a a - 6 A o tn 6 oo 0 WI o WI - in oo in 0 v) 0 v) I I - %A s %A - A - - A 6 A tA - - es C'} - C; C- (- o l C- osW W; o6 S6 c t- rfi 06 06 Qt Coa W fI o - -- - - -- - - 162 a - a a a a I a I I ' I ' - ' ' t" I I I I I I I I I I I I I I I I I I I I .-I V-4 0 I I I I oc.) .~Z C) 0 o g o- WE ! 8 4) Ca 00 e n- Q~ ) a~ Ca Uo CuC 04 .eCu QC- '0 0 4)0 0 8 CA Oglq I- U) o 4) ..-4 o I- r0 o0 00 ow 4- co C~ w GE, Q0 W-. o Ur o 0 4) Eu I - a I - -- - o eF e ->>o f}' Ft }es c t ON qe *N qi ON m I" m 14 m v m 14 m v m qe 0 z u* - . . 0 . . .0 0 . . . . el; . . . a . . . 4) 2 V o 0 O- O. V- V- .40 d-4 s1.4 ,l N. N.4 e4 CO ( W-4 W- - -N N NN N N N N N N " N N1 N 163 O W a 0% Ifn lt en r- - N C0Acfn fe en 00 -4 -oo in 00 ON -) m o t m - E -" -q a oo D o O C4' N~~~~~~ W)W)xDONem- I N t c4- t N N - -4 -4 - -4 - I I I I i I I I I I I I i I I I I I I I I I I I I I I I I I I I I i I I I I I I I I - * * * I I I I I I I I I i I i I I I I I I I I i I I I I I I I i I I I I .-q I W" I I I I I I I I I I I I I I I I I I I I I I I I I I I I I - .0 I I I I I I I I MN--N I - .-O MN N N-e Me N -MN - - M- - V-4 .4 0 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 1 6 I I I I I I I MOJAVE B RANGE A test evaluation project of 60 sites in the Mojave B Range of China Lake Naval Air Weapons Station (Map 2), located about 75 km southeast of the Coso Volcanic Field, was completed by Bouey and Mikkelsen (1989). Only small quantities of obsidian exist in prehistoric sites in this area, but virtually all of the obsidian was obtained from Coso. All but 31 of 221 (86.0%) pieces of obsidian from 23 sites submitted for geochemical sourcing were identified as Coso (Bouey and Miikkelsen 1989:78-82). By count only about 0.3% of all debitage is obsidian (Bouey and Mikkelsen 1989:195), and most of this is pressure flakes (70.7% of the diagnostic obsidian pieces). The 176 Coso hydration values (Figure 24a), form a unimodal distribution, with 51.7% of the values falling between 4.5-6.4 ,u (Table 82). The distribution is essentially the same when 21 formed tools are excluded. Applying the EHT correction factor derived for Fort Irwin (Table 83), the distribution shows a rapid rise in the amount of obsidian at the end of the Late Newberry period, maximal amounts during the Haiwee/Saratoga Spring period, with quantities trailing off rapidly thereafter. Persistently low amounts of obsidian were brought into the project area prior to ca. 2000 B.P. Table 83. Coso Obsidian EHT Corrections for Select Locations (adapted from Basgall 1990). Mean Annual Correction Locality Temperature in ?C Factor (6%) Grant Grove 7.25 1.3600 Long Valley 9.30 1.2328 Bishop 12.30 1.0666 Lone Pine 13.50 Cajon Pass 14.60 0.9445 Malibu 15.30 0.9946 Haiwee/Coso 15.70 0.8723 Randsburg 17.00 0.8211 Fort Irwin 17.50 0.8072 FORT IRWIN Fort Irwin is approximately 140 akm southeast of the Coso Volcanic Field (Map 2). Information provided in a compendium of chronological data from Fort Irwin (Gilreath et al. 1987), from excavations at 47 sites in Nelson Basin (Basgall 1991), and from seven sites in Tiefort Basin (Hall 1992), show that comparatively little obsidian is present in these prehistoric sites. Pieces occurring are typically pressure flakes or projectile point fragments - the vast majority of Coso obsidian. In Nelson Basin, for example, obsidian is less than 1 % (by count) of all debitage (Jones 1991:329); and in Tiefort Basin that proportion is slightly higher at about 2.5%. Debitage analysis identified 49.0% of the technologically diagnostic obsidian pieces from Nelson Basin as pressure flakes and an additional 29.7% as late stage thinning flakes (Jones 1991). Obsidian is better represented among the projectile points, accounting for about 10% (118 of 1156; Gilreath et al. 1987; Jones 1991, 1992). Basgall (1991:49) has indicated that obsidian is relatively more abundant in early Holocene deposits than in later contexts. Approximately 80% of the 1338 pieces of obsidian geochemically analyzed from Fort Irwin have been traced to Coso. Some 724 Coso hydration values from Fort Irwin are reported, ranging from 1.0-26.9 A (Table 82 readings compiled from Basgall 1991; Gilreath et al. 1987; and Hall 1992). The composite hydration curve shows a bimodal distribution (Figure 24b). A broad secondary peak spans hydration values from about 12.0-15.4 j (28.0% of all values, Table 82), and the primary peak spans hydration values from 4.5-6.4 I (24.4% of all values). The distribution is essentially the same when formed tools are excluded. Using the EHT correction factor for Fort Irwin 164 n=176 Fort Irwin EHT 60 so 20 - 10 0 2 4 S 10 11 12 1J 14 1S 16 lT17 14 2 ta 3a Fort Irrn 40 F "0- , I n=1 39 10 Haiwee EHT .- - o 0 1 2 3 6 7 9 10 1 1 12 13 14 1S 16 17 10 19 20 21 22 23 24 2S I I |Rose Spring, CA-INY-372, Locus 1 120 1 I1 I I 110I I I 100 90 i I 1 60 I I I I n=485 0I Haiwee EHT 301 20 I I I d 010 0 1 2 3 5 6 7 10 l1 12 13 14 15 16 17 18 19 20 21 22 23 24 2s 26 I Coso Juncteon Ranch, CA-INY-2284 e D r ii Figure 24. Frequency Distribution of Coso Hydration Values for Various Projects in the Western Mojave Desert. 165 40 30 20 10 a o I I I I I I I I I I I I I I derived by Basgall (1990), the broad peak pre-dates 6000 B.P., quantities decrease throughout the Little Lake period, then rises dramatically at the'end of the Newbeny period, to attain apparent maximal frequency during the Haiwee period. Quantities decrease throughout the Marana period. The discrepancy between the shape of the curve and Basgall's statement that obsidian is more prevalent in early Holocene deposits is a result of sampling bias. The greater quantities of obsidian recovered from early contexts typically require that a sample be drawn for hydration and sourcing analysis, while in later periods, the quantity is so low that nearly every piece recovered has received analysis. Were a less biased sample graphed, the pre-6000 B.P. values would simply rise to meet or exceed the Haiwee period peak. THE ROSE SPRING SITE (INY-372, LOCUS 1) The Rose Spring site lies only 15 km northwest of the Volcanic Field near the eastern base of the Sierra Nevada at the extreme north end of Rose Valley (Map 2). The site has long been heralded as a significant Rose Spring/Haiwee period village with an extensive midden (Lanning 1963). Yohe (1992) recently completed re- examination of the site, and determined, however, that the zenith of activities occurred ca. 1500 years ago, in late Newberry times. In total, 139 hydration rim values are reported on material for Locus 1: Ericson (1978) reports 39 items geochemically identified as Coso glass, and Yohe (1992:284-286) reports 68 values on debitage geochemically sourced to Coso and 32 on debitage reasonably presumed to be Coso obsidian. The composite profile (Figure 24c) indicates roughly a unimodal distribution, with 51.1 % of the rim values falling between 7.0-8.4 ly (Table 82). Using the Haiwee EHT correction factor (Table 83), fully two-thirds (67.6%) of the rim values fall between ca. 3300- 1350 B.P. (8.4-5.8 e), and only 17 (12.2%) between 5.84.2 , corresponding to 1350-650 B.P. It should be noted that Yohe (1992) clearly discounts hydration profiles from the deposit as temporally meaningful since they failed to confirm his assessment of well-stratified vertical deposits, giving preference instead to radiocarbon dates and projectile point distributions. Despite Yohe's (1992) reservations regarding the usefulness of obsidian hydration, a great deal of data support the contention that Locus 1 at the Rose Spring site was foremost a late Newberry obsidian reduction workshop and secondarily a Haiwee and Marana period occupation site. First, debitage from units X-1, X-2, and X-3 (Yohe 1992:230) is bimodally distributed with depth, the first peak occurs between 20-40 cm, followed by a sharp drop in density between 40-50 cm, and then an increase to previous or higher values between 50-140 cm (Table 84). Subsurfac te debsitage densities in the upper component (040 cm) calculate to 6,300 pieces/m3 (1/8" units), and 10,550 pieces/3 in the lower component (50-150 cm). The ver etical distribution of projectile points from the same block of units tracks the same upper and lower components (Table 84). All Elko and Gypsum series points were recovered below 60 cm, and 86.4 % of the Desert and Rose Spring series points were recovered from above 70 cm. Finally, hydration values corroborate two major occupations. Half of the 16 values from 0-50 cm measure <5.8 u, compared to only three of 48 rims below 50 cm. The overall pattern of hydration readings suggests that most of the debitage in the site was generated during Newberry period occupations. Due to the massive accumulation of this material, it was incorporated into later dating deposits, creating the clean Newberrty hydration profile below 50 cm and the mixed suite of readings nearer to the surface. Unfortunately, the projectile points recovered from the deposit have not been submitted for hydration analysis. But based on the evidence at hand, we would predict that when these data are available, most Rose Spring points will display rinds < 5.8 A, while Elko points, corresponding to the deeper deposits, will have rims > 6.0 1. Given the site's importance as an obsidian reduction workshop, the nature of the flintknapping activities that occurred on site are of much interest. Expecting a reduction in the size of bifaces to correlate with the introduction of the bow-and-arrow ca. 1500 years ago, Yohe's analysis indicated, istead, that "there is a reduction in biface size only in the last 600 years" (1992:219). Further, "the bifacial core reduction trajectory apparently remained largely the same throughout the most intensive period of use at the site, with exhausted bifacial cores or large, timmed flake blanks serving as the source of preforms" (1992:221). Regarding debitage characteristics in the deposit, Yohe (1992:234) comments "Probably the most striking thing about the debitage analysis is the apparent homogeneity maintained in most categories through time". From our interpretation of data from Locus 1 at the Rose Spring site, the overwhelming abundance of Newberry period debitage throughout the deposit would preclude documenting a diachronic change in reduction technologies. Yohe provides the following characterization: 166 Table 84. Subsurface Distribution of Projectile Points and Debitage from INY-372, Locus 1 (1987-89). Desert and Debitage Debitage Debitage Volume Rose Spring Elko and Unit X-1 Unit X-2 Unit X-3 Excavated Depth Series Gypsum Series (per k) (per k) (per k) (m3) 0-10 4 - 0.5 0.4 0.1 .675 10-20 7 - 1.3 1.4 0.3 .675 20-30 3 - 1.6 2.6 2.4 .675 30-40 3 - 1.4 3.5 1.5 .675 40-50 7 - 1.0 1.6 0.7 .675 50-60 7 - 2.2 2.6 1.0 .675 60-70 7 1 1.9 2.9 1.8 .675 70-80 2 - 2.0 2.8 1.7 .675 80-90 1 - 2.4 2.8 2.6 .675 90-100 2 - 2.3 3.0 unit terminated .450 100-110 1 - 3.1 2.8 .450 110-120 - - 2.5 2.5 .450 120-130 - - 2.3 2.2 .450 130-140 - - 2.6 1.8 .450 140-150 - 1 3.8 1.4 .450 150-160 - 1 1.7 unit terminated .225 160-170 - 1 1.6 .225 170-180 - 2 1.9 .225 180-190 - - 0.7 .225 190-200 - 1 1.1 .225 200-210 - 1 0.3 .225 210-220 - - unit terminated 220-230 - - - TOTAL 44 8 38,200 34,300 12,100 10.125 NOTE: Information derived from Yohe (1992:Table 24 and Figure 45); Excavation unit dimensions were 1.5 x 1.5 m, 1/8" mesh. The extremely low incidence of single-facet platforms, and the ubiquity of biface-thinning flakes throughout the deposit appear to support the original contention that bifacial cores were the most commonly used core type at the site. Platform abrasion prior to flake detachment seems to have been an uncommon practice. The small incidence of bulb- removal flakes suggests that the incipient bifacial cores were completely prepared prior to arrival on the sites, a contention that also is supported by the small numbers of alternate flakes and flakes bearing traces of cortex [1992:234]. It appears that less than 10% of the diagnostic flakes overall are attributed to pressure flaking (1992:235). It should be noted, too, that less than one percent of the chipping debris was of material other than obsidian. COSO JUNCTION RANCH SITE (INY-2284) Twelve lam due west of Sugarloaf Mountain, the Coso Junction Ranch site sits at the eastern base of the Sierra Nevada at a well-developed spring (Map 2). The site was tested by the UCLA Field School in Archaeology in the mid-1980s, and has been described as an extensive Rose Spring/Haiwee village with numerous structural remains, and voluminous amounts of chipping debris (Allen 1986). 167 Nearly 500 hydration rim values are reported from this deposit: Allen (1986:52-56) presents about 350 rim values on 287 pieces; Whitley (1988) presents 140 values on 129 additional specimens. All specimens are presumed to be Coso obsidian; the specimens reported by Allen are identified as debitage, while it is unknown if those reported by Whitley are from debitage or formed tools. The resultant sample shows a unimodal distribution with two-thirds of the values spag 5.0-6.9 I (Figure 24d, Table 82). Using thie Haiwee EHT correction factor, the most readings correspond to the recent end of the Late Newbeny period, ca. 1350 B.P., and frequencies decline as rapidly through the Haiwee and Marana period as they rose throughout the Early and Middle Newberrty periods. Also interested in monitoring the impact of the transition from dart to bow-and-arrow hunting technology on the consumption/reduction patterns of Coso glass, Allen segmented the vertical deposit into five phases based on hydration rim data. No appreciable differences were found to exist, however, in the debitage characterization from one phase to the next (Table 85). Enormous quantities of debitage occur in all phases, debitage exhibiting cortex is always rare, and in contrast to the Rose Spring site, tool finishing (i.e., pressure flaking) is always by far the overwhelming flintknapping activity. In sum, like the Rose Spring site, the deposit spans the Newberry and Haiwee periods, and when the transition from dart-sized to arrow-sized projectiles purportedly occurs, it fails to affect debitage profiles. Table 85. Hydration Profile and Debitage Characterization at INY-2284. Hydration Debitage Density per m3 % Early % Middle % Late % tool Phase Range (kg) (k) Reduction Reduction Reduction finishing % cortex IA a 5.2-5.7 6.4 23 0.6 3.2 17.3 78.8 0.7 b 5.8-6.1 6.1 19 0.9 3.2 22.4 73.4 0.5 c 6.4-6.6 6.4 22 1.0 2.9 19.3 76.7 0.1 d 6.74-6.9 4.0 14 0.6 4.0 17.0 78.3 0.7 e 7.3 1.9 10 0.4 1.5 9.6 88.5 0.7 NOTE: Data derived from Allen 1986:Tables 6, 7, and 10. Allen's debitage categories 14, 10 and 11 are combined for Early Percussion Reduction; categories 5, 6, 12, and 13 for Middle Percussion Reduction; categories 7, 8, 14, and 15 for Late Percussion Reduction; and category 9 for Finishing (pressure flaking). Percent on which cortex is present is the result of combining categones 1, 3, 5, 7, 10, 12, and 14. Debitage recovered from 1/8" mesh units. THIE CONTEL PROJECT Results of an evaluation of 23 sites on a north-south corridor between Little Lake and Big Pine (Map 2) indicate that Coso obsidian is common in prehistoric sites as far north as Lone Pine, but that it decreases rapidly in prevalence 10-15 om further north. In total, 182 hydration values are reported on 178 specimens of Coso glass recovered from 14 of the sites: 51 specimens were from sites located in the Little Lake vicinity, 104 specimens from sites located between Cottonwood Creek and Olancha, and the remainder from the vicinity of Lone Pine and northward. A composite profile of the hydration values shows three peaks of increasing magnitude, the lowest from 7.0-7.4 , the second at 5.5-5.9 , and the highest spanning 4.0-4.9 (Figure 25a, Table 82). Omission of readings on formed tools has little impact on the relative distribution. Using the Haiwee rather than the Lone Pine EHT correction factor (given that the majority of the readings derive from the south end of the corridor), the tertiary peak falls within the Newberry period, the secondary peak crosses the Haiwee/Newberry division, and the highest peak falls within the Haiwee period (Figure 25a). Minimal amounts of readings prior to ca. 3000 B.P. and post-650 B.P. are represented. 168 n=182 Haiwee EHT 60 50 40 30 20 10 a o 30 20 10 b o 70 60 - 50 - 40 - 30 20 - 10 - C o 30 20- 10 do 7 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 a I '57 ie Pine EHT 91 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 I Lubkin Creek CA-INY-30 I I I I I n=556 IGrant Grove EHT I Grant Grove EHT I - lI h rl-I - - I 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 Kern Plateau I I I I I I I I I I iLL,f o 1 2 3 4 5 6 7 S 9 I I I I La ! I 1 0I ml mI mI mI n=174 Bishop EHT 10 11 12 13 14 15 16 17 16 19 20 21 23 2 24 25 26 ke Isabella/rule River Indian Reservation Figure 25. Frequency Distribution of Coso Hydration Values for Various Projects in Southern Owens Valley and the Southern Sierra Nevada. 169 o I I Contel Project I I I I I n=2 1I Lon I I 0 1 2 3i I K rA J 1 2 I I ri A . . a u - - - I . . c AI t p t I . I .. . .. .- . I.. .. .. 1- .- - - - - - - -- ? I ,., A l A A I J A El - ~ ~ ~ ~ l I? - - --- - -- -- -- - I L JL I I I I I I 4 51 LUBKIN CREEK SITE (INY-30) At INY-30, seven km south of Lone Pine, (70 km NNW of the Volcanic Field), roughly 90% (by count) of the debitage recovered from excavation was obsidian, of which approximately 75% was visually identified as Coso, with pressure flakes accounting for 73.6% of the diagnostic pieces (Gilreath 1988:193). The site contains numerous Marana and late Newberry structures, with a minimal Haiwee period deposit represented in the area investigated. Subsurface debitage densities are highly variable across the site, ranging from 92 to 1663 pieces/m3 (1/8" units) in fill zones overlaying various structural remains. Just over 250 hydration values on Coso specimens are reported (Basgall and McGuire 1988), and the composite curve (Figure 25b) shows a bimodal distribution, with the earliest significant peak falling at values between 4.5-6.0, followed by a peak in values between 2.0-2.9 ,u (Table 82). Accounting for EHT, the quantity of readings increases throughout the Newberry period, peaking at the Newberry/Haiwee period. division, decreases in the latter part of the Haiwee and early Marana period, then rises again in late Marana times. Nearly 70% of the hydration values were obtained on formed tools (Table 82), a plot of hydration rim values on only debitage generates a fairly uniform distribution of values between 1.5-7.4 A. KERN PLATEAU The composite picture of Coso obsidian from the Ke Plateau is the uresult of combining data from several test evaluation reports (sites in the vicinity of Kennedy Meadows reported by Bard et al. 1985 [TUL-896, -897, - 898, and -899]; the Bear Mountain, Lamont, and Morris Peak segments of the Pacific Crest Trail reported by Garfinkel et al. 1980 [TUL-481, -482, -488, KER-743, -748] and McGuire and Garfinkel 1980 [TUL-619, -621, -623, -625, -629, and -636]; and at KER-878 [Schiffman 1988]); and data recovery investigations in Rockhouse Basin (reported by Garfinkel, et al. 1984 (UL-877, -879, -884, -889, and -894]); supplemented by hydration values at other sites in this portion of Tulare County as summarized by TCR/ACRS (1984:139; [TUL-511, -618, -630, -634, -877, -879, -882, -887, -889, -890, 891, -895). In suim, appronximately 550 rim values are reported on Kem Plateau specimens known or likely to be Coso obsidian (Table 82). The cumulative profile shows a bimodal distribution with a low secondary peak at 5.5-6.4 1t, and a broad primary peak centered on 3.0-3.4 IA (Figure 25c). Omitting hydration values on the approximately 100 formed tools has an impact on only the recent end of the hydration sample (Table 82). Approximately 60 artifacts (nearly exclusively Desert and Rose Spring series projectile points) have hydration rim values of less than 3.5 , and when excluded from the curve, the rate at which levels decline after the peak appears accelerated. An historic tenperature record of significant time depth does not exist for the Kern Plateau. In its absence, the record from Grant Grove is used to derive an EHT correction factor. Grant Grove is 90 km northeast of the Kern Plateau, and sits at an elevation of 2030 m (6600 ft), roughly comparable to that of the Tulare County sites listed above. Factoring the Grant Grove EHT correction (Table 83), the composite hydration distribution (Figure 25c) shows obsidian increased in frequency throughout the Newberry period, to a maximal amount in the Haiwee period, followed by a decline throughout the Marana period. A slight secondary peak occurs ca. 3500 years ago, suggesting obsidian from Coso has for a long time been carried up into the plateau. Data recovery excavations at five sites reported by Garfinke et al. (1984) provide robust information regarding the quantity of Coso obsidian and the nature of stone tool reduction practices represented. To engender an appreciation for the quantity of Coso obsidian ta ted about 45 km from the source, of approximately 32,000 pieces of debitage recovered from 1/8" screened deposit, only 126 (< 1.0%) were of material other than obsidian (Garfinkel et al. 1984:8-148). A maximal subsurface debitage density of 11,820 pieces of debitage/m3 is reported for TUL-889, Locus 1, though densities at the other four sites are less impressive, ranging from 25-536/m3. It should be noted, too, that the site area with extraordinary amounts of Coso glass was concluded to date between 4450-650 B.P., and all diagnostic points (n=7) from that deposit were Humboldts. Summary comments indicate that cortical pieces are rare; pressure filking was the prevalent reduction activity, followed by late stage percussion thinning; and only minor amounts of non-biface thinning flakes were present, leading to the interpretation that only bifacially worked blanks were transported into the sites, where they were firther reduced. They conclude: 170 A detailed analysis of the prehistoric lithic technology shows that the bifacial reduction of imported Coso obsidian was one of the main economic activities carried out at these sites. An examination of the lithic debitage and artifactual data clarified the nature of these sites as important lithic workshops in trans-Sierran obsidian manufacturing and exchange networks [Garfinkel et al. 1984:i]. LAKE ISABELLA AND THE TULE RIVER INDIAN RESERVATION On the westem slope of the southern Sierra Nevada, information regarding the relative abundance of Coso glass and associated hydration rim values are reported for two survey projects: Lake Isabella and the Tule River Indian Reservation, both approximately 75 km WNW and WSW of the Volcanic Field, respectively. Reporting results from the survey around LIke Isabella, which lays at an elevation of approximately 800 m (2600 ft), Dillon (1988) presents hydration values on approximately 50 specimens collected from the surface of seven prehistoric sites. Data are not available indicating the relative abundance of obsidian at the sites, other than his passing comment that "only a small percentage of the total amount available was collected, and of the sample, only a further fraction was measured in the hydration lab" (1988:65). Similarly whether the samples represent formed tools or pieces of debitage is not reported, suggesting they are most likely debitage. The Tule River Indian Reservation covers elevations ranging from 340-2215 m (1100-7200 ft), but most hydration specimens were collected from sites lying between 460-1385 m (15004500 ft) (Gehr 1981, 1988). Of 96 prehistoric sites identified in surveys, 28 had obsidian present, indicating an overall decrease in access to Coso glass, relative to that afforded people who lived around Lake Isabella and on the Kern Plateau. The full hydration sample from sites on the reservation consists of 121 items from the surface of the 28 sites. Geochemical analysis of 49 samples identified 88% as Coso obsidian, allowing the reasonable conclusion that, with rare exception, obsidian in this area derived from Coso. Recognizing that the magnitude of the hydration values decreased as elevation rose, Gehr (1988:25) developed a mean temperature gradient based on elevation, and correlated 13.590 C with an elevation of 835 m (3040 ft), and 11.160 C with an elevation of 1430 m (4640 ft). Of the various locations for which EHT correction factors have been figured, Bishop comes the closest to having a roughly equivalent annual temperature. At a value of 12.30 C, it fals between the two aestimates given above for select elevations in the Tule River Indian Reservation. For the sake of parsimony, the Bishop EHT correction factor is applied to the Lake Isabella and Tule River Indian Reservation hydration samples (Figure 25d). The composite hydration curve resulting from the combined samples shows a broad nmodal distribution, with the highest number of readings plotting between 3.5-7.9 g, and a decline in prior and subsequent periods. This distribution suggests maximal exploitation/access to Coso glass during the Newberry period, with availability decreased in the previous Little Lake and subsequent late Haiwee and Marana periods. GREATER LOS ANGELES VICINITY In general, obsidian is scarce in prehistoric sites withiin the greater Los Angeles vicinity. What little occurs has been shown to derive nearly exclusively from Coso (Meighan 1978:27), requiring that it be transported some 235 km from its origin. The composite profiles of Coso hydration values from archaeological sites in Los Angeles, Ventura, and Orange counties were developed by reviewing data reported by the UCLA Obsidian Hydration Laboratory (Meighan and Russell 1981; Meighan and Scalise 1988; Meighan and Vanderhoeven 1978). Though one or two specimens from a variety of sites have been noted, preference was given to sites from which 10 or more hydration values are reportd; in addition, sites with mission-period occupations were similarly excluded from the samle. References for the data compiled and presented in Figures 26a-c and Table 82 are presented in Table 86. The data set for Orange County is based on readings from four sites, totaling 156 readings on 147 specimens; that for Los Angeles County derives from five sites, totaling 90 readings; while two sites with a total of 62 readings constitute the Ventura County sample (Table 86). The EHT correction factor developed for Malibu (Table 83) is applied to convert these values into age estimates. Taking into consideration EHT, the distribution of hydration readings in Orange County indicates coastal groups appear to have had regular access to Coso glass as long ago as 171 I I I I I I I I I I I I I I I I I In .-~ i- ...-IlUNIl I UUf[l...Ir.. n=156 Malibu EHT I n --Jnl I I1 H H n I I1 f Ii la'- i-_ II o 1 2 3 4 5 6 7 8 9 110 11 12 13 14 151617 18 19 20 21 22 23 24 25 26 I I Orange County I I I I I I 1 2 3 4 56 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 28 l l | |Los Angeles County =I I I I I I n=62 i Malibu EHT 0 1 2 3 4 5 6 7 18 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 I I I 4 .-^, - o - 10 a o 30 20 10 b o 30 20 10 Co 420 400 380 360 340 320 300 280 280 240 220 200 10 160 140 120 100 80 60 40 20 d o .1 Li- I I I I I I I I I I I I . I I I Lm L 10 i I I I I I I I I I I I I I I I I I I I I I ventura county n=4111 Haiwee EHT r1' r"l_ nr'l 11 12 13 14 1S 16 17 18 19 20 21 22 23 24 26 26 Coso Volcanic Field II c R I A -4 Figure 26. Frequency Distribution of Coso Hydration Values for Various Projects in Coastal Southern Califonia and the Coso Volcanic Field. 172 30 F 20 I I' 5 ,6 7 I 0 I 2 3 41 I o4j It I I I r I . .--- - - . __ I I I I a t I I I I I I I I I I I I I I I I I I Table 86. Hydration Data Compiled from the Greater Los Angeles Vicinity. Number Number Site of rims of specimens Reference Orange County ORA-64 16 14 Koerper et al. 1986 ORA-119A 23 22 Koerper et al. 1986 ORA-119B 15 15 Cottrell 1988b ORA-244 34 33 Cottrell 1988a ORA-378 10 8 Koerper et al. 1986 ORA-378 58 55 Ericson et al. 1989 Los Angeles County LAN-59 22 20 Meighan 1988 LAN-61 37 34 Meighan 1988 LAN-264 11 11 Meighan 1978 LAN-702 10 10 Cottrell 1981 LAN-774 10 10 Singer 1981 Ventura County VEN-294 17 17 Rosen 1988 VEN-594 45 45 Cottrell 1988c ca. 4500 years ago, which stayed relatively stable throughout the Newberry period, and was disrupted in the mid- Haiwee period (Figure 26a). Distributions for Los Angeles and Ventura county sites indicate stable access to Coso glass began considerably later, during the middle Newberry period, but also was disrupted in the mid-Haiwee period (Figures 25b and c). ANTELOPE AND FREMONT VALLEYS Though a substantial amount of archaeological investigations have taken place in Antelope and Fremont valleys, few published reports document the fi . Consequently, the assessment of the relative abundance and temol distribuon of Coso glass in prehistoric sites here is considerably more impressionistic than that offered for other areas. As Sutton (1988a:61) points out, in c st to the southem Sierra Nevada/Tehachapi area, obsidian is uncommon in the desert area immediately to the south. Nonetheless, "all of the obsidian so far sourced from western Mojave Desert sites has come from the Coso Volcanic Field" (Sutton 1988b:43). Small quantities of Coso obsidian were brought into the area as early as the Pinto/Little Lake period, as evidenced by a cluster of 35 rim values in excess of 9.0 y from Edwards Air Force Base (Greenwood and McIntyre 1981:Appendix 10) and by the few Pinto series Coso points (one with a hydration band of 10.5 i) that have been found. In reviewing the archaeological record for the western Mojave Desert, Sutton notes that obsidian is more common in Fremont than Antelope valley, most obsidian occurs as finished tools, and a higher proportion of Elko and Rose Spring series items than Desert Series points found in the area were made of Coso glass (1988b). More recently, York (1992), working with data from Edwards Air Force Base, noted that Coso obsidian debitage tends to be late Newberry (mostly) or earlier in age, but that Coso projectie points recovered are most commonly late period forms. Citing burial pces, site structure, and the abundance of California marine-shell oraments in burial contexts, Sutton concludes "the western Mojave Desert was integrad into the mainsteam of events of southern California prehistory," and tentatively suggests that KER-303, and perhaps other similar large midden sites in the area, might have served as trade centers between 3000-300 B.P., where marine shell and Coso obsidian, among other commodities, were exchanged. Both Sutton and York perceive a major reduction in the quantity of Coso glass brought into this area in late times (cf. Byrd et al. 1994). 173 DISCUSSION Review of the distribution of Coso obsidian throughout southem California supports three major trends. First, Coso obsidian was the pnmaiy glass used by prehistoric peoples who occupied lands as far east as Fort Irwin, as far nordth as Lone Pine, throughout the southem Sierra south of Mt. Whitney (see Roper-Wickstrom 1992), west to Porterville, southwest toin the Paciefic coasnt in as far as the border, and as far south as Lancaster/Palmdale. Second, outside of the procurement zone, impressive quantities of Coso glass occur in only two areas, at major occupation sies within 15 km of Sugarloaf Mountain near the eastern base of the Sierra (e.g., INY-372 and -2284), and some 40 km west on the Kem Plateau. The quantity of Coso obsidian in sites in the southem third of Owens Valley, as far north as Lone Pine, indicate people living here also had ready access to the material. Only minimal qantities are recovered to the east and south. Third, nearly all areas considered show a growing increase in the quantity of obsidian during the Newbery period, with the quantity peaking near the Late Newberry/Haiwee period boundary, and declining, usually rapidly, in the latter part of the Haiwee period and continuing to fall during the ensuing Marana period. The composite hydration profile for the Volcanic Field (Figure 26d), without consideration of sampling bias, suggests exploitation/production was at its zenith during the Late Newberry period, with 35.9% of the rim values falling between 5.5-7.4 y (Table 82), and that the intensity of exploitation was comparatively low during the Haiwee period - a pattern at odds with overall trends outside of the Volcanic Field showing peak consumption typically in the 'Haiwee period. Reasons for this incongruity lay in the differing exploitation patterns and overall land-use practices operative during the Late Newberry and Haiwee periods, as well as the sampling bias inherent in the Volcanic Field's composite hydration profile. Figure 26d is based on samples from numerous sites (n=381) spread over an extensive area (15,000 acres), with many sites, regardless of how sparse or dense their flaling debris, having near equal representation. In large part, then, the profile reflects the relative number of sites by time in the Volcanic Field. An overview of the archaeological resources within the Volcanic Field (Gilreath and Hildebrandt 1991) reaffrms this contention, showing LTe Newberry off-quarry sites to be the most numerous type of resource in the Sugarloaf District (Table 87). Table 87. Single Period Loci Densities Standardized to 100-year Spans in the Coso Volcanic Field (from Gilreath and Hildebrandt 1991:Table 19). TEMPORAL PERIOD Late Middle Eady Little Marana Haiwee Newbery Newberty NewberIy Lake Early Quarry loci 0.67 1.60 2.87 4.0 1.86 1.65 0.63 0.18 Secondary Reductionloci 0.66 4.16 8.78 4.2 2.57 0.95 0.84 0.45 TOTAL 1.33 5.76 11.65 8.2 4.43 2.60 1.47 0.63 100 year spans 4.50 6.25 10.25 5.00 7.00 20.00 30.00 70.00 NOTE: ' - excludes "Milling Station/Features and Milling Camps. In the preceding chapters we showed that prior to the Middle Newberry period, widely dispersed, typically small pockets of lag deposits were used, with exploitation shifting to major obsidian outcops in Late Newberry times - a pattem continuing into the Haiwee period. That obsidian quarrying in the Late Newberry and Haiwee periods was confined to few massive obsidian exposures instead of less pleIl but more widespread deposits is a secondary source of distortion in the composite hydration profile. Te Newberry and, to a far grater degree, Haiwee quarries contribute little to the profile because relatively few exist. Results of our investigations parallel those of Gilreath and Hildebrandt (1991) in finding that Late Newberry groups exploited more primary deposits than did Haiwee period groups. It is suggested that obsidian exploitation in the Haiwee period was nearly exclusively restricted to the massive exposure on Sugarloaf Mountain, with other deposits, by comparison, largely ignored. A further 174 distinction between Late Newbeny and Haiwee quanrying practices involves the use (or lack thereof) of secondary reduction stations. Secondary reduction locations are numerous and widespread throughout the Volcanic Field, and the vast majority date to the Late Newberry period. In addition, the Late Newberry period saw the inception of several highly specialized stone tool production sites focused on the manufacture of biface blanks or preforms - a task differentiation not observed in prior times. Secondary reduction locations within the Volcanic Field are far less common in the Haiwee period. Secondary reduction must have occurred either at the quarry or at sites outside the Volcanic Field, like the Coso Junction Ranch and perhaps the Rose Spring sites. Assuming that the level of tool stone reduction in Middle Newberry and earlier times directly correlates to resident population's toolstone needs, then the significant increase in the production level in the Late Newberry period is excessive, even when a healthy population growth during this interval is conceded. Evidence of increased production levels during the Late Newberny period is found in the increased number of secondary reduction locations and intensity of manufacture at primary quarries. Subsurface debitage densities of 1000-3000 flakes/m3 are recorded at Late Newberry and Haiwee period primary quarries, while densities of 200-400 flakes/m3 are recorded at earlier lag quarries (Chapter 6, Table 61); i.e., one cubic meter of deposit at a primary quarry generates on average as much chipping debris as seven cubic meters of lag quarry deposit. The magnitude of differing production levels becomes more extreme when one appreciates the fact that primary quarries maintain such high densities over areas substantially larger than lag quarries. Up through the Early Newberry period, fairly scant lag deposits were exploited, production intensity was low, reduction activities and the flaked stone artifact repertoire were fairly diverse witiin any one location, and sites were widely spread across the lasape. All factors combine to suggest exploitation by mobile, small groups, who quarried as the opportunity afforded itself, to provide themselves with satisfactory quanttiies and kinds of obsidian for their own needs, within the constraints of their normal subsistence pursuits. The exploitation pattern undergoes changes roughout the Middle Newberry period, and by the Late Newberry period only pimay quarries are in use, the intensity of production increases magnificently, and secondary reduction locations are abundant with the focus of reduction activities varying from one location to the next. The stoneworking activities at Coso during this period represent a healthy mixture of reduction for personal consumption as well as surplus production for exchange. It is during this period that people living throughout sothem California have increasing access to Coso obsidian, with the quantity of Coso glass rising sharply in many distant locations. While access to Coso obsidian appears to have been unrestricted into the Latie Newbery period, several traits suggest some measure of control by the start of the Haiwee period. The radical decrease in the Late Newberry period in the number of obsidian deposits quarried provides the first indication of control. Obsidian is available at Coso over an area covering about 17 km north-south by 7 km east-west, and it is difficult to imagine that groups coming from diverse regional locations all would have ignored the lag deposits, choosing instead only the same few primary exposures. By the Haiwee period quarrying was restricted to even fewer and only the very best exposures. The location of nearby large Haiwee-period villages would have facilitated restricted access. People occupying the Coso Junction Ranch site (on the western front of the Volcanic Field) directly overlooking the Coso obsidian motherlode, and the Rose Spring site (at the narrows separating Owens and Rose valleys) would have been able to effectively monitor its access. Extraordinary amounts of secondary reduction took place at both these sites, suggesting they were key nodes in the exchange system that was well developed by the end of the Late Newberry period. It is easy to imagine that impressive quantities of Coso obsidian were moved north from the Rose Spring site, into the hands of neighboring, and probably closely related Owens Valley populations. In a similar fashion, large quantities of Coso obsidian moved from habitation/secondary reduction workshops like the Coso Junction Ranch site, up southeem Sierra canyons, into the Kern River drainage, and from there into the hands of south coast groups. 175 7iS CHAPTER 9: SUIMMARY AND CONCLUSIONS Obsidian toolstone is abundantly present in the Coso Volcanic Field. Nodules are available at numerous lag deposit quarries, while larger concentrations of higher quality material are also found at a limited number of primary seams. Subsistence resources are much less abundant, primarily restricted to a variety of seeds and small game. Lacking a broader range of foods, the Volcanic Field never supported more than a small portion of any subsistence-settlement system. These characteristics, although limiting on some levels, have produced an archaeological record sensitive to the changing role of obsidian within both local and inter-regional economic systems, as well as useful in determining how marginal environments relate to higher-order issues of subsistence-settlement pattern change. After providing a brief summary of the major project findings, we conclude the report with discussions of comparative data obtained from the Casa Diablo and Bodie Hills procurement zones, and address problems associated with exchange and its relationhip to sociopolitical complexity in California, the Great Basin, and other regions as well. LOCAL CHRONOLOGICAL SEQUENCE Because so much of the Coso archaeological record consists of stone working areas lacking materials traditionally used for dating (e.g., organics for radiometric dating, diagnostic artifacts for cross-dating), chronological ordering of the project assemblages relied largely on the use of obsidian hydration data. Conversion of hydration readings to calendric years was accomplished using a rate developed by Basgall (1990) based on a series of hydration-radiocarbon pairings obtained from a nearby site (INY-30), augmented with an EHT correction factor developed from pairings gathered from numerous other sites located in contrasting environmental settings. When applied to projectile points recovered from the Volcanic Field, the rate conversions successfully placed most diagnostic forms in their expected temporal interval. Beginning at around 3400 B.P. and moving forward, the sequence included Thin Elko forms (<6.5 mm thick [mean=7.4 i, s.d. =1.0]), Humboldt Basal-notched (mean=6.3 it, s.d.= 1.0), Rose Spring (mean=5.2 it, s.d.=0.8), Saratoga Spring (mean=4.8 it, s.d.=0.8), and Desert series (mean=3.0, s.d. = 1.2). Contrasng with this clean unilia progression, point types pre-dating 3400 B.P. included Thick Elko (mean=12.3 g, s.d.=3.3), Little Lake/Pinto (mean=14.2 A, s.d.=4.3), and Great Basin Stemmed (mean= 12.9 /, s.d. =2.7). All exhibited large, overlapping standard deviations, perhaps reflecting: (1) contemporaneous use over long periods of time (Flenniken and Wilke 1989); (2) problems in hydration rim development magnified by several thousand years of exposure to variable environmental circumstances (Rhode 1992); and/or (3) tool rejuvenation practices (Flenniken and Raymond 1986). Review of projectile point hydration data from surrounding areas produced the same relative sequence, varying only with respect to local differences in temperature (i.e., cooler areas produced relatively smaller rim values). Development of a local sequence for the Volcanic Field was based on hydration data obtained from several hundred single period loci recorded during extensive surveys of the area (Gilreath and Hildebrandt 1991). Organized into two populations (quarry versus off-quarry), hydration frequency distributions revealed several shifts in the intensity and character of obsidian production. These shifts resulted in the formulation of eight periods, most corresponding to Bettinger and Taylor's (1974) regional chronological scheme. Beginning at the recent end of the sequence, the Marana (650-200 B.P.) and Haiwee (1275-650 B.P.) periods correspond rather closely to the original construct. The Newbeny period was subdivided into Late (2300-1275 B.P.), Middle (2800-2300 B.P.), and Early (3500-2800 B.P.), while the Little Lake period was assigned to the 5500-3500 B.P. temporal interval. Although Gilreath and Hildebrandt (1991) defined two additional earlier periods, the standard deviations on hydration clusters reached such magnitudes that attempts at high resolution dating seemed inappropriate. Because of these difficulties, pre-5500 B.P. accumulations were simply considered Early. PREHISTORIC LAND-USE CHANGE WlTHIN THE COSO VOLCANIC FIELD Early (pre-5500 B.P.) and Litde la (5500-3500 B.P.) period use of the Volcanic Field was largely limited to short-term use of lag quarry deposits. Lacking firm evidence of subsistence oriented activities, most areas reflect the on-site production of cores, early stage bifaces, and a lesser number of unifacial items. The large number and variable size/quality of these areas indicates a rather generalized, opportunistic approach to the acquisition of 177 obsidian. This is also reflected by relatively low densities of debitage (dominated by shatter and core production debris) and the presence of variable tool morphologies, the latter showing little preference for one form over another. When combined with archaeological data obtained from the surrounding area, the current findings indicate that Early and Little Lake period lithic procurement activities were largely embedded within subsistence-settlement systems characterized by high degrees of residential mobility, minimal use of seed resources, a probable emphasis on hunting both large and small game, and little concern with the inter-regional exchange of obsidian. addition to the exploitation of lag quarries, Early Newberry and Middle Newberry period components (3500-2300 B.P.) showed a significant increase in the use of off-quary, biface production areas. Increased concern with biface manufacture was also found among the lag quarries, where bifaces were more abundant than cores, and biface reduction debris was also significantly present. The off-quanry areas, as would be expected, were dominated by Stage 2 and 3 bifaces, and variable densities of bifacial thinning debris. Some of these sites, as well as one of the quarry locations, also produced a variety of other artifacts ncluding projectile points, milling equipment, and numerous flake tools. Although some of the materials certainly indicate use of local subsistence resources, most occupations remained short-term, and primarily concerned with the acquisition of obsidian. The trend away from lag quarry use reached a peak during the Late Newberry period (2300-1275 B.P.). Of these 23 components, only two were quarries, both primary outcrops on Joshua Ridge. This shift corresponds with Elston and Zeier's (1984) proposals regarding the declining importance of lag quarries, as well as their discovery of two major Late Newberry (our dating) mi n i g areas on the Sugarloaf/West Sugarloaf primary quarry complex. The primary deposits investigated showed that stone working activities focused on the highest quality seams, and produced assemblages with relatively high frequencies of bifacial cores, and early stage (often triangular shaped) bifaces that were sub ally larger than equivalent items found at lag quarries. Off-quarry sites consisted almost entirely of biface production areas. Most were dominated by Stage 2 and 3 bifaces (significantly smaller than their quarry counterparts), and high densities of biface reduction debris. Some differentiation of the reduction activities occurs among these secondary reduction areas: biface blank manufacture was the focus at some, while preform manufacture was the focus at others. Some of these sites also produced subsistence-related materials (mostly millingstones). Lacking accumulations of midden, plant macrofossils, and faunal remains, most of the aforementioned tools represented casual use of plant foods to support individuals focused on the production of bifaces. The Late Newbeny changes in production organization were no doubt related to an expanded network of obsidian exchange. Increased deands for volcanic glass were no longer met by the expedient use of lag quarry deposits. Instead, the less accessible primary seams were mined for materials that were subsequently refined into blanks and preforms at nearby biface production areas. Corresponding changes in land-use, although present, do not evidence dthe emergence of sedentary populations or local control of the quarries. Data collected from Owens and Rose valleys (Basgall and McGuire 1988; Delacorte and McGuire 1993) reveal a Late Newberry subsistence- settlement system maintaining a relatively high degree of residential mobility. In contrast to previous systems, however, annual movements were less extensive and followed a more regularized schedule. Some groups traveled along a north-south axis at least 200 km in length (Basgall 1989), occupying a series of seasonal residential bases. Many of these sites were reused on a regular basis, judging by the presence of structures, features, and caches, and supported to some degree by logistically organized forays to distant resource areas (e.g., pinyon, mountain sheep, and marmots in lowland sites; see Basgall and McGuire 1988). Increased predictability of setdement locations during the annual subsistence-settlement cycle is hypothesized to have allowed regularized exchange relationships to develop. This is supported by increases in the frequency of Newberry period hydration readings at Rose Spring (Yohe 1992), Coso Junction Ranch (Allen 1986), several sites along Rose and Owens valleys (Basgall and McGuire 1988; Delacorte and McGuire 1993) and, importantly, more distant localities such as the Kem Plateau (Garfinkel et al. 1984; McGuire and Garfinkel 1980), west slope of the southern Sierra (Dillon 1988; Gehr 1981, 1988), and various locations within Ventura and Los Angeles counties. It is also important to note that desert areas to the soutieast did not produce similar increases (Basgall 1991; Bouey and Mikkelsen 1989; Gilreath et al. 1987; Hall 1992), rather the increases are largely confined to the Haiwee interval, while the composite profile from several sites in Orange County showed only a marginal Newberry period rise. Although far from conclusive, these trends suggest that most Late Newberry production surpluses were ultimately consumed by the relatively large population centers of coastal southem California. 178 Use of the Volcanic Field dropped significantly during the Haiwee period (1275-650 B.P.). Lithic production areas investigated were essentially limited to a single biface production station and one primary quarry. Seed processing areas, which dominate in the Marana period, were also quite rare, represented by a few scattered examples. This drop in obsidian production could be attributed to collapse of the Late Newberry exchange system, perhaps related to changes in technological orientation, i.e., a reduced demand for toolstone caused by the introduction of the bow and arrow (see Skinner 1989; Wilke and Flenniken 1988; cf. Singer and Ericson 1977). When moving to the hydration profiles developed for the outlying areas, however, a different picture emerges. The Newberry period rise in hydration values was maintained through the early half of the Haiwee period in nearly all settings, while new peaks were established only in the desert areas to the southeast. This apparent contradiction (i.e., little production activity within the Volcanic Field during the Haiwee period but ample consumption elsewhere) can be resolved by more closely examining Haiwee period production. Only two Haiwee period quarries have been investigated thus far, one during the current project on Joshua Ridge and the other by Elston and Zeier (1984) on West Sugarloaf Mountain. Both locations were intensively used, focusing on specific concentrations of high quality material, and the quaying was strctured in a way that eliminated the need for secondary reduction locations within the procurement zone. That large numbers of people from diverse localities would have traveled to the Volcanic Field, used such a limited number of quarry locations, and nor left indications of their stays in surrounding non- quarry contexts seem unlikely. Instead, these changes probably reflect production by a limited number of local people who regularly exploited favored seams of obsidian, and transported the products to residential bases outside the Volcanic Field for further reduction (e.g, Coso Junction Ranch). Reduced access to the Coso quarries by non- local people is also supported to some degree by a region-wide reduction in mobility. Bettinger (1989) and Delacorte and McGuire (1993), for example, argue that seasonal movements within the Owens Valley/Rose Valley area became increasingly confined during the Haiwee period, perhaps facilitating local control of an expanding system of inter-regional exchange. By the Marana period (post 650 B.P.), evidence for local flaked stone production is scant, as is consumption in outdying areas. Just when mobility had been reduced to a point where some level of control over the quarries could have been achieved, the obsidian production and exchange system collapsed, and the Volcanic Field became the focus of intensive seed processing activities. With the exception of a few locations where relatively long-term occupations took place, most Marana period areas were temporary seed camps consisting of milling equipment, hearhs, and little else. Judging by the high number and wide distribution of these sites, and the relatively large size of the millingstones contained within them, it appears they were established adjacent to productive resource patches and reused on a regular basis. RESOURCE INTENSIFNCATION IN THE COSO VOLCANIC FIELD Increased use of small seed resources within the Coso Volcanic Field is consistent with a major late period shift in subsistence orientation that Bettinger and Baumhoff (1982) think helped fuel the spread of Numic speaking populations throughout much of the Great Basin. According to the Bettinger and Baumhoff (1982) model, pre- Numic peoples used a "traveling strategy" where a substantial amount of tie was spent pursuing game and covering large tracts of land in search of highly-ranked plant resources. Numic groups followed a "processing strategy" where travel time was reduced due to an ineased use of lower-ranked plant resources (mostly small seeds) widely available in most environmental settings. Although encumbering less processing time per unit of output, the pre- Numic strategy supported relatively low population densities and required high degrees of mobility. The Numic strategy, in contrast, increased population density and lowered mobility through intensifying the use of a broader range of resources within progressively smaller foraging areas. Simply stated, the increas"img number of Numic peoples, combined with their ability to occupy environmental zones previously under-used by pre-Numics, established the competitive advantages that ultimately led to the spread of Numic speakers throughout the Great Basin. Late period expansion into previously under-used environments has also been addressed by Bettinger (1991) as part of his ongoing research project in the White Mountains. Prior to 1400 B.P., most upland areas were used almost exclusively for huntig. Later components are quite different, containing house structures associated with artifactual debris related to hunting, vegetal processing, and a variety of other domestic activities - no doubt reflecting relatively long periods of occupation. This land-use shift was thought to have been a response to regional 179 population increases that lowered the rates of return for lowland foods, thereby increasing the relative value of resources found in the uplands (Bettinger 1991). The Coso Volcanic Field also provides an excellent example of late period resource intensification within a previously under-used area. Largely restricted to seeds and small game, the local subsistence base was of limited value relative to surrounding areas from which a wide variety of both upland (acorns, pine nuts, deer) and lowland (fish, shellfish, brine fly larvae, waterfowl, greens and shoots, roots and tubers) resources could be had. Not surprising, little evidence exists in the project area for the use of local foods prior to the Late Newberry period. During the Late Newbery-Haiwee interval of expanding obsidian production, some subsistence pursuits did occur, but most were in support of the extensive stone working activities that took place. With the advent of the Marana period, temporary seed processing areas proliferate across the Volcanic Field. Largely restricted to milling equipment and hearths, the sporadic distribution of these sites suggests that most were located adjacent to seed producing areas rather than in places conducive to longer-term occupation. Moreover, most contained relatively large milkingstones selected for long term durability rather than ease of transport, clearly indicating the desire to reuse particular seed patches in the future. Due to the ephemeal nature of these areas, floral and faunal remains were rarely associated. Where such associations existed, animal bone was limited to rabbit-sized taxa and smaller, while the plant remains icluded nrcegrass, goosefoot, tansy muard, blazing star, chia, and desert tomato. As noted by Delacorte (1990), the interval of time between when seeds become ripe for harvest, and fall to the ground unused or are lost to competing animals, is extremely short. It follows, therefore, that a full use of this resource base required keen lmowledge of the distribution and condition of local seed-bearing plants, a knowledge attainable only by people occupying relatively limited tracts of land. COMPARISON WITH OTHER OBSIDIAN PRODUCTION SYSTEMS To gain better insight into the factors possibly involved with the collapse of the Late Newberry-Haiwee period production system at Coso, we look at general trends in th se use of obsidian from sources in northern Califor ia and southern Oregon, as well as those observed in the Casa Diablo and Bodie Hills procurement areas. Similar to Coso, Casa Diablo and Bodie have mlple obsidian flows (e.g., Hughes 1994a), and reveal production peaks during the Newberry and early Haiwee periods, followed by major declines thereafter (Bouey and Basgall 1984; Goldberg et al. 1990; Hall 1983; Hall and Basgall 1994; Jackson and Ericson 1994; Singer and Ericson 1977). The decline of obsidian production ca. 1300 B.P. also seems to have been widespread throughout northem Califo and Oregon. Fredrickson and White (1988:81) report a significant drop at about 1500 B.P. in obsidian production at the Borax Lake source in the North Coast Ranges of California. This is also the case for Medicine Lake obsidian sources in northeastemn California (Hildebrandt et al. 1994:8-58). The trend continues even further north, occurrng at Newberny Crater near Bend, Oregon (Skinner 1995a:5-39). Moreover, individual major obsidian flows formed since 1300 years ago - the Glass Mountain Flow in the Medicine Lake Highlands (Hughes 1982) and the Big Obsidian Flow at Newberry Crater (MacLeod et al. 1982) - show little evidence of use. Less than 2.0% of the 5,600 pieces (XRF) id fied as Medicine Lake Highlands glass match the geochemical profile for the Glass Mountain Flow (Gilreath 1997:7), while only 27 (1.0%) of over 2,700 pieces identified as Newberry Crater glass match the profile for the Big Obsidian Flow (Skinner 1995b:4-31). Obsidian hydration data on Casa Diablo materials collected from the west side of the Sierra Nevada roughly parallel the production profiles, and are thought to represent a trans-Sierran exchange system initiated by the development of large sedentary consumer populations living across the divide (Goldberg et al. 1990; King 1976; Moratto 1972; Moratto et al. 1978). Various scenarios have been forwarded to explain the late period decline of this system. Hall (1983) argues that recurrent volcanic eruptions in the eastern Sierra may have restricted access to the Casa Diablo quarries, thereby inhibiting the flow of obsidian across the mountains. Jackson and Dietz (1984), Moratto and Davis (1988), and Goldberg et al. (1990), altematively, suggest the rapid onset of drought during the Haiwee period disrupted the socio-political organization of west-side populations, in tur undermining existing trade relations by which Casa Diablo glass was dispersed. 180 Neither of the enviromental explanations proffered for Casa Diablo adequately accounts for the events that appear to have been occurring on an inter-regional scale. Looking first at Hall's hypothesis, if volcanism alone forced the decline in Casa Diablo exploitation, it is reasonable that exploitation increased or expanded in alternative quarrying zones to satisfy demand. Coso is the closest substantial obsidian quarrying zone south of Casa Diablo (Map 6, [Fish Springs is by comparison small]); it was not affected by late Holocene volcanism in the Inyo-Mono chain; it displayed no exploitation increase to off-set the decline in the availability of Casa Diablo obsidian. Although it is difficult to know the geographic extent of the Haiwee-period drought events (see Stine 1994), it is unlikely that obsidian quarries located considerably to the north (e.g., in southeast Oregon and northwest California) would have been affected in the same way as those in the more arid regions of southeastern California. Goldberg et al. (1990:179) also consider the importance of technological change, stating that "It is tempting to posit that reduction in production at the Casa Diablo quarries was related chiefly to technological changes, which, beginning during the Haiwee Peiod with dminished size of flakded stone tools, required smaller pieces of obsidian for tool production." Much of this material, it is further argued, could have been obtained through scavenging off local sites where obsidian was deposited during earlier periods of occupation (Skinner 1989; Wilke and Flenniken 1988). Ericson (1982) has also attributed production shifts to technological change, but argued that introduction of the bow and arrow increased demand for glass, resulting in export of unmodified cores that were subsequently reduced elsewhere, leaving little or no evidence at the quarries (see also Jackson 1988). We have shown that this latter proposal is not bome out by the consumer hydtion profiles obtained from much of southem California. Similar to our interpretations of the Coso sition, Bouey and Basgall (1984) argue that Newberry period adaptations along the eastem Sierra were characterized by high residential mobility and little or no territoriality qalites they view as incongruent with the development or persistence of a large-scale production system. They thought it more likely that groups living on the western slope traveled to Casa Diablo, directly procuring obsidian themselves, and t trading it to populaterions occupying Califoia's Centl Valley. The ultimate demise of the system occurred when east side populations bAecame increasingly territorial (see Bettinger and Baumhoff 1982) and direct access to the source was constrained. Given the stiking similarities between Casa Diablo, Bodie, and Coso production curves, as well as those from northern California and Oregon, it seems probable that similar sets of causative factors influenced their development and demise. A combination of two factors provides the best explanatory power: increased territoriality and technological change. Similar to the scenario forwarded by Bouey and Basgall (1984), the acquisition and distribution of obsidian reached peak proportions when various groups had access to the quarries. Whether they originated from across the Sierra Nevada, or were east-side people conducting exchanges with westerners and southemers during a regularized seasonal round, the crucial point is there was a demand for obsidian and little or no constraints inhibiting its acquisition. Later in time, decreased mobility accompanied by increased population density and territoriality (Bettinger and Baumoff 1982), may have constrained free movement across the landscape and ultimately complicated the flow of obsidian (i.e., the number of exchanges per unit distance traveled increased). These developments, coupled with decreasing demnds due to changes in flaked stone technology (Goldberg et al. 1990; Wilke and Flenniken 1988), compounded by increasing exploitation of plant resources, significantly decreased the importance of toolstone throughout much of southern California. EXCHANGE, CRAFT SPECIALIZATION, AND SOCIOPOLMCAL ORGANIZATION Overviews of California prehistory routinely tell a tale of population growth and increasing social complexity, resulting in the "extraordinary variability of California cultes" (Jennings 1975:176). This is evidenced by Heizer's (1978:ix) tribal territories map for Califonia. This is also indicated by the high degree of competition in late times for select resources with limited natural distributions. Aikens (1978:143), for example, states, "evidence of growing social complexity might be seen in the increase throughout central and southeem Califoria over the past several thousand years of both violent deaths and trade goods." From such accounts, we have come to assume that trade was more important in the late period that in earlier times, and since California ethnographic studies often list obsidian among the items traded, it is often accepted as fact that late period obsidian trade was widely practiced on a significant scale. Archaeologists working elsewhere in the world often have the same expectations. 181 Recent research in the American Southwest indicates that the production and exchange of obsidian reached peak proportions during the Classic period, after 850 B.P. (Mitchell and Shackley 1995; Shackley 1995). This period of cultural complexity is evidenced by several large Hohokam settlements in southern Arizona that were supported through the intensive use of irrigation agriculture and the widespread exchange of a variety of items. Common trade goods recovered from archaeological contexts include pottery vessels, marine shell, obsidian, and turquoise. This perceived correlation between late period cultural complexity and the increased importance of obsidian exchange has also influenced prehistoric studies in Alta California and the Baja peninsula, where several researchers have argued for parallel patterns of development among hunter-gatherer populations in these areas (e.g., Bettinger and King 1971; Ericson 1984; Fredrickson 1974; Heizer 1974; Shackley et al. 1996). Hughes (1994b:364) refers to this view as "the incremental model of culture change" where the "prevailing view was that the earliest peoples rarely, if ever, engaged in much trade, but that the frequency increased steadily through prehistoric time, culminating in the flow of commodities documented in ethnographies." This perspective is clearly illustrated by Ericson's (1984) attempt to explain the origin of formaized exchange systems among hunter- gatherers: Sedentism, which changes people-land relationships, requires increased scheduling and dependability of resources. Sedentism, population growth, and growing dependency of the population on regional resources as well as the establishment of territoriality within the region favor the growth of a regional exchange system [1984:7]. He also argued that large-scale exchange systems often used the labor of occupational specialists which usually required the centralized redistribution of surplus resources to support them. Torrence (1986) and Amold (1987), focusing on the archaeological implications of such phenomena, have proposed the following indicators of occupation specialization: (1) very high volumes of production materials; (2) identifiable workshops separated from other subsistence/residential areas; (3) distinct patterns in regional distribution, reflecting organized and controlled production and exportation; (4) high degrees of technological standardization and low rates of production failures; (5) control over critical resources; and (6) specialist's tools in certain burials. Direct access to quarry areas by non- specialized producers, in contrast, would be evidenced by a wider range of reduction technologies and end-products (Ericson 1984). While the latter prediction broadly corresponds to assemblages and adaptive patterns assigned to the Early, Little Lake, and Early/Middle Newberry periods at Coso, the Torrence-Arnold criteria - though closely approximating the Late Newberry archaeological record - would no doubt misinterpret the degree of organizational complexity involved. Late Newberry production specialization is clearly indicated by spatially discrete workshop areas, large volumes of obsidian debitage, high degrees of technological standardization and success, and distinct regional distributions of material (flowing largely to the west and souiwest). Rather than representing occupational specialization, we conclude that these characteristics are a likely outcome of logistically organized groups sent to the Volcanic Field for the specific purpose of producing bifacial blanks and preforms. Such forays probably originated from a seasonal residential base located outside the Volcanic Field, and would have been organizationally equivalent to hunting parties sent to upland areas in pursuit of deer. Haiwee period production organization presents a different set of interpretive difficulties. Lacking biface production areas within the procuremen zone, most toolstone (in reduced forms) was probably moved directly from the primary quarries to outlying residential bases where further modification took place (e.g., Coso Junction Ranch). Although substantial amounts of obsidian were reduced for exchange, perhaps by individuals controlling access to the quarries, the archaeological record provides few hints as to the degree of production specialization involved. Moving to the late prehistoric-ethnographic periods, Bettinger and King (1971) have also considered the importance of obsidian exchange in their analysis of sedentary settlement structure among the Owens Valley Paiute. Rather than attributing the late period emergence of permanent villages and non-egalitarian social organization to the rich resource base of Owens Valley (see Steward 1938), they ps that mtergroup exchange was a necessary component of the system because: 182 ... in the face of increasing social circumscription, trade provides a means whereby local populations can be maintained at levels higher than would be possible on the basis of local resources alone [Bettinger and King 1971:142]. Obsidian was viewed as an integral component of this system because it could be used to obtain shell beads from the west, which could in turn be traded for foodstuffs during periods of subsistence stress. Most bead-for-food exchanges occurred during inter-village fiestas where groups from a variety of locations would converge, and redistribute resources from areas with surpluses to those experiencing deficits. Bettinger and King (1971) further predicted that the development of the fiesta-shell bead economic system coincided with a dramatic increase in trans- Sierran obsidian trade. Although some aspects of the socio-economic relationships outlined above no doubt existed (e.g., the redistributive fiesta system is clearly documented in the ethnographic record), the role of obsidian - at least as seen from the archaeological records of Casa Diablo and Coso - appears to have been over stated. Obsidian exploitation at Coso was at its lowest level ever in the late period. The fact that most ethnographic accounts rarely mention obsidian, and when they do, rank it below more favored trade items such as salt, basketry, hides and pelts, shell beads, acoms, and fish (see Bettinger 1982a, 1982b; Davis 1961; Ericson 1977), provides further evidence of the reduced importance of volcanic glass as a late period commodity of exchange. 183 REFERENCES CITED Adams, C. J. 1986 CA-MNO-833, An Obsidian Stoneworking Camp in Long Valley, Mono County, California. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. Aikens, C. M. 1978 The Far West. In Ancient Native Americans, edited by J. D. Jennings, pp. 131-182. W. H. Freeman and Company, San Francisco. Allen, M. W. 1986 The Effects of Bow and Arrow Technology on Lithic Production and Exchange Systems: A Test Case Using Debitage Analysis. Unpublished Master's thesis, Department of Anthropology, University of California, Los Angeles. Arnold, J. E. 1987 Craft Specialization in the Prehistoric Channel Islands, California. University of Califomia Publications in Anthropology 18. University of California Press, Berkeley. Bacon, C. R., R. MacDonald, R. L. Smith, and P. A. Baedecker 1981 Pleistocene High-Silica Rhyolites of the Coso Volcanic Field, California. Journal of Geophysical Research 86:10233-10241. Bacon, C. R., D. M. Giovanetti, W. A. Duffield, G. B. Dalrymple, and R. E. Drake 1982 Age of the Coso Formation, Inyo County, California. U. S. Geological Survey Bulletin 1527. Bard, J. C., C. I. Busby, and R. J. Dezzani 1985 Kennedy Meadows Cultural Resources Data Recovery Project. Submitted to the USDA Forest Service, Sequoia National Forest, Porterville, California. Basgall, M. E. 1983 Archaeology of the Forest Service Forty site (CA-Mno-529), Mono County, California. Submitted to the USDA Forest Service, Inyo National Forest, Bishop, California. 1984 The Archaeology of CA-MNO-1529: A Secondary Reduction Site in Mammoth Lakes, Mono County, California. Submitted to the USDA Forest Service, Inyo National Forest, Bishop, California. 1987 Paleoindian Occupation in Centl-eastern California: The Komodo site. Current Research in the Pleistocene 4:50-54. 1989 Obsidian Acquisition and Use in Prehistoric Central-Eastern California: A Preliminary Assessnt. In Current Directions in California Obsidian Studies, edited by R. E. Hughes, pp. 111-126. Contributions of the University of California Archaeological Research Facility 48. 1990 Hydration Dating of Coso Obsidian: Problems and Prospects. Paper presented at the 24th meeting of the Society for California Archaeology, Foster City. 1991 The Archaeology of Nelson Basin and Adjacent Areas, Fort Irwin, San Bernardino County, California. Submitted to the U. S. Army Corps of Engineers, Los Angeles District Office, California. 185 Basgall, M. E., and M. C. Hall 1992 ArchaeoLogy of the Awl Site (CA-SBR4562), Fort Irwin, San Bernardino County, California: An Early Holocene Residential Base in the North-Central Mojave Desert. Submitted to U. S. Army Corps of Engineers, Los Angeles District Office, California. Basgall, M. E., and K. R. McGuire 1988 The Archaeology of CA-INY-30: Prehistoric Culture Change in the Southern Owens Valley, California. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. Basgall, M. E., and D. L. True 1985 Archaeological Investigations in Crowder Canyon, 1973-1984: Excavations at Sites SBR-421B, SBR421C, SBR421D, and SBR-713. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. Basgall, M. E., M. C. Hall, and W. R. Hildebrandt 1988 The Late Holocene Archaeology of Drinkwater Basin, Fort Irwin, San Bernardino County, Califonua. Submitted to U. S. Army Corps of Engineers, Los Angeles District Office, California. Bedwell, S. F. 1973 Fort Rock Basin: Prehistory and Environment. University of Oregon Books, Eugene. Bennyhoff, J. A., and R. E. Hughes 1987 Shell Bead and Orament Echange Networks between California and the Western Great Basin. Anthropological Papers of the American Museum of Natural History 64(2). New York. Bettinger, R. L. 1975 The Surface Archaeology of Owens Valley, Eastern California; Prehistoric Man-Land Relationships in the Great Basin. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Riverside. 1977 Aboriginal Human Ecology in Owens Valley, Eastem Califomia: Pmhistoric Change in the Great Basin. American Antiquity 43:3-17. 1982a Aboriginal Exchange and Territoriality in Owens Valley, California. In Contexts for Prehistoric Exchange, edited by J. Ericson and T. Earle, pp. 103-127. Academic Press, New York. 1982b Archaeology of the East Range of Light: Aboiginal Human Ecology of the Inyo Mono-Region of Central-Eastern California. Monographs in California and Great Basin Anthropology, No. 1. University of California, Davis. 1989 The Archaeology of Pinyon House, Two Eagles, and Crater Middens: Three Residential Sites in Owens Valley, Eastern California. Anthropological Papers of the American Museum of Natural History 67. 1991 Aboriginal Occupation at High Altitude: Alpine Villages in the White Mountain of Eastern California. American Anthropologist 93:656-679. Bettinger, R. L., and M. A. Baumhoff 1982 The Numic Spread: Great Basin Cultures in Competition. American Antiquity 47:485-503. Bettinger, R. L., and T. F. King 1971 Interaction and Political Organization: A Theoretical Framework for Archaeology in Owens Valley, California. University of California Archaeological Survey Annual Report 13:187-195. 186 Bettinger, R. L., and R. E. Taylor 1974 Suggested Revisions in Archaeological Sequences of the Great Basin and Interior Southem California. Nevada Archaeological Survey Research Paper 5:1-26. Bettinger, R. L., M. G. Delacorte, and K. R. McGuire 1984 Archaeological Excavations at the Partridge Ranch Site (CA-INY-2146), Inyo County, California. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. Bettinger, R. L., J. F. O'Connell, and D. H. Thomas 1991 Projectile Points as Time Markers in the Great Basin. American Anthropologist 93(1):166-172. Billings, W. D. 1951 Vegetational Zonation in the Great Basin of Western North America. Union Internationale des Sciences Biologiques, Serie B9:101-122. Bostwick, T. W., and J. H. Burton 1993 A Study in Sourcing Basalt Ground Stone Implements. Kiva 3:357-372. Botkin, S. G., T. A. Clewlow, M. C. Brown, and C. W. Clewlow 1987 Final Report on Archaeological Invetigations Along the CLIV 28.5 Mile Transmission Corridor, China Lake Naval Weapons Center, Inyo and Kern Counties, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Bouey, P. D. 1991 Recoging the Limits of Archaeological Applications of Non-Destructive Energy-Dispersive X- Ray Fluorescence Analysis of Obsidians. Materials Research Society Synmposium Proceedings 185:309-320. Bouey, P. D., and M. E. Basgall 1984 Trans-Sierran Exchange in Prehistoric California: The Concept of Economic Articulation. In Obsidian Studies in the Great Basin, edited by R.E. Hughes, pp. 135-172. Contributions of the University of California Archaeological Research Facility No. 45. Berkeley. Bouey, P. D., and P. J. Mikkelsen 1989 Survey and Test Evalwation for the China Lake-Fort Irwin Joint Land Use Area, San Bernardino County. Submitted to the U. S. Army Corp of Engineers, Los Angeles District Office, California. Bouscaren, S., and P. J. Wilke 1987 Excavation at Mammoth: Archaeological Data Recovery at Four Sites near Mammoth Creek, Mono County, California. University of California Archaeological Research Unit Project Report No. UCRARU773, Riverside. Byrd, B. F., D. Pallette, and C. Serr 1994 Prehistoric Settlement along the Eastern Margin of Rogers Dry Lake, Western Mojave Desert, California. Brian F. Mooney Associates Anthropological Technical Series 2. Callahan, E. 1979 The Basics of Biface Knapping in the Eastem Fluted Point Tradition: A Manual for Flinappers and Lithic Analysts. Archaeology of Eastern North America 7(1):1-180. Cleland, J. H. 1988 Problems in the Hydration Dating of Coso Obsidian at the Source. Paper presented at the 22nd Annual Meeting of the Society for California Archaeology, Redding. 187 1989 Induced Hydration Rates for Coso Obsidian: An Update. Paper presented at the 23rd Annual Meeting of the Society for California Archaeology, Santa Rosa. 1990 Sugarloaf Archaeological District Cultural Resources Management Plan. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Cleland, J. H., R. M. Apple, and E. Nilsson 1990 Sugarloaf Mountain in Prehistory: Archaeological Testing and Data Recovery for the Exploratory Drilling Program II and the Unit No. 1 Project. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Clewlow, C. W., R. F. Heizer, and R. Berger 1970 An Assessment of Radiocarbon Dates for the Rose Spring Site (CA-INY-372), Inyo County, California. Anthropology of the Great Basin Contributions of the University of California Archaeological Research Facility No. 7, pp. 19-27. Clewlow, C. W., H. F. Wells, and D. S. Whitley 1980 Cultural Resources Technical Report on the Coso Geothemal Study Area. Rockwell International Technical Report (EMSC8312.20) for the 1980 BLM Coso Leasing EIS. Submitted to Environmental Branch, Naval Air Weapons Station, China Lake, California. Collins, M. B. 1975 Cottrell, M. G. 1981 Lithic Technology as a Means of Processual Inference. In Lithic Technology: Makng and Using Stone Tools, edited by E. Swanson, pp. 15-34. Mouton Publishers, The Hague, Paris. In Obsidian Dates III: A Compendium of the Obsidian Hydration Determinations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and G. S. Russell, pg. 22. University of California, Los Angeles. 1988a Obsidian Hydration: An Analysis of Readings Obtained from CA-ORA-183, A Coastal Shell Midden, Huntington Beach, California. In Obsidian Dates IV:. A Compendium of the Obsidian Hydration Determinations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 30-34. University of California, Los Angeles. 1988b Tomato Springs Site (CA-ORA-244): Obsidian Hydration Readings and Radiocarbon Dates. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Detenminations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 35-39. University of California, Los Angeles. 1988c Analysis of Radiocarbon Dates and Obsidian Hydration Readings Obtained for Two Sites in the Southern California Coastal Region. In Obsidian Dates IV:. A Compendium of the Obsidian hydration Determinations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 4041. University of Califoria, Los Angeles. Crabtree, D. E. 1973 Davis, J. T. 1961 DeDecker, M. 1984 Experiments in Replicating Hohokam Points. Tebiwa 16:1045. Trade Routes and Economic Exchange Among the Indians of California. University of California Archaeological Survey Reports 54:1-71. Berkeley. Flora of the Northern Mojave Desert, California. California Native Plant Society Special Publication 7. 188 Delacorte, M. 1988 G. Beyond the Hoizon: Adaptive Variation in Central Eastern California. Manuscript prepared for University of California, Los Angeles, Institute of Archaeology Monographs. 1990 The Prehistory of Deep Springs Valley, Eastern California: Adaptive Variaion in the Western Great Basin. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Davis. Delacorte, M. 1993 Dillon, B. D. 1988 G., and K. R. McGuire Archaeological Test Evaluaton at 23 Sites Located Along a Proposed Fiber-Optic Telephone Cable Route in Owens Valley, California. Submitted to Contel of California Inc., and the Department of Interior, Bureau of Land Management, California Desert District, Ridgecrest, California. Southern Sierra Nevada Obsidian Hydration: Kern County's Isabella Basin. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Determinations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 64-69. University of California, Los Angeles. Drews, M., and R. G. Elston 1983 An Archaeological Investigation of Drilling Locations and Power Plant Site in the Coso Known Geothemal Resource Area, China Lake Naval Weapons Center. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Duffield, W. A., and C. R. Bacon 1981 Geologic Map of the Coso Volcanic Field and Adjacent Areas, Inyo County, Ca. U.S. Geological Survey Miscellaneous Investigations (series map 1-1200, scale - 1:50,000). Duffield, W. A., C. R. Bacon, and G. B. Dalrymple 1980 Late Cenozoic Volcanism, Geochronology, and Structure of the Coso Range, Inyo Count, California. In Geothermal Investigations in the Coso Range, California. Journal of Geophysical Research 85(5):2381-2404. Elston, R. G., and C. D. Zeier 1984 The Sugarloaf Obsidian Quarry. Naval Weapons Center Administrative Publication 313, China Lake, California. Elston, R. G., S. Seck, and S. James 1981 An Intensive Archaeological Investigation of Two Proposed Drilling Locations in the Coso Known Geothemal Resource Area, China Lake Naval Weapons Center. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Ericson, J. E. 1977 Prehistoric Exchange Systems in California: The Results of Obsidian Dating and Tracing. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Los Angeles. 1978 Date lists, INY-372. In Obsidian Dates II, a Compendium of the Obsidian Hydration Deteminations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and P. I. Vanderhoeven, pp. (3)18-19. Monograph VI, Institute of Archaeology. University of California, Los Angeles 1982 Production for Obsidian Exchange in California. In Contexts for Obsidian Exchange, edited by J. E. Ericson and T. K. Earle, pp. 129-148. Academic Press, New York. 189 1984 Toward the Analysis of Lithic Production Systems. In Prehistoric Quarries and Lithic Production, edited by J. E. Ericson and B. A. Purdy, pp. 1-9. Cambridge University Press, Cambridge. 1989 Toward Flow-Specific Obsidian Hydration Rates: Coso Volcanic Field, Inyo County, California. In Current Directions in California Obsidian Studies, edited by R. E. Hughes, pp. 13-22. Contributions of the University of California Archaeological Research Facility No. 48. Ericson, J. E., H. C. Koerper, C. E. Drover, and P. E. Langenwalter II 1989 Advances in Obsidian Hydration Dating and Obsidian Exchange in Prehistoric Orange County. Pacific Coast Archaeological Society Quarterly, 25(2):45-60. Flenniken, J. J., and A. W. Raymond 1986 Morphological Projectile Point Typology: Replication, Experimentation, and Technological Analysis. American Antiquity 51(3):603-614 Flenniken, J. J., and P. J. Wilke 1989 Typology, Technology, and Chronology of Great Basin Dart Points. American Anthropologist 91(1):149-158. Fowler, C. S. 1986 Subsistence. In Great Basin edited by W. L. d'Azevedo, pp. 64-97. Handbook of North American Indians, Vol. 11, W. C. Sttevant, general editor. Smithsonian Institution, Washington D.C. Fratt, L., and J. L. Adams 1993 New Trends in Ground Stone Research: Its Not the Same Old Grind. Kiva 58(3):313-317. Fredrickson, D. A. 1974 Cultural Diversity in Early Central California: A View from the North Coast Ranges. Journal of California Anthropology 1(1):41-53. Fredrickson, D. A., and J. Grossman 1977 A San Dieguito Component at Buena Vista Lake, California. Journal of California Anthropology 4(2):173-190. Fredrickson, D. A., and G. G. White 1988 The Clear Lake Basin and Early Complexes in Califoia's North Coast Ranges. In Early Human Occupation in Far Western North America: The CZois-Archaic Interface, edited by J. A. Willig, C. M. Aikens, and J. L. Fagan. Nevada State Museum Anthropological Papers 21:75-86. Friedman, I., and W. Long 1976 Hydration Rate of Obsidian. Science 191:347-352. Friedman, I., and J. Obradovich 1981 Obsidian Hydration Dating of Volcanic Events. Quaternary Research 16:37-41 Friedman, I. and R. L. Smith 1960 A New Dating Method Using Obsidian: Part 1, The Development of the Method. Ameican Antiquity 48: 544-547. Frison, G. C., and B. A. Bradley 1980 Folsom Tools and Technology at the Hanson Site, Wyoming. University of New Mexico Press, Albuquerque. 190 Garfinkel, A. P., and K. R. McGuire 1980 Subsistence-Settlement Systems of the Southern Sierra Nevada. In Archaeological Investigations in the Southern Sierra Nevada: The Lamont Meadow and Morris Peak Segment of the Pacific Crest Trail, by A. P. Garfinkel, R. A. Schiffman, and K. R. McGuire. Submitted to the USDA Forest Service, Sequoia National Forest, Porterville, California. 1981 Chronometrics. In Archaeological Investigations in the Kennedy Meadows/Rockhouse Basin Segment of the Pacific Crest Trail, by K. R. McGuire. Submitted to the USDA Forest Service, Sequoia National Forest, Porterville, California. Garfinkel, A. P., C. A. Schiffman, and K. R. McGuire 1980 Archeological Investigations in the Southern Sierra Nevada: The Lamont Meadow and Morris Peak Segments of the Pacific Crest Trail. Bureau of Land Management, Cultural Resources Publications, Archaeology. Bakersfield Garfinkel, A. P., L. S. Kobori, J. C. Bard, and R. J. Dezzani 1984 Rocouse Basin Data Recovery Project. Submitted to the USDA Forest Service, Sequoia National Forest, Porterville, California. Gehr, E. A. 1981 Obsidian Hydration Data from Various Tulare County Sites: Tule River Indian Reservation. In Obsidian Dates IlI A Compendium of the Obsidian Hydration Determinations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and G. S. Russell, page 35. University of California, Los Angeles. 1988 Mean Temperature Scaling of Source Specific Hydration Rates. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Determinations Made at the Obsidian hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 19-26. University of California, Los Angeles. Gilreath, A. J. 1987 Survey and Evaluation of Cultural Resources on a Portion of Federal Lease CA-11402 Lands (Parcel 20) mwithin the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1988 Survey and Evaation of Cultural Resources on a Portion of the Navy Contract Lands (Navy 2) within the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1997 Cultural Resources Sample Survey and Sensitivity Assessment for the Fourmile Hill Geothermal Project, Modoc and Kamah National Forests. Submitted to the U.S.D.A. Forest Service, Modoc National Forest, Alturas, California. Gilreath, A. J., 1988 and W. R. Hildebrandt Survey and Evaluation of Cultural Resources on a Portion of the Navy/C.L.J. V. Contract (Navy 1) Lands Within the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1991 National Register of Historic Places Registration Form: Sugarloaf Archaeological District. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1995 Prehistoric Use of the Coso Volcanic Field, Volumes I-HI. Submitted to the Environmental Branch, China Lake, NAWS, and Eastem Information Center, Department of Anthropology, University of Califomia, Riverside. 191 Gilreath, A. J., and K. R. McGuire 1987 Preliminary Report of Investigations Conducted at Seven Archaeological Sites in Conjunction With the Placement of Eight California Energy Company, Inc. Drill Pads in the Coso K.G.R.A. B.L.M. Lease CA-11402, Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Gilreath, A. J., M. E. Basgall, and M. C. Hall 1987 Compendium of Chronologically Indicative Data From Fort Irwin Archaeological Sites, San Bernardino County, California. Submitted to the U. S. Army Corps of Engineers, Los Angeles District Office, California. Gilreath, A. J., B. P. Wickstrom, and W. R. Hildebrandt 1991 Survey and Evaluation of Cultural Resources on Federal Lease CA-11402 (Residual B.L.M.) Within the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Glennan, W. S. 1971 A Glimpse at the Prehistory of Antelope Valley - Archaeological Investigations at the Sweetser Site (KER-302). Kern-Antelope Historical Society. Lancaster. Goldberg, S. K., E. J. Skinner, and J. F. Burton 1990 Archaeological Excavations at sites CA-MNO-574, -577, -578, and -833: Stoneworking in Mono County, California. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. Goldberg, S .K., S. S. Salzman, E. J. Skinner, J. Burton, M. E. Scully, J. J. Holson, and M. J. Moratto 1986 Cultural Resources of the Crane Valley Hydroelectric Project Area, Volume III: Archaeological Testing, Resource Evaluation, Inpact Assessment, and Management Planning. Submitted to Pacific Gas and Electric Company, San Francisco. Grant, C., J. W. Baird, and J. K. Pringle 1968 Rock Draings of the Coso Range, Inyo County, Caifornia Maturango Museum Publication No 4, Ridgecrest, California. Greenwood, R. S., and M. J. McIntyre 1981 Cultural Resources Management Plan for Edwards Air Force Base. Submitted to Edwards Air Force Base, California. Hall, M. C. 1983 Late Holocene Hunter-Gatherers and Volcanism in the Long Valley-Mono Basin Region: Prehistoric Culture Change in the Eastern Sierra Nevada. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Riverside. 1984 Obsidian, Paleoeconomy, and Volcanism in the Eastern Sierra Nevada. Paper presented at the 18th Great Basin Anthropological Conference, Boise, Idaho 1992 Fnal Report on the Archaeology of 7iefort Basin, Fort Irwin San Bernardino County, California. Submitted to the U. S. Army Corps of Engineers, Los Angeles District Office, California. Hall, M. C., and M. E. Basgall J9! Dib)o CA i And GX ?t Bi rthit Pape p eted at tl 24&( Gr Basin Anthropological Conference, Elko, Nevada. Harrington, M. R. 1951 A Colossal Quarry. Masterkey 25(1):15-18. 192 1957 A Pinto Site at Little Lake, California. Southwest Museum Papers No. 17. Los Angeles. Hartzell, L. L. 1992 Heizer, R. F. 1974 Hunter-gatherer Adaptive Strategies in Lacustrine Environments in the Buena Vista Lake Basin, Kern County, California. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Davis. Studying the Windmiller Culture. In Archaeological Researches in Retrospect, edited by G. A. Willey, pp. 179-204. Winthrop Publishers, Cambridge, Massachusetts. 1978 California. Handbook of North American Indians, Volume 8, W. C. Sturtevant, general editor. Smithsonian Institution, Washington D.C. Heizer, R. F., and M. A. Baumhoff 1962 Prehistoic Rock Art of Nevada and Easem Caifomia. University of California Press, Berkeley and Los Angeles. Heizer, R. F., and T. R. Hester 1978 Great Basin. In Chronologies in New World Archaeology, edited by R. E. Taylor and C. W. Meighan, pp. 147-199. Academic Press, New York. Hester, T. R. 1973 Chronological Ordering of Great Basin Prehistory. Contributions of the University of California Archaeological Research Facility 17. Berkeley. Hildebrandt, W. R., and A. J. Gilreath 1988 Survey and Evaion of Cultural Resources on a Portion f the Navy/C.LJ. V. Contract (Residual Navy 2) Lands within the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Hildebrandt, W. R., P. J. Mikkelsen, and D. A. Jones 1994 Toolstone Procurement and Lithic Production Technology, California. In Archaeological Investigations PG-PG&E Pipeine Eansion Project Idaho, Washington, Oregon, and California; edited by M. J. Moratto, Volume IV:8.1-8.63. Report submitted to Bechtel Corporation, San Francisco. Hughes, R. E. 1982 Age and Exploitation of Obsidian from the Medicine Lae Highland, Califoia. Journal of Archaeological Science 9(2):173-185. 1988 The Coso Volcanic Field Reexamined: Implications for Obsidian Sourcing and Hydration Dating Research. Geoarchaeology 3(4):253-265. 1994a Intrasource Chemical Variability of Artifact-Quality Obsidians from the Casa Diablo Area, California. Journal of Archaeological Science 21(2):263-271. 1994b Mosaic Patterns in Prehistoric California - Great Basin Exchange. In Prehistoric Exchange Systems in North America, edited by T. Baugh and J. E. Ericson, pp. 363-383. PlenumPress, New York. Hughes, R. E., 1985 and D. L. True Perspectives on the Distribution of Obsidians in San Diego County, California. North American Archaeologist 6(4):325-339. 193 Hunt, A. P. 1960 Archaeology of the Death Valley Salt Pan, Califorma. University of Utah Anthropological Papers 47. Hunt, C. B. 1975 Death Valley: Geology, Ecology, Archaeology. University of California Press, Berkeley. Iroquois Research Institute 1979 A Land Use History of Coso Hot Springs, Inyo County, Califomia. Administrative Publication 200, Naval Weapons Center, China Lake. Irwin, C. N. (editor) 1980 The Shoshoni Indians of Inyo County, California: The Kerr Manuscript. Edited, annotated and with introductoty preface by C. N. Irwin. Ballena Press Publications in Archaeology, Ethnology, and History No. 15, edited by P. J. Wilke and A. B. Elsasser. Jack, R. N. 1976 Jackson, R. J. 1985 Jackson, T. L. 1988 Prehistoric Obsidian in Califomia I: Geochemical Aspects. In Advances in Obsidian Glass Studies: Archaeological ad Geochemical Perspectives, edited by R. E. Taylor, pp. 183-217. Noyle Press, Park Ridge, New Jersey. An Archaeological Survey of the Wet, Antelope, Railroad, and Ford Timber Sale Compartments in the Inyo National Forest. Submitted to the USDA Forest Service, Inyo National Forest, Bishop, California. Amending Models of Trans-Sierran Obsidian Tool Production and Exchange. Journal of California and Great Basin Anthropology 10(1):62-72. Jackson, T. L., and S. A. Dietz 1984 Archaeological Data Recovery Excavations at CA-FRE-798 and CA-FRE-805, Siphon Substation 33kV Distribution Line and Balsam Meadow Hydroelectric Project. Submitted to Southern California Edison Company, Rosemead, California. Jackson, T. L., and J. E. Ericson 1994 Prehistoric Exchange Systems in California. In Prehistoric Exchange Systems in North America, edited by T. Baugh and J. E. Ericson, pp. 385415. Plenum Press, New York. Jenkins, M. 1975 Glass Trade Beads in Alaska. Bead Journal 2(1):23-26. Jennings, J. D. 1975 Prehistory of North America. Second Edition. McGraw-Hill Book Company, New York. Jones, D. A. 1991 Flaked Stone Artifacts from Nelson Basin. In The Archaeology of Nelson Lake Basin and Adjacent Areas, by M. E. Basgall. Submitted to the U. S. Army Corps of Engineers, Los Angeles District Office, California. 1992 Flaked Stone. In Final Report on the Archaeology of Tiefort Basin, Fort Irwin, San Bernardino County, California, by M. C. Hall. Submitted to the U. S. Army Corps of Engineers, Los Angeles District Office, California. 194 Kelly, M. S., J. H. Cleland, and A. L. York 1989 Cultural Resources Investigation for the Shallow Underground Tunnel/Chamber Explosion Test (S. U.T.C.E.T.) Facility. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Kelly, R. L. 1988 The Three Sides of a Biface. American Antiquity 53:717-734. King, T. F. 1976 Political Differentiation Among Hunter-Gatherers: An Archaeological Test. Unpublished Ph.D. Dissertation, Department of Anthropology, University of California, Riverside. Knaus, E., and J. J. Mazer 1991 Homogeneity of Water Content in Obsidianfrom the Coso Field, Submitted to the U.S. Department of Energy, Washington, D.C. Inyo County, California. Koerper, H. C., J. E. Ericson, C. E. Drover, and P. E. Langenwalter II 1986 Obsidian Exchange in Prehistoric Orange County. Pacific Coast Archaeological Society Quarterly 22(1):33-69. Kuchler, A. W. 1977 Map of the Natural Vegetation of Califoia. n Terrestrial Vegetation of California, edited by M. G. Barbour and J. Major. Wiley and Sons, New York. Lamphere, M. A., G. B. Dalrymple, and R. L. Smith 1975 K-Ar Ages of Pleistocene Rhyolitic Volcanism in the Coso Motains, Califonia. Geology, 3:339- 341. Lanning, E. P. 1963 The Archaeology of the Rose Spring Site (INY-372). University of California Publications in American Archaeology and Ethnology 49(3):237-336. Berkeley. Laylander, D., and L. E. Christenson 1988 Corral Canyon and Late Prehistoric Exchange in Inland San Diego County, California. Proceedings of the Society for California Archaeology 1:134-158. Lee, R. 1969 Chemical Temperature Integration. Journal of Applied Meteorology 8:423430. Leitner, B. M., and P. Leitner 1988 Biological Resources of Certain Lands Within the Coso K.G.R.A., Including Portions of Navy/C.L.J. V. Contract Lands and B.L.M. Lease CA-11402. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Mazer, J. J., J. K. Bates, C. M. Stevenson, and C. R. Bradley 1992 Obsidians and Tektites: Natural Analogues for Water Diffusion in Nuclear Waste Glasses. Materials Research Society Symposium Proceedings 257:513-520. McClean, D. D. 1944 The Prong-horned Antelope in California. California Fish and Game 30:221-241. McDonald, A. M. 1992 Indian Hill Rockshelter and Aboriginal Cultural Adaptation in Anza-Borrego Desert State Park, Southeastern California. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Riverside. 195 McGuire, K. R. 1981 Archaeological Investigations in the Kennedy Meadows/Rockhouse Basin Segment of the Pacific Crest Trail. Submitted to the USDA Forest Service, Sequoia National Forest, Porterville, California. McGuire, K. R., and A. P. Garfinkel 1980 Archaeological Investigations in the Southern Sierra Nevada: The Bear Mountain Segment of the Pacific Crest Trail. Bureau of Land Management, Cultural Resources Publications, Archaeology. Bakersfield. McGuire, K. R., A. P. Garfinkel, and M. E. Basgall 1982 Archaeological Investigations in the El Paso Mountains of the Western Mojave Desert: The Bickel and Last Chance Sites, CA-KER-250, -261. Submitted to the Department of the Interior, Bureau of Land Management, California Desert District, Ridgecrest, California. MacLeod, N. S., D. R. Sherrod, and L. A. Chitwood 1982 Geologic Map of Newberry Volcano, Deschutes, Klamath, and Lake Counties, Oregon. U. S. Geological Survey Open File Reports 82-847. Meighan, C. W. 1978 Obsidian Dating of the Malibu Site. In Obsidian Dates II: A Compendium of the Obsidian Hydration Determinations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and P. I. Vanderhoeven, pp. 158-161. University of Califomia Institute of Archaeology Monograph 6. Los Angeles. 1981 The Little Lake Site, Pinto Points, and Obsidian Hydration Dating in the Great Basin. Journal of California and Great Basin Anthropology 3:200-214. 1988 Obsidian Dates from the Loyola Site, LAN-61. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Determinations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 27. University of California, Los Angeles. Meighan, C. W., and G. S. Russell (editors) 1981 Obsidian Dates III: A Compendium of the Obsidian Hydration Determinations Made at the U.C.L.A. Obsidian Hydration Laboratory. University of California, Los Angeles. Meighan, C. W., and J. L. Scalise (editors) 1988 Obsidian Dates IV: A Compendium of the Obsidian hydration Deteminations Made at the U.C.L.A. Obsidian Hydration Laboratory. University of California, Los Angeles. Meighan, C. W., and P. I. Vanderhoeven (editors) 1978 Obsidian Dates II: A Compendium of the Obsidian Hydration Determinations Made at the U.C.L.A. Obsidian Hydration Laboratory. University of California, Los Angeles. Michels, J. W. 1965 Lithic Serial Chronology through Obsidian Hydration Dating. Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Los Angeles. 19%3 The Hy&atiou Rate for Coso Sugaloat) Obsidian at Mbaedoogvcal Sites in the China 1ke Tea of California. MOHLAB Technical Report 23, State College, Pennsylvania. Mille, P. 1975 An Historical Explanation of Alaskan Trade Beads. Bead Journal 2(2):20-24. 196 Mitchell, D. R., and M. S. Shackley 1995 Classic Period Hohokam Obsidian Studies in Southern Arizona. Journal of Field Archaeology 22(3):291-304. Mone, S. L. 1986 Moratto, M. J. 1972 CA-MNO-574, A Secondary Reduction Site in Long Valley, Mono County, California. Submitted to the California Department of Transportation, Office of Environmental Analysis, Sacramento. A Study of Prehistory in the Southern Sierra Nevada Foothills, California. Unpublished Ph.D. dissertation, Department of Anthropology, University of Oregon, Eugene. Moratto, M. J., T. F. King, and W. B. Woolfenden 1978 Archaeology and California's Climate. Th7e Journal of California Anthropology 5(2):147-161. Moratto, M. J., and O. K. Davis 1988 Muto, G. R. 1971 Paleoenvironment. In Culture Change in the Central Sierra Nevada 8000 B. C. - A.D. 1950, edited by M. J. Moratto, L. H. Shoup, and J. D. Tordoff, pp. 39-88. Submitted to the National Park Service, Washington D.C. A Technological Analysis of the Early Stages in the Manufacture of Lithic Artifacts. Unpublished Master's thesis, Department of Anthropology, Idaho State University, Pocatello. Newcomer, M. H. 1971 Some Quantitative Experiments in Hand-axe Manufacture. World Archaeology 3:85-94. Rhode, D. 1992 The World According to a Hydrating Obsidian Flake: Steps Towards Chronological Control of an Archaeological Landscape. Paper presented at the 23rd Great Basin Anthropological Conference, Boise. Roper-Wickstrom, K. C. 1992 A Study of High Altitude Obsidian Distribution in the Southern Sierra Nevada, California. Unpublished Master's thesis, California State University, Sonoma, CA. Rosen, M. D. 1988 Ross, L. A. 1993 Samples Submitted from Ven-294. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Determnations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, p. 331. University of California, Los Angeles. Glass Beads from Native American Protohistoric Sites in Inyo County, California. In Archaeological Test Evaluion at 23 Sites Located Along a Proposed Fiber-Optic Telephone Cable Route in Owens Valley, California, by M. G. Delacorte and K. R. McGuire, Appendix V. Submitted to Contel of California Inc., and the Department of Interior, Bureau of Land Management, California Desert District, Ridgecrest, California. Schiffman, R. A. 1988 An Experiment in the Use of Obsidian Hydration: Effect of oratory Variability on the Research and Interpretation of Sites. In Obsidian Dates IV: A Conmpendium of the Obsidian Hydration Determnations Made at the Obsidian Hydration Laiboratory, edited by C. W. Meighan and J. L. Scalise, pp. 13-18. University of California, Los Angeles. 197 Schneider, J. S., and G. D. Everson 1989 The Desert Tortoise (Xerobates agassizii) in the Prehistory of the Southwestern Great Basin and Adjacent Areas. Journal of California and Great Basin Anthropology 11:175-202. Shackley, M. S. 1995 Sources of Archaeological Obsidian in the Greater American Southwest: An Update and Quantitative Analysis. American Antiquity 60(3):531-551. Shackley, M. S., J. R. Hyland, and M. de la Luz Guti6rrez 1996 Mass Production and Procurement at Valle Del Azufre: A Unique Archaeological Obsidian Source in Baja California Sur. American Antiquity 61(4):718-731. Sharrock, F. W. 1966 Prehistoric Occupation Patterns in Southwestern Wyoming and Cultural Relationships with the Great Basin and Plains Culture Area. University of Utah Anthropological Papers 77. Singer, C. 1981 Date List, Horse Flats Sites, LAN474A, 474B, Los Angeles County. In Obsidian Dates III: A Compendium of the Obsidian Hydration Detenninations Made at the U.C.L.A. Obsidian Hydration Laboratory, edited by C. W. Meighan and G. S. Russell, p. 21. University of California, Los Angeles. Singer, C. A., 1977 Skinner, C. E. 1995a and J. E. Ericson Quarry Analysis at Bodie Hills, Mono County, California: A Case Study. In Exchange Systems in Prehistory, edited by T. K. Earle and J. E. Ericson, pp. 171-190. Academic Press, New York. Obsidian Hydration Studies. In Archaeological Investigations PGT-PG&E Pipeline Expansion Project Idaho, Washington, Oregoand Califoia; edited by M. J. Moratto, Volume V:5.1-5.51. Submitted to Bechtel Corporation, San Francisco. 1995b Obsidian Characterization Studies. In Archaeological Investigations PGT-PG&E Pipeline Expansion Project Idaho, Washington, Oregon, and California; edited by M. J. Moratto, Volume V:4.1-4.54. Submitted to Bechtel Corporation, San Francisco. Skinner, E. J. 1989 Lithic Analysis. In Cultural Resources of the Crane Valley Hydroelectric Project Area, Madera County, California. Volume V, Bass Lake Erosion Control Project, Limited Archaeological Site Data Recovery at CA-MAD-223, -224, and -392, by S. K. Goldberg and E. J. Skinner. Submitted to Pacific Gas and Electric Co., San Francisco. Skinner, E. J., and P. W. Ainsworth 1990 Unifacial Bifaces: More Than One Way to Thin a Biface. Paper presented at the 43rd Northwest Anthropological Conference, Eugene, Oregon. Stevenson, C. M. 1987 Hydration Rate Development for Selected Obsidians from the Coso Volcanic Field, Inyo County, California. Submitted to Dames and Moore, San Diego Office, prepared by Archaeological and Historical Consultants, Centre Hall, Pennsylvania. 1990 Vapor Hydration Rates for the Coso Obsidian Source, Inyo County, California. In Archaeological Testing and Data Recovery for the Exploratory Drilling Program II and the Unit No. 1 Project, by J. H. Cleland, R. M. Apple, and E. Nilsson, Appendix E. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 198 Stevenson, C. M., and B. E. Scheetz 1989 Induced Hydration Rate Development of Obsidian from the Coso Volcanic Field: A Compendium of Expeinental Procedures. In Current Directions in California Obsidian Studies, edited by R. E. Hughes, pp. 23-30. Contributions of the University of California Archaeological Research Facility No. 48. Stevenson, C. M., J. J. Mazer, and J. K. Bates 1990 Obsidian Hydration Dating as a Function of Temperature, Compositions, and Relative Huimidity. Paper presented at the 24th Annual Meeting of the Society for California Archaeology, Foster City. Stevenson, C. M., E. Knaus, J. J. Mazer, and J. K. Bates 1993 Homogeneity of Water Content in Obsidian from the Coso Volcanic Field: Implications for Obsidian Hydration Dating. Geoarchaeology 8(5):371-384. Steward, J. H. 1933 Ethnography of the Owens Valley Paiute. University of California Publications in American Archaeology and Ethnology 33(3):233-350. 1938 Basin-Plateau Aboriginal Sociopolitical Groups. Bulletin 120, Smithsonian Institution, Bureau of American Ethnology. 1954 Types of Types. American Anthropologist. 56(1):54-57. Stine, S. 1994 Extreme and Persistent Drought in Califomia and Patagonia During Mediaeval Time. Nature 369:546-549. Sutton, M. Q. 1988a An Introduction to the Archaeology of the Western Mojave Desert, California. Archives of California Prehistory No. 14. 1988b Obsidian Analysis in the Mojave Desert, California: Results, Cautions, and Comments. In Obsidian Dates IV: A Compendium of the Obsidian Hydration Determinations Made at the Obsidian Hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 51-63. University of California, Los Angeles. 1988c Insects as Food: Aboriginal Entomophagy in the Great Basin. Ballena Press Anthropological Papers No. 33. 1991 Archaeological Investigations at Cail, Fremont Valley, Western Desert, California. Museum of Anthropology, Occasional Papers in Anthopology No. 1. California State University, Bakersfield. Theodoratus Cultural Research, Inc. and Archaeological Consulting and Research Services, Inc. (TCR and ACRS) 1984 Cultural Resources Overview of the Southern Sierra Nevada. Submitted to the USDA Forest Service, Inyo National Forest, Bishop, California. Thomas, D. H. 1981 How to Classify the Projectile Points from Monitor Valley, Nevada. Journal of California and Great Basin Anthropology 3:7-43. 1986 Points on Points: A Reply to Flenniken and Raymond. American Antiquity 51(3):619-627. Thornthwaite, C. W. 1948 An Approach Toward a Rational Classification of Climate. Geographical Review 38:55-94. 199 Torrence, R. 1986 Productian and Exchange of Stone Tools. Cambridge University Press, Cambridge. Tremaine, K. J. 1989 Obsidian as a Time Keeper: An Investigation in Absolute and Relative Dating. Unpublished Master's thesis, Department of Anthropology, Sonoma State University, California. Tremaine, K. J., and D. A. Fredrickson 1988 Induced Obsidian Hydration Experiments: An Investigation in Relative Datng. Matenal Research Society Symposium Procedures 123:271-278. Troxell, H. C., and W. Hofmann 1954 HydrologyoftheMojaveDesert. InGeology of SouthernCalifornia, editedbyR.H.Jahns,pp. 13-17. California Division of Mines Bulletin 170. W and S Consultants 1985 Archaeological Survey of Eight Proposed Geothermal Well Pads within Federal Lease CA-11402, Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1986a Archaeological Survey of the Proposed C.L.J. V. 28.5 Mile Transmission Line Coridor in the Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. 1986b Archaeological Investigations of Well Pads C and D Within Federal Lease CA-11402, Coso K.G.R.A., Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Wallace, W. J. 1958 Archaeological Investigations in Death Valley National Monument. University of California Archaeological Survey Reports 42:7-22. 1977 A Half-century of Death Valley Archaeology. he Jounal of California Anthropology 4:249-258. Warren, C. N. 1967 The San Dieguito Complex: A Review and Hypothesis. American Antiquity 32:168-185. 1984 The Desert Region. In Califoia Archaeology, by M. J. Moratto, pp. 339430. Academic Press, Orlando, Florida. 1986 Early Holocene Cultural Adaptation in the Mojave Desert, California. Paper presented at the World Archaeological Congress, Southampton, England. Warren, C. N., 1986 and R. H. Crabtree Prehistory of the Southwestem Area. In Great Basin, edited by W. L. d'Azevedo, pp. 183-193. Handbook of the North American Indians, Vol. 11, W. C. Sturtevant, general editor. Smithsonian Institution, Washington, D. C. Warren, C. N., M. M. Lyneis, and J. H. Cleland 1984 Draft Historic Preservation Plan, Fort Irwin, California. Submitted to the Interagency Archaeological Services, National Park Service, San Francisco. Wedel, W. R. 1941 Archaeological Investigations at Buena Vista Lake, Kern County, California. Washington, D.C.: Bureau of American Ethnology Bulletin 130. 200 Whitley, D. S. 1988 Whitley, D. S., 1987 Obsidian Hydration Dates from the Coso Range. In Obsidian Dates IV: A Compendium of the Obsidian hydration Determinations Made at the Obsidian hydration Laboratory, edited by C. W. Meighan and J. L. Scalise, pp. 75-76. University of California, Los Angeles. G. Gumerman IV, J. M. Simon, and E. H. Rose The Late Prehistoric Period in the Coso Range and Environs. Paper presented at the 21st Society for California Archaeology Meetings, Fresno. Wilke, P. J., and J. J. Flenniken 1988 Experimental Reduction of a Bifacial Core. Paper presented at the 21st Great Basin Anthropological Conference, Boise. 1991 Missing the Point: Rebuttal to Bettinger, O'Connell, and Thomas. American Anthropologist 93(1):172-173. Yohe, R. 1992 A Reevaluation of Western Great Basin Cultural Chronology and Evidence for the Timing of the Introduction of the Bow and Arrow to Eastern California Based on New Excavations at the Rose Spring Site (CA-INY-372). Unpublished Ph.D. dissertation, Department of Anthropology, University of California, Riverside. York, A. 1992 Archaeological Investigations at CA-KER-2816 and -2817 Edwards Air Force Base, California. Paper presented at the Society for California Archaeology Annual Meetings. Zeier, C. D., 1984 and R. G. Elston An Analysis of Obsidian Hydration Processes at the Sugarloaf Obsidian Quarry, Inyo County, California. Submitted to the Environmental Branch, Naval Air Weapons Station, China Lake, California. Zigmond, M. L. 1981 Kawaiisu Ethnobotany. University of Utah Press, Salt Lake City. 201 Amy J. Gilreath and William R. Hildebrandt are Principals at Far Western Anthropological Research Group, Inc. in Davis, California. CONTRIBUTIONS OF THE UNIVERSITY OF CALIFORNIA ARCHAEOLOGICAL RESEARCH FACILITY NUMBER 56 AUGUST 1997 PREHISTORIC USE OF THE COSO VOLCANIC FIELD by AMY J. GILREATH and WILLIAM R. HILDEBRANDT UNIVERSITY OF CALIFORNIA AT BERKELEY SAN ANGELES 0 00 ML ETES ?- - - - - ' - 00 KlLOUETER9 ISBN 1-882744-09-8 PREHISTORIC USE OF THE COSO VOLCANIC FIELD by AMY J. GILREATH and WILLIAM R. HILDEBRANDT 1997 NUMBER 56 CONTRIBUTIONS OF THE UNIVERSITY OF CALIFORNIA ARCHAEOLOGICAL RESEARCH FACILITY BERKELEY Library of Congress Catalog Number 97-61421 ISBN 1-882744-09-8 ? 1997 by the Regents of the University of California Archaeological Research Facility University of California at Berkeley Printed in the United States of America. All rights reserved. No part of this publication may be reproduced or trnmintted in y form or by any means, electronic or mechanical, including photocopy, recording, or any inforion ra or retrieval estamn without the written permission of the publisher. PREFACE CalEnergy, Inc., today operates four geothermal power plants and maintains over 70 supporting wells in the Coso Known Geothermal Resources Area (KGRA) of China Lake Naval Air Weapons Station in southern California. Between 1986 and 1993, the authors directed numerous archaeological studies within the Coso KGRA, to support CalEnergy's efforts to comply with federal and state historic preservation law and policy. Many technical reports of these studies are filed with regulatory agencies. These reports typically are extremely descriptive, very dry, and filled with compliance jargon. Their distribution is, unsurprisingly and mercifully, restricted. Our e ig motivation for writing this monograph was to share the noteworthy results of "compliance archaeology" studies in this fascinating part of the western Great Basin with you, the interested public. Many preceding studies at Coso helped lay the groundwork for this report, and a brief history of the studies is in order here, identifying the most substantive ones. Cultural studies related to geothermal development at Coso began in the late 1970s with an ethnographic and historic overview of Coso Hot Springs (Iroquois Research Institute 1979). Soon thereafter, Clewlow et al. (1980) conducted a sample survey of the KGRA, providing data necessary to develop a predictive model of site density across variable environmental settings. Numerous small-scale studies were then accomplished as prospective developers began exploratory drilling in the early 1980s. Most of this early work was by Intermountain Research (e.g., Drews and Elston 1983; Elston and Zeier 1984; Elston et al. 1981; Zeier and Elston 1984), and focused on obsidian acquisition patterns across time and space. In the late 1980s and early 1990s, exploration accelerated. The roster of archaeologists working in the KGRA expanded, and is reflected by the number of reports filed, such as Botkin et al. 1987, Cleland et al. 1990, Gilreath and McGuire 1987, Kelly et al. 1989, and W&S Consultants 1985, 1986a, 1986b. Once exploration determined that full-scale geothermal development was viable, small-scale studies were replaced by extensive surveys blanketing over 15,000 acres (Gilreath 1987, 1988; Gilreath and Hildebrandt 1988; Gilreath et al. 1991; Hildebrandt and Gilreath 1988). CalEnergy's preluinary development plans were then overlaid on archaeology base maps. Where feasible, cosuction and project designs were redrawn to avoid sites andminimize disturbance to others. In the end, all or portions of 34 prehistoric sites required excavation. This report describes the results of those excavations. Under ideal circumstances, sites might have been chosen for investigation based on a strategy designed to obtain a presentaive sample of the full range of site types present. Although geothermal development plans in large part derined the sites investigated, we were fortuitous in obtaining robust samples of both quarry and off-quarry sites. The reader interested in individual site reports and/or analytical raw data is directed to a three-volume report by Gilreath and Hildebrandt (1995), filed with the Eastern Inforaion Center of the California Historic Resources File System, of the California State Office of Historic Preservation. Volume I is similar in many ways to the report in hand, but necessarily describes the project within the cultural resources management regulatory framework. Individual site reports are pr ed in Volume II. Volume m includes analytical raw-data appendices for projectile points, other flaked stone artifacts (e.g., bifaces, cores, flake tools, etc.), ground and battered implements (e.g., handstones, millingsnes, cobble tools, etc.), and faunal remains. Excavation unit summaries are also provided, and obsidian sourcing (X-ray fluorescence) and obsidian hydration analyses conclude the appendices. i ACKNOWLEDGMENTS Completing an archaeological project of this magnitude is quite a challenge. Many people have contributed to the effort, beginning with project planning, continuing through multiple seasons of field work, and concluding with years of analysis and write-up. Carolyn Shepherd (China Lake Naval Air Weapons Station) was a primary player throughout the process. She was instrumental in developing this study and, through determined interactions with numerous parties, made sure we were able to finish the work. Her enduring patience was always welcomed. CalEnergy, who financed the undertaking, also deserves a great deal of credit. Dave McClain and Phil Essner are commended for their efforts to incorporate archaeological concerns into the development process. Greg Halsey, with the help of Jony Homer and Becky Baker, assisted in all aspects of the project, ranging from field logistics to budgetary concerns. Laurie McClenahan (MHA Environmental Consulting, Inc.) and Jean Hopkins (formerly of MHA) developed the environmental documents necessary for our work to go forward; we appreciate their efforts. We also thank Rob Jackson (then of the State Office of Historic Preservation) and Joan Oxendine (Bureau of Land Management) for their assistance during early phases of the project. Many people participated in the field effort, working hard under a wide range of climatic conditions. We thank Laura Barrett, John Berg, Gary Buck, Kimberly Collins-Roscoe, Jery Doty, Willie Duddles, Rick Fitzgerald, Mark Giambastiani, David Glover, Leslie Glover, Glenn Gmoser, Larry Hause, Jeff Hall, Jennifer Hider, Liz Honeysett, Cristi Hunter, Mark Hylkema, Marilyn Jasmain, Robert Jobson, Kirk Johnson, Bob Johnson, Debbie Jones, Ed Kaler, Eric Kaufnn, the late Ron King, Jan Lawson, the late James N. Leavitt, Susan McCabe, Kelly McGuire, Pat Mikkelsen, Rochelle Mink, Peter Mundwiller, Jim Nelson, Daryl Noble, Bill Norton, Traci O'Brien, Matt O'Connell, Jill Onken, Rex Palmer, Cindy Park, Bert Rader, Allise Rhode, D. Rogers, James Roscoe, Kristina Roper, Gary Sauner, Barry Scott, Jill Shannon, Bill Slater, Linda Sickler-Hylkema, Bill Stillman, Chris Sublett, Miranda Warburton, Laurie Walsh, Brian Wickstrom, Glen Wilson, and Eric Wohlgemuth. We are grateful to all involved with analysis and report p tion, including Marguerite Badovinac, Kevin Bell, John Berg, David Boe, Chris Carter, Bill Combs, Jerry Doty, Deborah Dyer, Jelmer Eerkens, Christine Franco, David Glover, Leslie Glover, Carl Hansen, Susan Harris, Jennifer Hider, Liz Honeysett, Cristi Hunter, Kandi Gillespie, Martha Jackson, Eric Kaufmann, Joe Mayeux, Bill Norton, Jessie Piper, Jennifer Richman, Barry Scott, Paul Strader, Krislyn Taite, Nnika Tinney, Sharon Waechter, Patricia Welsh, and Derek Wheeler. We especially appreciate the efforts of Eric Wohlgemuth throughout this project, leading excavation crews, directing laboratory studies, and contributing his ethnobotanical expertise. Finally, we thank Tammara Ekness and Peter Mundwiller for their fine graphics work, and Michael Delacorte for sharing information from ongoing projects. The authors particularly appreciate Robert Bettinger and William Clewlow, Jr., for their efforts reviewing, editing, and discussing an early draft version. We thank two anonymous peers for the time and attention they paid our revised draft. Comments hey provided made us work harder at ing certain aspects of this monograph, and we are much obliged for their efforts. ii ABSTRACT Excavations at 34 sites in the Coso Volcanic Field reveal a dynamic history of obsidian use spanning at least 10,000 years. Obsidian is common and abundant throughout the area, both in primary and secondary contexts. Secondary lag deposits are widely distributed due to numerous pyroclastic eruptions that sent obsidian nodules and other debris over much of the local laInscape. Primary flows, though less numerous and of more difficult access, have greater quantities of material, often forming thick bands of glass extruded from the sides of rhyolite domes. Dated primarily on the basis of obsidian hydration, analysisstl of over 90 chronologically discrete assemblages indicates that prior to 2800 B.P., use of the area was largely restricted to generalized, short-term exploitation of lag quarries by groups using an expansive, highly mobile subsistence-settlement system. Beginning around 2800 B.P., but reaching peak proportions between 2300-1275 B.P., a major reorganization in production strategy took place. Lag quarries were essentially ignored, and primary flows were intensively exploited, supplying obsidian to a co ding proliferation of off-quarry biface manufacturing areas. Increases in production volume were at least partially related to long distance exchange. Contemporaneous parallel increases in Coso obsidian use among consumer populations in the Owens-Rose Valley area, southern Sierra Nevada and, more importantly, several locations in coastal southern. California provide evidence of this exchange. Contrary to expectations of some, development of obsidian surpluses for exchange did not correlate with the emergence of sedentary populations and local control of the source. Rather, subsistence-settlement pattern data from surrounding areas indicate that residential mobility remained relatively high, but followed a more regularized schedule, allowing more predictable interactions among neighboring populations. Local control of the Volcanic Field appears to have occurred for only a short period between ca. 1275-800 B.P., co-occurring with a reduction in settlement mobility region-wide. During this short period, flaked stone production became largely restricted to few primary deposits, even though obsidian consumption continued unimpeded in outlying areas. After 800 B.P., continued decreases in settlement mobility, often accompanied by higher degrees of territoriality, did not lead to an expanded inter-regional exchange system. Instead, production and exchange of Coso obsidian items essentially stopped, and the Volcanic Field was used for subsistence, largely the exploitation in unprecedented proportions of small seeds. Collapse of the Coso production-exchange system is attributed to decreased demand caused by a variety of factors including technological change and a shifting subsistence focus, while increases in territoriality may have inhibited its supply, not only by restricting direct access, but also by increasing the number of exchanges required to move material equivalent distances across the land. iii TABLE OF CONTENTS PREFACE ....................................................... ACKNOWLEDGMENTS ........................................... ABSTRACT ..................................................... LIST OF MAPS ............................................ LIST OF FIGURES .......................................... LIST OF PLATES ........................................... LIST OF TABLES ........................................... ................ ii ............... ii ............... 1u ................ vii ................ vii ............ viii .. .......... viii CHAPTER 1: INTRODUCTION ........................................ CHAPTER 2: RESEARCH ISSUES ..................................... ENVIRONMENTAL PRODUCTIVITY ................................. CHRONOLOGY ............................................... RADIOCARBON DATES FROM THE VOLCANIC FIELD ................. CROSS-DATING OF DIAGNOSTIC ARTIFACTS FROM THE VOLCANIC FIELD OBSIDIAN HYDRATION DATING IN THE COSO VOLCANIC FIELD ......... LITHIC PRODUCTION AND EXCHANGE ............................. WHAT QUARRY DEPOSITS WERE EXPLOITED ...................... WHEN DID EXPLOITATION OCCUR .............................. WHAT WAS BEING PRODUCED AND IN WHAT QUANTITIES ............ WHY WAS COSO OBSIDIAN EXPLOITED ........................... WHO HAD ACCESS TO COSO OBSIDIAN............................ SUBSISTENCE AND SETTLEMENT .................................. NORTHERN MOJAVE DESERT .................................. WESTERN MOJAVE DESERT .................................... OWENS VALLEY .......................................... SOUTHERN SIERRA NEVADA AND WHITE MOUNTAINS ............... DEATH AND DEEP SPRINGS VALLEYS ............................ COSO REGION .............................................. DISCUSSION .............................................. CHAPTER 3: FIELD AND ANALYTICAL METHODS . . FIELD METHODS ....................... ANALYTICAL METHODS .................. FLAKED STONE ANALYSIS .............. GROUND AND BATTERED STONE ANALYSIS Millingstones ..................... H adstones........................ Miscellaneous Ground Stone ............ Other Milling Equipment .............. Non-Obsidian Cobble-Cores ............ MISCELLANEOUS ARTIFACTS ........... FLOTATION ANALYSIS (by E. Wohlgemuth) ... FAUNAL ANALYSIS ................... CHAPTER 4: ENVIRONMENTAL PRODUCTIVITY .... PLANT RESOURCES ..................... PLANT RESOURCE SUMMARY ........... ANIMAL RESOURCES .................... DISCUSSION ............................ v ee.......e 1 7 7 9 9 9 10 16 16 16 18 21 21 22 22 24 25 26 26 28 28 .31 .31 .33 .34 .38 .41 .41 .41 .42 .42 .42 .50 * 50 . 51 .54 57 .58 .58 CONTENTS - Continued CHAPTER 5: CHRONOMETRICS ............................. RADIOCARBON ASSAYS ................................ PROJECTILE POINTS................................... OTHER TEMPORALLY DIAGNOSTICITEMS ................... SUMMARY ........................................... CHAPTER 6: LITHIC PRODUCTION PATTERNS ................... EARLY PERIOD QUARRY ASSEMBLAGES .................. EARLY PERIOD OFF-QUARRY ASSEMBLAGES .............. EARLY PERIOD DISCUSSION .......................... LITTLE LAKE PERIOD QUARRY ASSEMBLAGES ............. LITTLE LAKE PERIOD OFF-QUARRY ASSEMBLAGES .......... LITTLE LAKE PERIOD DISCUSSION ...................... EARLY NEWBERRY PERIOD QUARRY ASSEMBLAGES......... EARLY NEWBERRY PERIOD OFF-QUARRY ASSEMBLAGES ...... EARLY NEWBERRY PERIOD DISCUSSION .................. MIDDLE NEWBERRY PERIOD OFF-QUARRY ASSEMBLAGES. MIDDLE NEWBERRY PERIOD DISCUSSION ................. LATE NEWBERRY PERIOD PRIMARY QUARRY ASSEMBLAGES ... LATE NEWBERRY PERIOD OFF-QUARRY ASSEMBLAGES ....... LATE NEWBERRY PERIOD DISCUSSION ................... HAIWEE PERIOD PRIMARY QUARRY ASSEMBLAGES .......... HAIWEE PERIOD LAG QUARRY ASSEMBLAGES ............. HAIWEE PERIOD OFF-QUARRY ASSEMBLAGES .............. HAIWEE PERIOD DISCUSSION........................... MARANA PERIOD LAG QUARRY ASSEMBLAGES ............. MARANA PERIOD OFF-QUARRY ASSEMBLAGES ............. MARANA PERIOD DISCUSSION ......................... DISCUSSION......................................... TOOLSTONE WORKING PRACTICES AT COSO QUARRIES ....... TOOLSTONE WORKING PRACTICES IN OFF-QUARRY CONTEXTS CHAPTER 7: SUBSISTENCE-SETTLEMENT PATTERNS .. ARTIFACT AND FEATURE INVENTORIES ........ EARLY PERIOD ........................ LITTLE LAKE PERIOD ................... EARLY NEWBERRY PERIOD ............... MIDDLE NEWBERRY PERIOD .............. LATE NEWBERRY PERIOD ................ HA1WEE PERIOD ....................... MARANA PERIOD ...................... SUBSISTENCE-SETTLEMENT SUMMARY ......... DIACHRONIC TRENDS.................... SPATIAL DISTRIBUTIONS.................. GROUND AND BATE RED STONE ASSEMBLAGES Millingstones ....................... Handstones .......................... Non-Obsidian Cobble-Cores .............. FLORAL REMAINS ...................... FAUNAL REMAINS ..................... DISCUSSION.............................. vi 61 68 70 87 88 . 89 . 89 .94 . 95 . 97 100 102 102 107 108 109 110 110 112 118 118 121 123 124 124 124 128 128 129 130 .... 131 .... 131 .... 131 .... 132 .... 133 .... 133 .... 134 .... 135 .... 135 .... 139 .... 139 .... 141 .... 141 .... 141 .... 150 .... 152 .... 153 .... 157 .... 158 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . I . . I . . I . . I . . I . . I . CONTENTS - Continued CHAPTER 8: COSO OBSIDIAN EXCHANGE PATTERNS ................................ 161 MOJAVE B RANGE .................................................. 164 FORT IRWIN ..... ................................................ 164 THE ROSE SPRING SITE (INY-372, LOCUS 1) ..................................... 166 COSO JUNCTION RANCH SITE (INY-2284) .................................. 167 THE CONTEL PROJECT .................................................... 168 LUBKIN CREEK SITE (INY-30) ............................................... 170 KERN PLATEAU ...................................................... 170 LAKE ISABELLA AND THE TULE RIVER INDIAN RESERVATION .................. 171 GREATER LOS ANGELES VICINITY ....................................... 171 ANTELOPE AND FREMONT VALLEYS ...................................... 173 DISCUSSION ........................................................ 174 CHAPTER 9: SUMMARY AND CONCLUSIONS ................................... 177 LOCAL CHRONOLOGICAL SEQUENCE ....................................177 PREHISTORIC LAND-USE CHANGE WITHIN THE COSO VOLCANIC FIELD ............ 177 RESOURCE INTENSIFICATION IN THE COSO VOLCANIC FIELD ................ 179 COMPARISON WITH OTHER OBSIDIAN PRODUCTION SYSTEMS .................... 180 EXCHANGE, CRAFT SPECIALIZATION, AND SOCIOPOLITICAL ORGANIZATION ....... 181 REFERENCES CITED ..................................................... 185 LIST OF MAPS Map 1 Project Location .................................................2 Map 2 Important Project Locations in the Region .................................3 Map 3 Prehistoric Sites in Study Area ........................................4 Map4 Geologic Map ................................................. 8 Map 5 Known Obsidian Quarries within the Coso Volcanic Field ........................12 Map 6 Major Obsidian Sources in Eastern California ...............................17 Map 7 Regional Vegetation ...............................................52 Map 8 a Vegetation...................................................... 53 Map 9 Distribution of Early and Little Lake Period Components .......................142 Map 10 Distribution of Early Newberry and Middle Newberry Period Components ............ 143 Map 11 Distribution of Late Newberry Period Components ...........................144 Map 12 Distribution of Haiwee and Marana Period Components ........................145 LIST OF FIGURES Figure 1 Pairings of Coso Hydration Values and Uncorrected Radiocarbon Dates from the Western Mojave Desert........................... ................ 14 Figure 2 Hydration Profiles for Coso, Casa Diablo, and Bodie Hills Source Areas ..............19 Figure 3 Obsidian Stage 2 Bifaces ............................................36 Figure 4 Obsidian Stage 3 Bifaces ad Stage 2 Uniface-B ..............................37 Figure 5 Complete Obsidian Stage 3 Flake Blanks ..................................38 Figure 6 Complete Obsidian Cores............................................39 Figure 7 Complete Obsidian Flake Tool and Unifaces ................................40 Figure 8 Complete Unifacial Slab of Granite ......................................42 Figure 9 Select Complete Unifacial Granite Millingstones .............................43 Figure 10 Select Complete Unifacial Millingstones ....................................44 vii VII CONTENTS - Continued Figure 11 Select Handstones ................................................45 Figure 12 Select Handstones ................................................46 Figure 13 Select Handstones ................................................47 Figure 14 Select Complete Non-Obsidian Cobble-Cores ................................48 Figure 15 Olivella Beads........................................................ 49 Figure 16 Miscellaneous Artifacts .............................................49 Figure 17 Temporal Distribution of Single Period Loci in 250 Year Increments ...................63 Figure 18 Hydration Data Summary from Single Period Deposits in the Coso Volcanic Field ......... 65 Figure 19 Coso Obsidian Hydration Values on Projectile Points from the Volcanic Field ............ 83 Figure 20 Coso Obsidian Hydration Values on Elko and Little Lake Points..................... 85 Figure 21 Formalized Rock-lined Hearths at INY4329, Locus 1 ......................... 137 Figure 22 Informal Hearths at INY-1906, Locus 2...................................... 138 Figure 23 Seasonal Availability of Recovered Seeds ................................. 156 Figure 24 Frequency Distribution of Coso Hydration Values for Various Projects in the Western Mojave Desert ................. ................... ........ 165 Figure 25 Frequency Distribution of Coso Hydration Values for Various Projects in Southern Owens Valley and the Southern Sierra Nevada ......................... 169 Figure 26 Frequency Distribution of Coso Hydration Values for Various Projects in Coastal Southern California and the Coso Volcanic Field ....................... 172 LIST OF PLATES Plate 1 Desert Side-notched, Cottonwood, and Saratoga Spring Projectile Points ............... 74 Plate 2 Rose Spring Projectile Points .........................................75 Plate 3 Humboldt Basal-notched and Undifferentiated Humboldt Projectile Points .............. 76 Plate 4 Elko Series Projectile Points ...........................................77 Plate 5 Elko Series Projectile Points .......................................... 78 Plate 6 Gypsum and Little Lake Series Projectile Points .............................. 79 Plate 7 Silver Lake and Lake Mohave Variants of Great Basin Stemmed Projectile Points...........80 Plate 8 Lake Mohave Variants of Great Basin Stemmed Projectile Points ...................81 Plate 9 Concave-based and Leaf-shaped Projectile Points .............................82 LIST OF TABLES Table 1 Excavated Sites ........................................................1 Table 2 Types of Loci Investigated ........................................... 5 Table 3 Intra-source Hydration Relationships at the Lubkin Creek Site (CA-INY-30), from Basgall 1990. ................................................. 13 Table 4 Pairings of Coso Hydration Values and Uncorrected Radiocarbon Dates from the Western Mojave Desert . .............................................. 15 Table 5 Flaked Stone Assemblages from Selected Casa Diablo Sites.......................... 20 Table 6 Materials Recovered from Project Sites . .....................................32 Table 7 Average Monthly and Mean Annual Precipitation and Temperature at Select Locations in the Region ........................................... 54 Table 8 Important Economic Plants of Southeastern California ..........................55 Table 9 Chronological Data from Project Sits .................................... 62 Table 10 Chronological Periods for the Coso Volcanic Field ............................64 Table 11 Chronometric Summary from Component Areas by Site .........................66 Table 12 Uncorrected Radiocarbon Dates from Project Sites ............................... 69 Table 13 Radiocarbon and Hydration Pairings from Project Sites .........................70 Table 14 Projectile Points by Material Type ......................................71 viii CONTENTS - Continued Table 15 Summay Metrics of Obsidian Projectile Points ..................................72 Table 16 Hydration Data Summary of Coso Obsidian Projectile Points from the Volcanic Field ....... 73 Table 17 Frequencies of Coso Obsidian Projectile Points by Type and Period .................84 Table 18 Hydration Data on Coso Obsidian Projectile Points from Various Projects.............. 86 Table 19 Diagnostic Projectile Points in Components.................................... 88 Table 20 Assemblage Inventories in Early Period Components ...........................90 Table 21 Summa Metics of Cores in Early Period Components ......................... 91 Table 22 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Early Period Components ............................................. 92 Table 23 Summary Attributes of Bifaces in Early Period Components .......................93 Table 24 S ma Metrics of Unifaces in Early Period Components.......................... 93 Table 25 Debitage S ma Data for Early Period Co onents .......................... 94 Table 26 Summary Attributes of Simple Flake Tools at Off-Quarry Contexts by Period ............ 96 Table 27 Assemblage Inventories in Little Lake Period Components .......................98 Table 28 Summary Metrics of Cores in Little Lake Period Components .....................99 Table 29 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Little Lake Period Components ....................................... 100 Table 30 Summary Attributes of Bifaces in Little Lake Period Components .....................101 Table 31 Summary Metrics of Unifaces in Little Lake Period Components .................. 101 Table 32 Debitage Summary Data for Little Lake Period Components ..................... 102 Table 33 Assemblage Inventories in Early Newberry Period Component ................... 103 Table 34 Summary Metrics of Cores in Early Newberry Period Components ................. 104 Table 35 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Early Newberry Period Components ............................. ...... 105 Table 36 Smmry Attributes of Bifaces in Early Newberry Period Components............... 106 Table 37 Summary Metrics of Unifaces in Early Newberry Period Components .................. 106 Table 38 Debitage Summary Data for Early Newberry Period Components ..................... 107 Table 39 Assemblage Inventories in Middle Newberry Period Components .................. 110 Table 40 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Middle Newberry Period Components ................................... 111 Table 41 Summary Attributes of Bifaces in Middle Newberry Period Components.............. 111 Table 42 Debitage Summary Data for Middle Newberry Period Components ................. .111 Table 43 Assemblage Inventories in Late Newberry Period Component .................... 113 Table 44 Summary Metrics of Cores in Late Newberry Period Components .....................114 Table 45 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Late Newberry Period Components ..................................... 115 Table 46 Summary Attributes of Bifaces in Late Newberry Period Components ................. 116 Table 47 Smmary Metrics of Unifaces in Late Newberry Period Components ................... 116 Table 48 Debitage Summary Data for Late Newberry Period Components................... 117 Table 49 Projectile Points and Biface/Points from Late Newberry Off-Quarry Components ........ 117 Table 50 Assemblage Inventories in Haiwee Period Components ......................... 119 Table 51 Summary Metrics of Cores in Haiwee Period Components ...................... 120 Table 52 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Haiwee Period Components ......................................... 121 Table 53 Summary Attributes of Bifaces in Haiwee Period Components ....................... 122 Table 54 Debitage Summary Data for Haiwee Period Components ....................... 122 Table 55 Summary Metrics of Unifaces in Haiwee Period Components ..................... 123 Table 56 Assemblage Inventories at Marana Period Comonents ......................... 125 Table 57 Summary Metrics of Cores in Marana Period Components ...................... 126 Table 58 Summary Metrics of Bifaces, Uniface-B's, and Flake Blanks in Marana Period Components ......................................... 127 Table 59 Summary Attributes of Bifaces in Marana Period Components ....................... 127 Table 60 Debitage Summary Data for Marana Period Components ....................... 128 ix CONTENTS - Continued Table 61 Subsurface Debitage Densities by Period and Context (CU-1/4" m3) elf"%1L AE ~','qt 1 Uale o Table 63 Table 64 Table 65 Table 66 Table 67 Table 68 Table 69 Table 70 Table 71 Table 72 Table 73 Table 74 Table 75 Table 76 Table 77 Table 78 Table 79 Table 80 Table 81 Table 82 Table 83 Table 84 Table 85 Table 86 Table 87 ............... 129 Fvarean renuo reatur.pes ................................................ Frequency Distribution of Component Types Across Time ................... Frequency Distribution of Selected Artifacts Across Time ........................ Frequency Distribution of Feature Types Across Time ...................... Millingstone Surface Configuration by Form ............................ Millingstone Surface Configuration by Wear Type ........................ Millingstone Form by Bifacial and Unifacial Wear ........................ Millingstone Surface Configuration by Bifacial and Unifacial Wear .............. Millingstone Attributes Across Time ................................. Millingstone Surface Configuration Across Time ......................... Millingstone Form by Condition ................................... Handstone Wear by Material Type .................................. Frequency of Unifacial and Bifacial Handstones Across Time ................. Handstone Shaping, Secondary Wear, and Condition by Material Type ............ Handstone Surface Configuration Across Time ........................... Non-Obsidian Cobble-Core Material Types Across Time .................... Non-Obsidian Cobble-Core Wear Intensity Across Time ..................... Non-Obsidian Cobble-Core Wear Type Across Time ....................... Plant Macrofossils from Select Features ................................ Faunal Remains from Components .................................. Coso Glass Hydration Values for Various Archaeological Projects ............... Coso Obsidian EHT Corrections for Select Locations (adapted from Basgall 1990) .... Subsurface Distribution of Projectile Points and Debitage from INY-372, Locus 1 (1987-89). ...................................... Hydration Profile and Debitage Characterization at INY-2284 ................. Hydration Data Compiled from the Greater Los Angeles Vicinity ............... Single Period Loci Densities Standardized to 100-year Spans in the Coso Volcanic Field (from Gilreath and Hildebrandt 1991:Table 19) .......................... ... 136 ... 139 ... 140 140 146 ... 146 ... 147 . 147 ... 148 ... 149 ... 149 ... 150 ... 151 ... 151 ... 152 ... 153 154 ... 154 .. 155 ... 157 ... 162 ... 164 ... 167 ... 168 ... 173 ... 174 x CHAPTER 1: INTRODUCTION This report summarizes recent archaeological excavations in the Coso Volcanic Field, Inyo County, California (Maps 1 and 2). Adniistered by China Lake Naval Air Weapons Station, the area is characterized by vast deposits of naturally occurring obsidian that were extensively quarried throughout prehistory. It also contains a great underground reservoir of super-heated water, currently evident at active fumaroles. In addition to providing hot springs used by both prehistoric and historic peoples, the reservoir has been developed and generates enough electricity for over 240,000 urban dwellers. Known sites in the Coso Volcanic Field include about 150 quarry sites, identified as such by the natural presence of workable obsidian, 300 off-quarry sites, and 100 segregated reduction locations. This report presents results of excavations at 34 of 62 prehistoric sites within part of the Coso Volcanic Field (Map 3, Table 1). The sites investigated include 20 quarries and 14 off-quarry sites, but no segregated reduction locations. It is apparent from previous descriptions of these segregated reduction locations, where one or several cobbles were reduced to a few cores and/or bifaces, that the reduction activities at them differed only in intensity, not in kind, from activities carried out at quarry and off-quarry sites. Consequently, their absence in our data base is of minimal importance since it creates little interpretive biases. Of the 11 prehistoric site types identified in the Coso Volcanic Field, our data base includes at least one example of all but one of these types (Table 2). Brief Encampments are the most common type of off-quarry loci in the Volcanic Field; over half of the off-quarry sites we investigated were previously classified to this type (Table 2). Perfunctory Quarries are the most common type of quarry loci; nearly half of the quarries studied here were of this type. In a general sense, then, the sites excavated and described herein, include a variety of site types, with those most prevalent, best represented. Excavations were extensive (435.4 mi3). Combined with surface collection, over 7,500 artifacts and 185,000 pieces of debitage were recovered. Table 1. Excavated Sites. Site Number Site Number Site Number INY-1816 INY-3015 INY-4320 INY-1824 Main INY-3299 INY-4322/H INY-1824-B INY-3300 INY-4324/H INY-1824-X INY-3456 INY-4325 INY-1906 INY-4239 INY-4327 INY-1907 INY-4240 INY-4328 INY-1984 INY-4243 INY-4329 INY-2103 INY-4244 INY-4330 INY-2825 INY-4246 INY-4331 INY-2826 INY-4252 INY-4378 INY-3004/3005 INY-4267 INY-3012 INY-4319 NOTE: California State Trinomial site designations are employed. This volume begins with a summary of our research goals, outlining the outstanding issues addressed by the study (Chapter 2). Field and analytical methods follow, summarizing approaches to excavation, artifact classification and sampling, and other procedural matters (Chapter 3). The next five chapters present data directly corre ing to the research issues outlined in Chapter 2. These include a review of critical local and regional environmental variables (Chapter 4), development of a local chronological sequence (Chapter 5), a diachronic accounting of lithic production activities (Chapter 6), and an evaluation of how these activities related to patterns of subsistence and settlement (Chapter 7). Finally, Coso obsidian and its relevance to prehistoric exchange patterns is addressed in Chapter 8, followed by a synthesis of important project findings (Chapter 9). 1 SAN FRANCISC ~'-. o 100 MILES 0 200 KILOMETERS ~..+ {? ... :: O~~ ti~.E ? 200 tOMEWR AREA RIDGECREST LOS ANGELES Map 1. Project Location. 2 VALLEY *MOJAVE B RANGE *FORT IRWIN *ANTELOPE/FREMONT VALLEYS ANGELES COUNTY 0 Cw 0I d.,- 0 100 MILES 0 Map 2. Important Project Locations in the Region. 3 Alts. I? also 10 0- V-0 *;p (""I V-A V-t V-A C"I 13 C -Ari CA- 1 rl S1300 FEET I - E l ETN Map 3. Prehistoric Sites in Study Area. 4 CA4 Table 2. Types of Loci Investigated. Coso Volcanic Field Sample Excavated No. % No. % Quarry Loci/Sites Limited Residential Quarry 34 18.6 5 26.3 Perfunctory Quarry 131 71.6 9 47.3 Quarry Location Encampment 13 7.1 3 15.8 Large Lag Quarry 3 1.6 1 5.3 Quarry Milling Zone 2 1.1 1 5.3 Subtotal 183 100.0 19 100.0 Unclassified/Unknown 50 -- 2 -- Off-Quarry Loci/Sites Brief Encampments 247 69.5 17 56.7 Limited Habitation Encampments 35 9.9 7 23.3 Production Station 11 3.1 2 6.7 Milling Camp 13 3.7 3 10.0 Isolated Feature/Milling Station 49 13.8 1 3.3 Subtotal 355 100.0 30 100.0 Unclassified/Unknown 54 -- 1 -- Segregated Reduction Loci 89 -- -- -- NOTE: Many sites contain multiple loci. 5 CHAPTER 2: RESEARCH ISSUES The major goal of this project is to monitor the prehistoric production, use, and exchange of Coso obsidian, and determine the relationship of these activities to other socioeconomic developments in the region. To achieve this goal, several research problems have been developed. Derived from previous studies in eastern California and similar projects conducted in other parts of the world, these issues reflect current perspectives in hunter-gatherer research as applied to prehistoric use of obsidian-rich environments. The first of four issues considered is the area's environmental productivity, both in terms of hard resources (e.g., obsidian) and soft resources (e.g., food). The second research problem revolves around chronology and determining the age of project sites. Documenting prehistoric lithic exploitation patterns is the third problem domain. The final issue concerns subsistence and settlement practices. ENVIRONMENTAL PRODUCTIVITY The Coso Volcanic Field lies in the southwestern portion of the Coso Range, east of the Sierra Nevada, south of Owens Valley, west of the Argus Range, and north of Indian Wells Valley (Map 2). It is characterized by numerous explosion craters, rhyolitic domes, debris flows, and pyroclastic deposits, most originating from volcanic events that occurred between 1.5 million and 33,000 years B.P. (Map 4; Duffield et al. 1980). Most of the original basement rock consists of Mesozoic granitics which protrude the more recent volcanic material in the form of sporadic, discontinuous ridge systems. Pyroclastic fallout fills most of the small valleys that have formed between the ridges and domes, creating a landscape that clearly reflects its violent past. Due to the complexity of these multiple volcanic events, many of the valleys have been enclosed, resulting in several small, internally drained playas. Naturally occurring obsidian is abu ly present throughout the Coso Volcanic Field, and available from a variety of contexts. Survey of approximately 24,000 acres has identified over 150 quarry sites (Gilreath and Hildebrandt 1991). These include several primary outs of glass flowing from the sides of steep rhyolitic domes, with high quality obsidian often occurring as large boulders and slabs. Most of these are on Sugarloaf Mountain where many tons of high quality material are still available. The most impressive of this group was originally termed the "Colossal Quarry" by Harrington (1951) and described as: More than two miles of ancient diggings, plainly visible along the edge of the blufl.! A hillside covered with hundreds of tons of obsidian quarry refuse left by prehistoric workers! It did not seem possible, yet there it was... We stood there in amazement that day ...thinking of the inm erable human hands, the countless generations that must have been required to produce such a result [1951:15]. Secondary quarries occur as obsidian float in major debris flows or in concentrations of air fall sporadically scattered across the land during volcanic eruptions. These quarries are numerous and usually more accessible than the others. They vary from small pockets that could have been played-out during a single visit, to extensive nodule scatters encompassing tens of acres. The amount of useable obsidian in the most extensive secondary deposits, however, seems inconsequential when compared to the amount at extensive primary quarries. These secondary quarries occur on ridgetops, exposed as lag after fine-grained sediments have eroded away (Elston and Zeier 1984). The obsidian ranges from pebbles to cobble-sized fragments reaching 30 cm in diameter. Every major natural obsidian deposit discovered during survey was exploited (i.e., contains chipping debris), but there are many tons of useable obsidian still available both at primary and secondary deposits. Subsistence resources are less abundant. Owing largely to the rain shadow effect of the Sierra Nevada, precipitation averages only 14 cm (5 in.) per year, with the majority (75 %) occurring between December and March (Troxell and Hofman 1954). Convectional thunderstorms originating from the Southwest and Gulf of Mexico provide additional moisture from July through October, however, potential evapotranspiration exceeds annual 7 ANo~ K. 7 IH ALLUVIUM ALLUVIAL PAN AND MINOR FLUVIAL DEOSr BASE ROCKS - PRINCIPALLY GRANITIC INTRUSIVE ROCKS OP MESOZOIC AGE. PYROCLASnC DEPOSITS WELL-BEDD!D FRAGOM AL DEPOSITS OF PUMICE: OSIDIAN, AND BASIMENT ROCKS; LOCALLY REWORKED FROM HILLSIDES, FORMS EXPLOSIVE RINGS AROUND SOME DOMES AND GENERALLY MANTIES ENTIRE AREA OF VOLCANIC FIELD; INCLUDES MINOR PLAYA DEPOSITS STEEP-SIDED FLOWS AND DOMES MOST COVERED BY A CARAPACE OF VESICULAR RHYOLME TIIROUGI I WHIaCt OBSIDIAN PROTRUDES LOCALLY. 0 3 anOER VOLCANIC MILES MILETERS 0, 3 ~CEt~s. a Map 4. Geologic Map. 8 precipitation by a wide margin, resulting in a sparse vegetative cover of Mojave Creosote Bush Scrub, Mojave Mixed Woody Scrub, and Desert Saltbush Scrub (Billings 1951; Leitner and Leitner 1988; Thomrthwaite 1948). A wide range of plant and animal resources were exploited within these environmental zones, but their availability varied considerably from one season to the next, and from year-to-year depending on local climatic conditions. To evaluate the importance of these resources, and ultimately determine their influence on long-term strategies of subsistence and settlement, it is first necessary to measure local resource abundance and diversity relative to that of surrounding regions (e.g., the well-watered lands of Owens Valley). This will be accomplished through tracing the distribution of local vegetation communities (Leitner and Leitner 1988) and translating this information into measures of plant and animal productivity. Successful completion of this process requires review of ethnographic accounts of subsistence technology and ecological data concerning seasonal and annual cycles in resource pctivity. Much of this work has already been completed by Delacorte (1990) who organized particular habitat zones common to eastern California into seasonally specific resource packages. Combining each package with the most-likely techniques used in their exploitation allows estimates of search, pursuit, and processing costs for each, and ultimately an overall ranking of the various environmental circumstances existing in the region. CHRONOLOGY Our ability to place prehistoric sites in their temporal context is prerequisite to developing a use-history for the Volcanic Field. Because we want to know when the area was occupied, if occupational intensity and focus changed through time, and if diachronic variability in obsidian exploitation patterns existed, we want to estimate the age of as many sites as possible. The nature of the prehistoric record within the Volcanic Field is such that only three dating techniques are useful in positing age estimates: radiocarbon assay, cross-dating of diagnostic artifacts, and obsidian hydration. RADIOCARBON DATES FROM THE VOLCANIC FIELD Unfortunately, prehistoric cultural deposits containing organic residues are the exception, not the rule, within the Volcanic Field. True midden deposits have accumulated at very few locations, and structural remns are equally rare. Consequently radiocarbon samples are primarily obtained from hearths, roasting pits, and other such comparatively small features. Furthermore, rarely are such features discovered in buried context. An inordinately high percentage of those features excavated in the Volcanic Field could be seen on the ground, suggesting in and of itself that they are fairly recent. Notwithstanding these problems of visibility, an occupational history of the Volcanic Field reliant on the radiocarbon dates would also, in effect, document when specific kinds of activities involving hearth usage occurred, but those episodes need not have been in phase with activities related to stone working. CROSS-DATING OF DIAGNOSTIC ARTIFACTS FROM THE VOLCANIC FIELD Prehistoric artifacts from the Volcanic Field most amenable to cross-dating include projectile points, several types of beads, and pottery. Projectile points occur at more sites than either beads or pottery, but only about 150 have been discovered. Stating what many North American archaeologists before them had come to conclude, Heizer and Hester wrote "projectile points are subject to distinctive morphological variation through time, and these changes have made them extremely important as 'time markers' (the 'historical-index' types of Steward 1954)...no one can deny the value of projectile point types as guides to the chronological ordering of prehistoric cultural development" (1978:153). By compiling lists of radiocarbon dated deposits in the Great Basin where various projectile point types had been recovered, Heizer and Hester reaffirmed that different styles occurred during different intervals, and offered temporal ranges for various commonly recognized types: Pinto, Humboldt, Elko, Rose Spring, Eastgate, Desert Side-notched, and Cottonwood, among others. 9 In seeking to develop a local chronological sequence for Monitor Valley, Nevada, Thomas (1981) devised a classification scheme for projectile points that relied on a variety of metric attributes. This approach served to remove much of the subjectivity from classifications, and allowed researchers working with different collections to directly compare the morphological characteristics of their points to those found elsewhere. The Monitor Valley projectile point chronological sequence has come to serve as a standard for cross-dating points throughout the Great Basin. Several years ago, Flenniken and Raymond made the observation that "the contingencies of point manufacture, hafting, use, and rejuvenation create morphological changes that may render questionable use of these morphological typologies as prehistoric cultural markers" (1986:603). In the course of repairing a break, the morphology of the point may change such that the repaired one ill-conforms to the ideal type. This rather plain observation has generated considerable debate (Bettinger et al. 1991; Flenniken and Wilke 1989; Thomas 1986; Wilke and Flennien 1991) regarding the degree to which reworking is of concern to chronology: is there evidence that prehistoric individuals actually did rework broken points in a fashion that mutes their veracity as chronological markers; are particular types of projectile points more susceptible/amenable to reworking than others? The argument has expanded, since, to question the temporal placement of dart-points (special emphasis given to Elko, Little Lake, Gypsum, and Humboldt series) ine the Great Basin, with one side contending that varaious typies are derivatives of one another (changing form during the use-life of the artifact), and that the types are not consecutively occurring temporally diagnostic forms, rather they co-exist in deposits ranging from 8500-1500 B.P. (Flenniken and Wilke 1989). The other side maintains that they are temporally consecutive types that serve as chronological markers (Bettinger et al. 1991). Data from the Coso Volcanic Field may provide some insight into this issue. The projectile points recovered from the Volcanic Field will be described in a fashion consistent with the Monitor Valley projectile point key (Thomas 1981), and referenced using the point type nomenclature that has developed in the archaeological literature for the region. Hydration rim values recorded on the projectile points will be presented, and used to independently evaluate the temporal significance of morphologically distinct types. OBSIDIAN HYDRATION DATING IN TE COSO VOLCANIC FIELD That moisture penetrates volcanic glass at a predictable and, hence, quantifiable rate is at the foundation of obsidian hydration dating. The assumption holds that the hydration rind on a surface of an obsidian flake broken off from the parent mass some 2000 years ago will be consistently wider than the rind on a flake broken from the same piece only 1000 years ago, all else being equal. Since the inception of obsidian hydration studies in North America, Coso glass has commonly been a subject of discussion, resulting in it being one of the most thoroughly investigated obsidians in North America. Friedman and Smith (1960) were the first to recognize the chronometric potential of obsidian. Their initial study concluded that: (i) "hydration thickness increased as the square root of time, (ii) the rate was independent of relative humidity but increased with temperature, and (iii) the chemical composition of the obsidian appeared to affect the rate" (Friedman and Long 1976:347). Picking up where the initial study left off, and through a series of expments, Friedman and Long concluded: 'it is possible to calculate the hydration rate of a sample from its silica content, refractive index, or chemical index and the knowledge of the effective temperature at which the hydration occurred" (1976:347). These and other potentially key variables have been the focus of many subsequent studies. In developing a hydration rate for Coso glass at the source, Zeier and Elston (1984) made pioneering attempts to control for temperature variability, giving consideration to such factors as ambient air temperature, ground surface temperature, slope, exposure, vegetation canopy, and elevation. Their results were far from conclusive regarding many of these conceivably influential factors, but they were able to adequately reaffirm that effective hydration temperature (EHT) greatly influenced hydration rates. Ambrose cells placed in the Sugarloaf Mountain area verified that the EHT was high compared to many other parts of California and the Great Basin, and quite variable within the Volcanic Field. 10 Hughes (1988) and Ericson (1989) have spent considerable energy examning obsidian geochemical variability within the Volcanic Field. In spite of early geological studies that showed obsidian was available at numerous localities within the field (Bacon et al. 1981, 1982; Duffield and Bacon 1981; Duffield et al. 1980; Lamphere et al. 1975), early archaeological sourcing research was satisfied to recognize "Coso" as geochemically dit from other major obsidian sources in eastern California and Nevada. Interest in the magnitude of intra-field geochemical variability grew in response to increasing appreciation of the fact that glass from different sources absorbs water at differential rates. To study intra-field variability at Coso, Hughes processed samples from 15 locations, and disrated four groups based on differing ratios of rubidium (Rb) and zirconium (Zr): West Sugarloaf, Sugarloaf, West Cactus Peak, and Joshua Ridge. Map 5 (taken from Gilreath and Hildebrandt 1991) shows the distribution of recorded obsidian quarries in the Volcanic Field. West Sugarloaf glass occurs in the south- central portion of the project area, and Sugarloaf in the northern half of the project area. West Cactus and Joshua Ridge quarries occur at the northern and southern ends of the Volcanic Field, respectively. Having shown that these four subgroups have different chemical indices (Hughes 1988:Table 4), he concluded that "each of them should hydrate at a slightly different rate" (Hughes 1988:262). Regarding prehistoric exploitation of the obsidian available in the Volcanic Field, his analysis of projectile points found at the Rose Spring (CA-INY-372) and Stahl (CA-INY- 182) sites, which are located within 12 km of the Volcanic Field and each other, suggested West Sugarloaf and Sugarloaf deposits were more heavily utilized in prehistoric ties than either West Cactus or Joshua Ridge deposits. Sourcing data from the Lubkin Creek site (CA-INY-30 [Basgall and McGuire 1988]) provided additional support for this pattern, with comparatively few West Cactus or Joshua Ridge specimens identified in archaeological materials. The prevalence of West Sugarloaf and Sugarloaf obsidian in prehistoric sites prompted subsequent hydration and sourcing investigations to concentrate on these two subgroups, largely ignoring West Cactus and Joshua Ridge. in an attempt to independently verify the geochemical distnctveness of West Sugarloaf and Sugarloaf glass, Bouey (1991) re-analyzed 30 archaeological aricts htat had been previously ascribed to these two subgroups, and found that many rubidium/zirconium ratios fell outside of the 95% confidence intervals Hughes (1988) delimited that distinguished one from the other. Bouey offers two explanations for why his results diverged from Hughes': (1) Hughes' study may have suffered from inadequate sample size (i.e., had Hughes used larger sample sizes and/or mre sampling localities, he might have documented more variability and hence greater overlap between the two subgroups); and (2) the geochemical fingerprinting technique is simply not precise enough to render such accurate distinctions. Comparatively simple factors that influence precision include "slight alterations in sample surface configuration and placement" relative to the angle at which the X-ray diffraction beam strikes the artifact or flake's surface (Bouey 1991:309). We will leave to others the issue as to whether analytical techniques are acure enough to distinguish true West Sugarloaf glass from true Sugarloaf specimens. However, an important chronological issue remains: does intra-field geochemical variability influence he rate of hydration in a fashion that we need concern ourselves with? Data from the iLubkin Creek site have been ied (Basgall 1990; Basgall and McGuire 1988) to address this specific question. At that site south of Lone Pine, arfactal specimens of West Sugarloaf and Sugarloaf glass were recovered together in some of the most intact depositional contexts imaginable. In two cases, tool caches were found on the floors of burned prehistoric houses. As Table 3 shows, hydration measures on co-associated West Sugarloaf and Sugarloaf specimens are not significantly different. The data are admittedly meager, but it is unlikely that more robust, inarguably co-associated rits will be recovered in the near future. These data suggest that the magnitude of influence that intra-source variability has on hydration rim formations is inconsequential. On a different tack, Stevenson (1987), and Stevenson and Scheetz (1989), among others (Ericson 1989; Tremaine 1989; Tremaine and Fredrickson 1988) have been emining variability in rim formation processes through induced-hydration experiments. Results of the experiments conducted by Stevenson (1987, 1990) discussed by Cleland (1988, 1989, 1990) raise serious questions about the replicability of induced-hydration experiments. In 1988, Cleland presented hydration rates for Sugarloaf and West Sugarloaf glass based on the results from a liquid bath experiment involving one specimen ascribed to each subgroup. The rate for Sugarloaf was stated as 27.2 p3/1000 years and for West Sugarloaf at 38.7 p2/1000 years - the latter hydrating 42% faster than the former. This set of experiments was followed by another that attempted to replicate the results using a vapor rather than liquid bath matrix. The vapor matrix e rimnts resulted in ted rates of 23.3 p2/1000 years for Sugarloaf glass and 12.9 p2/1000 years for West Sugarloaf glass (Cleland 1989), reversing the order of which subgroup hydrates fastest. 11 WEST SUGARLOAF SUGARLOAF * WEST CACTUS i JOSHUA RIDGE SUGARLOAF, WEST SUGARLOAF, AND WEST CACTUS + MATERIAL 1 A SUGARLOAF/CACTUS WEST CACTUS AND SUGARLOAF/CACTUS * NON-OBSIDIAN * NO SOURCING DATA Map 5. Known Obsidian Quarries within the Coso Volcanic Field (from Gilreath and Hildebrandt 1991). 12 Table 3. Intra-source Hydration Relationships at the Lubkin Creek Site (CA-INY-30), from Basgall 1990. West Sugarloaf Sugarloaf Mountain mean hydra- mean hydra- n tion (i) s.d. n tion (A) s.d. Cache 1 (Humboldt Basal-notched Points) 2 5.65 0.21 3 5.70 0.53 Cache 2 (bifacial preforms) 2 5.35 0.35 3 5.30 0.70 Desert Side-notched Points 11 3.45 0.69 1 3.00 -- Cottonwood Triangular Points 9 2.78 0.82 2 3.75 1.06 Humboldt Basal-notched Bifaces 6 5.36 0.50 4 5.45 0.66 Structure 11 8 5.69 0.50 7 5.14 0.67 The profound difference in the liquid bath and vapor matrix experiments results for the West Sugarloaf samples prompted the analyst to remeasure the liquid bath sample, and revise the 38.7 //1000 years to 10.9 /2/1000 years. Depending on the results one prefers, Sugarloaf glass might be hydrating 80% or 161 % faster than West Sugarloaf glass. A third set of experiments attempted to replicate results obtained in the vapor matrix tests (Cleland 1990). The single West Sugarloaf and single Sugarloaf sample this time produced results comparable to one another - results that were also similar to that of the West Sugarloaf sample in the first vapor matrix test. Supplemental geochemical analysis of the piece of glass classified as "Sugarloaf" that was used in the liquid bath experiment and the first vapor matrix experiment prompted Cleland (1990) to hypothesize that it derived from a not-previously- identified, fifth, geochemical subgroup in the Volcanic Field. In short, the confusion resulting from the various results reported for liquid bath and vapor matrix experiments on West Sugarloaf and Sugarloaf obsidian, leave us doubtful that these induced-hydration results are replicable. Because of these problems, we do not consider their results relevant to the current project. Most recently, some glass scientists have turned their attention to inherent moisture content of the toolstone itself as a potential factor influencing hydration rim formation (Knaus and Mazer 1991; Stevenson et al. 1990, 1993). Citing Mazer et al. 1992, Stevenson et al. (1993:371) report "Recent experimental results have shown that the intrinsic water content of an obsidian is the dominant hydration rate-controlling compositional parameter." Analyzing specimens from the four subgroups defined by Hughes (1988), Stevenson et al. (1993) show that nodules of obsidian from lag deposits within the Volcanic Field, in contrast to primary deposits, contain great amounts of water due to the explosive events associated with their deposition. But, the intrinsic water content found within each geochemical subgroup was found to be extremely variable. The authors suggest that this variability may in part account for the numerous Coso rates that have been offered. Likthe iuced-hydration experiments, the e'intrinsic water content" studies are nascent, and, as such, do not influence our interpretation of hydration results. Continued refinement in interpretng hydration analysis is likely to result from studies such as the induced- hydration experiments and intrinsic water content analyses. It is doubtful, though, that they will radically alter the basic tenet that wider hydration bands result from exposure over longer periods of time, and, all other things being equal, rim formation is slower in a cool depositional environment than in a warm environment. In spite of all the potential sources of variability of Coso hydration, the archaeological record from the western Mojave Desert indicates that Coso hydration rim values and age are strongly correlated (Figure 1). We can glean from the literature 28 pairings of Coso hydration values and radiocarbon dates from the western Mojave Desert (Table 4). Other chronological indicators found in association with these pairings (e.g., projectile points and pottery) provide independe support of each locality's chronological integrity (see Chapter 5, Table 18). These data clearly indicate that wider micron values mean older samples. Confronted with such compelling data, we are not inclined to join the ranks of the hydration nay-sayers. It is clearly our position that Coso glass hydrates at a predictable and quantifiable rate, and that EHT of the area in which sites containg Coso glass occur must be factored into the hydration rate equation. Giving further examination to Figure 1, recognize that southern Owens Valley, where INY- 30 (depicted by triangles), and INY-3806/H and INY-3812 (depicted by hexagons) are found, has a lower mean annual temperature than Fort Irwin; and that Rose Valley, where INY-372 occurs, has a mean annual temperature nearer to that of southern Owens Valley than Fort Irwin. Since Fort Irwin has a higher EHT than the other areas, 13