UC Merced Journal of California and Great Basin Anthropology

Title Lithic Raw Material Prospects in the Mojave Desert, California

Permalink https://escholarship.org/uc/item/8np7g12w

Journal Journal of California and Great Basin Anthropology, 11(2)

ISSN 0191-3557

Authors Wilke, Philip J. Schroth, Adella B.

Publication Date 1989-07-01

Peer reviewed

eScholarship.org Powered by the California Digital Library University of California Journal of California and Great Basin Anthropology Vol. 11, No. 2, pp. 146-174 (1989).

Lithic Raw Material Prospects in the Mojave Desert, California

PHILIP J. WILKE and ADELLA B. SCHROTH, Dept. of Anthropology, Univ. of California, Riverside, CA 92521.

A HIS paper discusses lithic raw material THEORETICAL BACKGROUND prospects (or simply "prospects"), places where potentially flakeable tool stone was Acquisition of tool stone by aboriginal assayed or tested for quality. It characterizes peoples was an industry that in terms of scale this site type and contrasts it with quarries, varied greatly from one situation to another. places where stone was obtained consistently The large and well-known quarries of the and in quantity, and places where stone was western United States represent one end of picked up, used, and discarded with little the spectrum. These include the AUbates modification. We believe prospects represent silicified dolomite quarries, Texas; Spanish a major archaeological site type that has re­ Diggings quartzite quarries, Wyoming; ceived inadequate attention in the literature. Tosawihi opalite quarries, Nevada; and Casa We describe here a prospect site (CA- Diablo and Coso obsidian quarries, California, SBr-5872), characterize its assemblage, and to name a few. The thousands of metric tons interpret the behavioral context represented of tool stone that must have been taken from by it. We believe the activities represented at these and other major quarries suggest for­ the site were fully embedded within some pri­ malized acquisition, reduction, and distribu­ mary (probably subsistence-related) strategy, tion of stone, perhaps in some cases by but that the activity was more structured than specialists, throughout much of prehistory. the casual selection of useful pieces of stone. The physical and chemical tracing of stone We also believe the site is typical of many on from such sources to elucidate patterns of the landscape of the western United States prehistoric exchange is a major research effort and that such assemblages can only be in archaeology. interpreted with respect to the behavioral At the other end of the spectrum is non- contexts responsible for them. We also draw quarried stone, that obtained in a casual and comparisons with several other examples of expedient manner as the occasion required by raw material prospects in the Mojave Desert. people engaged in other activities. Few The principal site described here is one at studies have focused on such informal indus­ which people occupied a short-term camp, tries in the aboriginal western United States. probably many times. During such occupa­ Gould (1977) discussed acquisition of non- tions, they tested cobbles and obtained tool quarried stone as a common aspect of Aus­ stone in limited supplies and of varying tralian aboriginal lithic technology. He wrote: quality on the immediately adjacent terrain. Consistent acquisition of high-quality tool At the outset it is important to distinguish between stone materials gathered by the desert stone occurred at actual quarries 1-2 km. away Aborigines from definite quarries, that is where good-quality chert, chalcedony, and specific localities where usable stone is available jasper were obtained in quantity. known to the Aborigines and visited by them LITHIC RAW MATERIAL PROSPECTS 147

[sic], and non-quarried stone, which is obtained schedules. Very rarely, and then only when from the surface of the ground at or near the things have gone wrong, does one go out into the spot where it is needed for a particular task. environment for the express and exclusive purpose In this latter case, the stone comes from a non- of obtaining raw material for tools [Binford localised source which may be visited only once 1979:259, emphasis in original]. [Gould 1977:163]. The extent to which these comments were Gould went on to describe the landscape of intended to be specific to the Nunamiut, or the Western Desert of Australia as one where more generally applicable to interpretation of stone suitable for tool use generally can be archaeological assemblages, is uncertain. obtained from the surface with little search or Binford's ideas may approximate Gould's effort. He (1977:164) then contrasted the description of nonquarried stone in aboriginal behavioral situation at quarries and non- Australia, but we believe his comments were quarried-stone-acquisition localities where directed more at the "hidden agenda" aspect Aborigines obtained stone: of stone acquisition. That is to say that we At quarry sites one sees Aborigines obtaining believe his comments referred to side-trips flakes and small lumps or cores which are made to quarries to obtain stone whUe people carried away and further trimmed for specific were en route to some other destination, such uses. ... At more generalised non-quarry as on some subsistence-related pursuit. What­ localities, however, stones were used for immediate tasks on the spot. ... In every case ever the case, Binford's ideas about raw observed the Aborigines always disposed of the material acquisition are completely at odds tools that were manufactured and used at non- with common sense when one stands on any quarry locations at these same places. They of a hundred renowned aboriginal quarries in were never observed to carry the tools away to a habitation camp or some other locality for the western United States. The number and further retouch and/or use. magnitude of excavated quarry pits at some Having built a convincing case for the expe­ sites (such as Tosawihi and Alibates), the dient acquisition, reduction, use, and on-site standardized approach to cobble/boulder discard of tool stone, Gould (1977:167) con­ reduction, staging of quarry blanks and cores, cluded that actual quarried stone represented and long-distance distributional data all point "only a tiny fraction of the total amount of to formalized, highly structured, and intensive lithic material used within the cultural stone acquisition in prehistory. system." Between these two extremes (organized, Binford (1979:259-261, 270) did not dis­ industrial, primary-strategy quarrying and tinguish raw material extracted for transport distribution vs. casual, expedient, embedded- and use elsewhere from that obtained on the strategy picking up of stone off the land­ landscape for immediate use. He did say that scape), and certainly closer to the latter, was among the Nunamiut Eskimo, with whom he another strategy by which stone was acquired conducted ethnoarchaeological research, most in prehistory. This strategy was one of inten­ raw material was obtained in the context of an tionally prospecting for stone at places that extractive strategy embedded within some sometimes yielded useful material but that primary subsistence-related strategy: never did so consistently or in sufficient quantity to result in what most archaeologists Raw materials used in the manufacture of would call aboriginal quarries. We doubt implements are normally obtained incidentally anyone consistently went to such places to get to the normd execution of basic subsistence tasks. Put another way, procurement of raw tool stone, but people did try to find useful materials is embedded in basic subsistence material on occasion when they happened to 148 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY be at the spot for other reasons. In other thereby better understand these data. cases, they may have located outstanding veins or nodules of material and returned with CA-SBR-5872: A LITHIC RAW appropriate tools and attempted, successfully MATERIAL PROSPECT or otherwise, to remove such material. We SBr-5872 is located at the south end of the refer to these places as lithic raw material Castle Mountains, a small Tertiary volcanic prospects, or simply/»TO.spec^s. range in eastern San Bernardino County, Tool stone seldom was quarried at California, and adjacent parts of Clark prospects, and it never was quarried at County, Nevada (Fig. 1). The geologic history nonquarry sites of the kind described by of the range is complex, and involved exten­ Gould. Therefore, a clear terminology is sive volcanism and geothermal activity (Linder needed to distinguish the two site types. We 1989). Deposits of siliceous sedimentary are satisfied with the term "lithic raw material stone, apparently formed through geothermal prospect" because it clearly describes the activity by replacement of existing rhyolite behavior that occurred at such sites. We with chalcedonic quartz, outcrop along the believe, however, that nonquarried-stone- eroded base of the range. Flanking alluvial acquisition sites of the kind discussed by fans contain chert and chalcedony clasts, Gould should be identified by a label that which vary widely in occurrence and quality. more satisfactorily describes the ephemeral Extensive use was made of these materials for and nonredundant activities that occurred at tool stone by the prehistoric inhabitants of the them. Toward that end, we term such places region. Rhyolite and other rocks suitable for ephemeral stone acquisition and use sites. use in tool manufacture occur also, but were In the context of the above-referenced sought less often for tool stone. discussions, it should be pointed out that The site is located on a low ridge (eleva­ neither Gould nor Binford commented on the tion 1,275 m.) that extends south from the difficulty of pressure flaking most raw cherts, range, and consists of a highly discontinuous, chalcedonies, and jaspers, and the degree to light scatter of tested cobbles, cores, debitage, which the flakeabiHty of such stone is and other artifacts. To the east and west are improved by careful heat treatment. With ephemeral drainages. Erosion on the crest of reference to the Australian situation, this is the ridge has exposed the material that is the understandable; many (perhaps most) tools focus of the industry reported here. Flakeable were percussion-flaked from raw stone and stone is not evident on the surface in the area were intended for rough-service use such as of the flanking washes where it probably is working hardwoods. In the American West, covered by finer-grained alluvium. most of the crude quartzes were heat-treated Vegetation in the area of the site is open and intended for pressure flaking into small Joshua Tree Woodland. Plants that might be tools. This distinction makes it a bit difficult called the "overstory" are Joshua tree (Yucca to apply the ideas offered by Gould and brevifolia) and Mojave yucca {Y. schidigera). Binford directly to aboriginal California desert The "understory" consists of scattered contexts. Nevertheless, these models are creosote bush (Larrea tridentata) and abun­ important for interpreting our research. In dant blackbush (Coleogyneramosissima). Also part because they fit the aboriginal California present are silver cholla (Opuntia echinocar- desert situation imperfectly, the models force pa), galleta grass {Hilaria rigida), and many us to explore our own data more fully and other species. Galleta grass is especially LITHIC RAW MATERIAL PROSPECTS 149

Fig. 1. Location of SBr-5872. 150 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY common across the minor drainage that of collected material later revealed that the borders the site on the east side. There are objects from Locus 10 consisted entirely of no weather stations in the immediate area, rhyolite fractured by natural weathering but rainfall is very low, probably under 20 cm. processes. Part of the material collected at per year. In the winter, frosts are common that locus was employed in experiments and and snowfall occasionally occurs. Water the remainder was discarded, leaving a total sources in the region consist of isolated of 16 loci. Most were small clusters not over springs. Piute Spring, ca. 20 km. to the 2-3 m. across, and represent places where raw southeast, gives rise to a permanent stream material was found and tested for quality. that flows for a kilometer or more before Following the initial surface examination, disappearing in the sand. collection of artifacts by locus began at the The site is in the area occupied by the southeast end of the site and proceeded historic Chemehuevi (Laird 1976; Kelly and toward the northwest. The loci were num­ Fowler 1986). These people, an offshoot of bered in the order collected. At each locus, the Southern Paiute, foraged in the area the center point was plotted and aU artifacts during the last several hundred years. No within 2 m. of that point were collected for substantive archaeological work has been analysis. Following the completion of the conducted in the region to adequately surface collection, subsurface testing was characterize their prehistoric Ufeway or that conducted by the excavation of six shovel test of their Archaic predecessors. However, for units. Four were placed in the general site a synthesis of the archaeology of the Califor­ area and two at Locus 12. No cultural re­ nia deserts, see Warren (1984). mains were found below 10 cm. depth in any of the test units. Two shovel test units yielded Field Procedures no cultural material at aU. We concluded that Fieldwork at SBr-5872 was undertaken in the entire assemblage was confined essentially advance of a proposed mining project. The to the surface. procedures followed in fieldwork included This procedure was used for loci 1 through establishing a permanent datum point,^ 9, 11, and 13 through 17. These loci consti­ identifying areas of artifact concentrations, tute what is termed hereafter "the main site mapping the site, collecting a representative area." At Locus 12, during the surface collec­ sample of lithic detritus and all formed flaked tion, several aboriginal ceramic sherds and stone artifacts, and sampling for subsurface fragments of at least three block millingstones cultural remains with 50-cm.^ shovel test units were found. More extensive mapping and col­ using a l/8th-inch screen. lecting occurred at this locus. While we have Total site area is about 5,000 m.^ when an no way of demonstrating that aU these site oval is drawn around all observed site loci loci were used at approximately the same (Fig. 2). The overall site measures about 110 time, we assume they were. m. NW-SE by 70 m. NE-SW. The actual site Whereas the loci with debris resulting configuration is, however, perhaps better from material testing essentially determined described by the occurrence of the site loci the extent of SBr-5872, it is important to than by a perimeter line that encloses them. realize that tested cobbles and a few associat­ Seventeen surface concentrations or loci ed flakes occur as isolated archaeological of lithic detritus were identified during a occurrences throughout the immediate region. series of initial transect walks. Examination How often such cobble testing resulted in the LITHIC RAW MATERIAL PROSPECTS 151

CM 13 1^ 00 in ^ CD CO < E o o 152 JOURNAL OF CALIFORMA AND GREAT BASIN ANTHROPOLOGY successful discovery of quality stone is not Raw chert and chalcedony are exceedingly known, but the number of frustrated attempts tough and produce a stronger cutting edge for is considerable. simple flake tools than can be obtained from heat-treated material. Raw material is, how­ Analysis and Discussion ever, difficult to pressure flake. As with most In this report, the term artifact refers to siliceous sedimentary rocks, raw chert and any object displaying evidence of having been chalcedony such as occurs at SBr-5872 frac­ altered by human activity. Artifacts found at tures conchoidaUy, but flake scars usually the main site area include cobbles of raw display a sugary texture that lacks luster. As chert and chalcedony, and occasionally of these crude quartzes fracture, the force rhyolite, tested for quality by the removal of rebounds around microcrystaUine or crypto- one or more flakes; flake cores of bifacial or crystalline structures and impurities. The multidirectional form; and flakes removed fracture scar therefore is not smooth. (True from tested cobbles and cores. Tested cobbles flints, with higher SiOj content, display better were very common; formed artifacts were fracture properties, but still can be improved rare. Large boulders still embedded in soil by heat treatment.) matrix had been struck in attempts to test the Substantial force is required to detach quality of the stone or to detach usable pieces. flakes from raw chert and chalcedony. When No evidence was seen of attempts to excavate flakes are detached from such material, they and remove such boulders from the site or sometimes display split cones of force^; they further reduce them; apparently the quality of have heavily battered and crushed platforms; stone in them did not warrant such efforts. the platforms often break away as the flake No evidence of quarry pits, heat-treatment bends during detachment; and such flakes may facilities, or intensive-reduction loci was have overshot (outrepasse), hinge, or step observed. terminations. Hinge and step terminations Characteristics of the Raw Stone. Nat­ indicate that the force applied to the stone urally occurring chert and chalcedony at SBr- was inadequate, or improperly directed, to 5872 are highly variable in quality. Incipient- carry the flake through to a feather termina­ cone cortex is absent, indicating that the ma­ tion. Overshot terminations usually indicate terial has not been transported any great dis­ excessive force. All of these characteristics tance from its primary geologic context. In are common in the assemblage from SBr-5872 general, the surfaces of such material are and illustrate the difficulty of working the raw heavUy weather-checked, and flakes detached stone that occurs there. from them either contain fractures or were Stone obtained on the surface usually has broken on detachment from the parent clast. numerous weather checks and cracks, caused Internally, most clasts contain vugs or crystal by wetting and freezing. Elsewhere in the pockets. Testing cobbles by detachment of western United States, where stone of good flakes enabled quality control, permitted selec­ quality occurred in abundance, it was exten­ tion of the best stone for tool use, and result­ sively quarried, and quarry pits mark the ed in discarding substantial quantities of poor- efforts of aboriginal miners to obtain high- quality material. Presumably, the best material quality stone at depths where weathering by was removed from the site for further reduc­ frost action was minimal. The quality and tion elsewhere, perhaps where its flaking abundance of stone at SBr-5872 did not quality could be enhanced by heat treatment. warrant such effort. AU stone obtained there LITHIC RAW MATERIAL PROSPECTS 153 was surface material, and therefore its quality equivalent to large bifacial flake cores was compromised to one degree or another by common throughout the prehistory of western weathering. North America (Wilke and Flenniken 1988). Cores vs. Tested Raw Material. In the Quality and configuration of the raw stone discussion that follows, cores are defined as were controlling factors in core production. formed artifacts reduced from raw material, Presence of vugs, crystal pockets, weather- and that served, or could have served, as checks, and variable lithology all precluded sources for the detachment of flakes for the production of large cores of bifacial form. cutting tools or flake blanks for tool produc­ While these imperfections may have led to tion. Many archaeologists consider the term their being discarded on the site without being "core" to describe any mass of stone from used for tool production, cores retain little which flakes or blades are struck (including cortex, display evidence of platform prepara­ what we term "tested raw material"), whether tion, have numerous flake-detachment scars, or not the mass has been prepared as a core. and convey the impression of attempts at We equate the term "core" with "prepared intentional flake production. Once produced, core," and distinguish objects so labelled from a core could have been removed from the site "tested raw material." The latter refers or discarded there because of imperfections specifically to masses of naturally occurring that became apparent while it was being pro­ stone from which one or more flakes have duced or reduced. Presumably the cores been struck, not for use as tools, but to assess recovered represent those discarded because the quality of the stone. of poor-quality stone or those expended by Our distinction between cores and tested the detachment of flakes for use as tools or to raw material admittedly is not satisfactory be made into tools. with reference to ephemeral stone acquisition Tested raw material is stone that displays and use sites discussed above, at which pieces flake detachment scars but still exhibits of raw stone are picked up, flaked to a func­ substantial cortex. Examples of these artifacts tional edge, used, and discarded on site. This at SBr-5872 are not considered cores because situation, however, points to the need for their quality precluded the production of more sophisticated terminology as the study cores. They were discarded because they of lithic technology grows. could not be reduced into useful cores. The We reject the notion that cores often distinction obviously is one of degree, and no served as rough-service tools, such as chop­ operational definition can be formulated to pers or cleavers. Impact from use likely consistently distinguish tested raw material would have emplaced incipient cone fractures from cores. Some investigators consider "test­ in such cores, compromising the quality and ed raw material" synonymous with "incipient intended function of stone to supply flakes core" (J. Binning, personal communication useful for tools or blanks. 1989). We do not favor this usage because At SBr-5872, cores are small and are the term implies that the material actually was either generally bifacial or of multidirectional in the process of being reduced into a core. form (i.e., flakes were struck from almost any The material was not in the process of being available platform). Bifacial cores simply so reduced; it was partially reduced in an at­ have two faces; apparently no deliberate tempt to discover if it was of adequate quality attempt was made to standardize their form to warrant further reduction into a core. or size, and they should not be considered (One other alternative is that the material we 154 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY term "tested cobbles" represents clasts from debitage recovered, 64.8% is over 4 cm. in which flakes were struck to be used as expe­ maximum dimension. Specimens measuring dient tools. Had this been the case, the between 2 and 4 cm. amount to 33.4% of all situation would closely resemble that de­ debitage recovered (Table 2). The tendency scribed in Australia by Gould [1977], and we toward larger size in part reflects the fact that believe the amount of worked material at the no pressure flakes were recovered. site would be minor. We do not accept this Common debitage categories include (in interpretation because of the large amount of decreasing order of frequency) interior flakes, worked material at the site and the lack of secondary decortication flakes, and primary observable evidence of use-wear on the flakes. decortication flakes. Also, shatter and non- We note, however, that agencies such as classifiable flake fragments are well represent­ trampling can grossly alter surface assemblag­ ed. These debitage categories, together with es and lead to erroneous conclusions of tool the overall size of individual pieces, indicate use based on edge-wear studies [Flenniken or are expected in the course of early stages and Haggarty 1979].) in the testing and reduction of raw material. Aside from cores and a few formed arti­ A single edge-preparation flake suggests facts found at Locus 12, all specimens recov­ biface reduction. ered from SBr-5872 are debitage. Four are Tested Raw Material. Tested raw material glossy with a waxy luster, suggesting detach­ includes clasts and clast fragments from which ment from cores that had been heat-treated one or more flakes were detached, but which and presumably brought to the site as supplies were discarded without having been reduced of tool stone for daily use. All other debitage into cores (Fig. 3). (Larger boulders tested specimens lack the waxy luster that usually for quality were not collected.) Examples of results when siliceous stone is heat-treated tested raw material (30 specimens) represent (see below). They are flakes and shatter that the early-stage nature of the lithic industry in resulted from the testing of raw material and the main site area. the production of cores. The cores them­ Cores. Although seven of the 10 cores are selves likely would have been transported for small and of rough bifacial configuration (Fig. heat treatment to a place that offered more 4), there is no clear evidence of a preferred firewood. core form. The material would not consistent­ Loci 1-9, 11, 13-17. Table 1 lists the ly yield good bifacial cores, and multidirec­ characteristics of the flaked stone assemblage tional ones may have been obtained as often collected at each of the various site loci and as, or more often than, those of other at each of the shovel tests in the main site configurations. A small core remnant of area. chalcedony is lustrous and waxy, possibly from Debitage. The debitage analysis presented heat treatment. The remaining cores are of here employs, with minor revisions in nomen­ unstandardized form, most being multidirec­ clature and description, technological types tional (Fig. 5). Dimensions of these are given commonly employed in lithic technology in Table 3. (Crabtree 1982; Lithic Analysts MS). Cortex. As noted, a common occurrence Overall, the assemblages from the surface in the recovered assemblage is raw material loci and the shovel test units are similar in tested for quality by the detachment of one or that the same debitage types are represented more flakes (7.9% of the total assemblage). in approximately the same ratios. Of all Debitage categories support this idea by LITHIC RAW MATERIAL PROSPECTS 155

o o I ^ o a< e s 88

S « S S P

00 r^ Op ^ 00 ^ m rt •-I m •? IS ss m 00 Sea I I I I I 1 1 1 1

1 1 - 3 0 1 1

I I I n t-i I I rH f^ (^ 1 1 m f) D "1 I I I I I I I I I 1 1 r^ '^ 1 r» o

"1 r4 o O I I I - I I I I I 1 1 1 '^ 1 "O rq - I I 1 ^ 1 t^ rq CA D .J U i-i I es '- I (N v^ 1 1 1 S^ x u m fn ^ SO r- 00 1 1

^& I I I I rt.-, I I - 1

I I 1 o m 1 1 g<-

'^ I I I I ^ I I 1 <^ 1 1 1 or4

I I I I I I I I I 1 1 1 •0 i I •* ^ 1 •» Tf 1 1 p)00 Q

I I I I n n I I I I I 1 1 r-t 1 .-c VO .-1

I .-< e- ^s O £ £ 2 'Sk'S ra a I 8 & Z i/] 0L. Zio fl •? 2 "at! S2 u «• S i2 CQ III - Z i/5 BL- (/J 156 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Table 2 SIZE RANGE OF DEBITAGE FROM SBr-5872, MAIN SITE AREA"

Chert/Chalcedony Rhyolite Size Range (cm.) Sub­ Size Range (cm.) Sub­ Total Percent Locus <1 1-2 2-4 >4 total <1 1-2 2-4 >4 total Debitage of Total

1 _ _ 12 14 26 — — _ 1 1 27 8.0 2 - - 2 1 3 - 1 - 1 4 1.2 3 - - 9 16 25 - - - 3 3 28 8.3 5 - - 3 7 10 - - - - 0 10 3.0 6 — - 4 22 26 - — - - 0 26 7.7 7 — 3 23 29 55 - - - - 0 55 16.3 8 — — 1 13 14 — — — — 0 14 4.1 9 — — 1 — 1 - 1 8 9 10 3.0 11 — 1 30 51 82 - - - 5 5 87 25.7 13 - - 7 16 23 - - - 1 1 24 7.1 14 — — 3 14 17 — — - - 0 17 5.0 15 - - - - 0 - - - 1 1 1 0.3 17 — — 3 8 11 - - - - 0 11 3.3 STV 11'1 1 12 8 21 - - - - 0 21 6.2 STU B - 1 - 1 2 - 1 - 1 3 0.9

Total 0 6 110 200 316 0 0 3 19 22 338 100.0 Percent 0.0 1.8 323 59.2 93.5 0.0 0.0 0.9 5.6 6S 100.0

" Locus 4 and Locus 16 have been omitted because neither contained debitage. '' STU = Shovel Test Unit, 50 cm. s

Bifacial cores, SBr-5872. Left three specimens were rejected because of inferior quality. The right specimen is an expended core that appears to have been heat treated. Length of left specimen, 9.5 cm.

Fig. 5. Multidirectional cores, SBr-5872. Length of left specimen, 8 cm.

Table 3 weathering, occur in this area. The scatter DIMENSIONS OF CORES FROM SBr-5872, has been named the Juan obsidian source EXCLUSIVE OF LOCUS 12 area (Wilke and Schroth 1988b). Apparently Catalog Length Width Thick. Weight it was the source of local obsidian for many Number Artifact (cm.) (cm.) (cm.) (&) late prehistoric sites in the area of the Castle

138-2-1 Unstandardized core 6.14 6.07 3.02 108.1 Mountains. Many such sites have surface 138-4-1 Bifacial core 7.96 5.75 2.32 106.6 assemblages containing these small obsidian 138-7^ Bifacial core 9.38 5.92 3.04 158.1 clasts or the bipolarly reduced remnants of 138-7-7 Bifacial core 7.78 4.94 3.36 83.9 138-8-3 Bifacial core 6.14 5.16 3.11 83.3 them. The specimen found at SBr-5872 is 138-8-4 Bifacial core 4.89 5.03 2.05 53.6 believed to be a lost object. 138-14-2 Unstandardized core 6.98 4.72 3.20 84.4 Locus 12. Cultural remains from this 138-14-3 Unstandardized core 7.45 5.39 2.65 103.1 138-15-1 Bifacial core 9.55 5.91 3.73 170.0 locus (Fig. 2) include debitage, cores, formed 138-16-2 Bifacial core 12.42 9.67 3.64 403.6 flaked-stone artifacts, a hammerstone, block 158 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

millingstones, and ceramics. The assemblage measuring 11.8 x 10.6 x 3.5 cm., with battering suggests a seed-collection camp. on the edges (Fig. 7). Cobbles of this nature Debitage. Debitage types generally mirror are foreign to the local environment. those from the other site loci except for the Millin^tones. Large block metates, or absence of primary decortication flakes (Table millingstones, of basalt are represented by 4). Secondary decortication flakes are present fragments only. One specimen (Fig. 8) con­ in low frequency. This suggests that the locus sists of six fragments and originally measured contains products of behavior somewhat about 40 X 30 X 12 cm. with a nearly flat different from that represented elsewhere on milling surface. The remaining fragments the site. The only two readily evident biface- could represent as few as one or as many as thinning flakes found at the site occurred at three individual millingstones. No manos this locus. were found. Apparently, blocks of basalt, or In addition to local chert and chalcedony, prepared millingstones, were brought to the three flakes of red jasper and two of black site and left there; they were too cumbersome jaspagate, material not seen elsewhere at the to have been carried about the landscape on site, were recovered. The black jaspagate a daily basis. This suggests that the milling­ appears waxy and lustrous, suggesting heat stones were left at the site as site furniture treatment; the red jasper lacks these proper­ (Binford 1979) intended for use on subse­ ties, but may have been heat treated. (Even quent trips, and that repeated visits occurred. after heat treatment, red jasper from Lavic, We cannot account for the breakage of the near Ludlow, 100 km. to the southwest, often block millingstones, but it does not appear to fails to fracture with a shiny luster and a waxy be the result of natural weathering processes. texture.) Approximately the same percentages Ceramics. Fifty-four ceramic sherds of sizes of debitage pieces occur at Locus 12 probably came from two vessels. Most sherds as at the remainder of the site (Table 5). occurred in a cluster measuring about 3 x 4 m. Cores. One bifacial core of chalcedony AU but one appear to be from a bowl deco­ and one expended multidirectional core of rated on the interior with a black-on-gray rhyolite were recovered. Dimensions of the design. The exterior is reddish brown. Rim cores and other formed artifacts are given in profile and decoration indicate a bowl that Table 6. was straight-sided, neither incurvate nor (Mier Formed Flaked-stone Artifacts. excurvate to any large degree, of undeter­ Other formed artifacts include one preform mined height, and with an estimated diameter fragment, two undiagnostic biface fragments, of 40 cm. The rim sherds are somewhat and one complete unifacial tool or edge- flattened. modified flake of apparently heat-treated The sherds were examined by Margaret chalcedony (Fig. 6). The long pressure-flake Lyneis (personal communication 1988), and scars on the preform fragment suggest that the results of her analysis suggest that the the specimen had been subjected to heat vessel represented by most of the sherds was treatment. Pressure flaking of this kind manufactured in the general tradition of cannot be accomplished on raw stone from Lower Colorado Buff Ware, but the design is the area. reminiscent of the western Anasazi. Heavy Hammerstone. The only hammerstone black lines, finer lines, lines with pendant dots, found at the site came from Locus 12. It is a and a design that is not distributed evenly smooth, discoidal, quartzite stream cobble about the rim all reflect Black Mesa style LITHIC RAW MATERIAL PROSPECTS 159

Table 4 FLAKED STONE ARTIFACTS FROM SBr-5872, LOCUS 12

STU* STU Sub­ Percent Artifact Category and Type S urface 12a Ub total Total of Total

Secondary decortication flakes 13 19.4 Natural platform - 1 - 1 Single-facet platform 3 - - 3 Platform absent 8 - 1 9 Interior flakes 27 40.3 Natural platform 1 - - 1 Single-facet platform 10 - - 10 Multifacet platform 2 - - 2 Platform absent 14 - - 14 Biface flakes 2 3.0 Early stage percussion 1 - - 1 Late stage percussion 1 - - 1 Shatter 9 13.4 With cortex 2 - 1 3 Without cortex 2 2 2 6 Undiagnostic flake fragment 6 9.0 With cortex 1 - - 1 Without cortex 5 - - 5 Tested raw material 4 - - 4 4 6.0 Cores 2 3.0 Bifacial 1 - - 1 Multidirectional, expended 1 - - 1 Other formed artifacts 4 6.0 Preform fragments 1 - - 1 Undiagnostic biface fragment 2 - - 2 Complete unifacial tool 1 — — 1

Locus 12 Total 60 3 4 67 67 100.1 Percent of Total 89.6 4.5 6.0 100.1

STU = Shovel Test Unit, 50 cm. square.

Table 5 SIZE RANGE OF DEBITAGE FROM SBR-5872, LOCUS 12

Chert/Chalcedony Jasper/Jaspagate Size Range (cm.) Size Range (cm.) Total Percent <1 1-2 2-4 >4 Subtotal <1 1-2 2-4 >4 Subtotal Debitage of Total

Surface — - 20 25 45 - - 4 1 5 50 87.7 STU 12a — 1 2 - 3 - - - - 0 3 5.3 STU 12b - - 2 2 4 - - - - 0 4 7.0

Total 0 1 24 27 52 0 0 4 1 5 57 100.0 Percent 0.0 1.8 42.1 47.4 91.2 0.0 0.0 7.0 1.8 8.8 100.0

STU = Shovel Test Unit, 50 cm. square. 160 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 6. Formed artifacts from Locus 12, SBr-5872. Left, uniface; center, early-stage biface tool fragment; right, late- stage (pressure-flaked) biface fragment. Right specimen is of a material not found elsewhere at the site. All appear to have been heat treated. Length of right specimen, 30 mm.

Fig. 7. Hammerstone from Locus 12, SBr-5872. Maximum dimension, 11.8 cm.

(Fig. 9). Temper is coarse, angular feldspars date perhaps as much as 100 years later. and probably some quartz. Although the Lyneis suggested that some potter trained in sherds cannot be identified to specific type. the ceramic tradition of the Lower Colorado Black Mesa Black-on-White was thought by River may have been inspired by, and at­ Colton (1955) to date about A.D. 900-1000, tempted to mimic, western Anasazi pottery, and Ambler (1985) assigned it a beginning perhaps with imperfect results because good LITHIC RAW MATERIAL PROSPECTS 161

Table 6 Comparison of Flaked Stone Assemblages DIMENSIONS OF FORMED ARTIFACTS FROM SBr-5872, LOCUS U Comparison of the flaked stone artifact assemblage from the main site area with that Catalog Length WidthThicknessWeight Number Artifact (cm.) (cm.) (cm.) (g.) from Locus 12 reveals several important differences. Locus 12 yielded no primary 138-12-2 Preform fragment >2.71 2il OJl 4.7 decortication flakes and a lower percentage of 138-12-t Hammerstone 11.72 10.61 3-51 540.0 138-12-6 Multidirectional core, 3.46 3.29 1.68 18.5 secondary decortication flakes. Although the expended numbers are small. Locus 12 yielded greater 138-12-8 Bifacial core 7.36 5.70 1.81 90.3 percentages of biface-reduction flakes,interio r 138-12-10 Biface fragment, >2S6 3J0 0.97 8.9 unfmished flakes, and shatter, as well as the only formed 138-12-11 Biface fragment, >2.34 3.21 0.74 4.8 flaked stone artifacts (Fig. 10). These differ­ unflnished ences suggest that activities on the main site 138-12-17 Uniface 4.33 336 0.65 10.0 area were directed more toward stone acqui­ sition and early-stage reduction, and that those at Locus 12 were directed more toward core reduction and tool production, in associ­ ation with subsistence activities. Heat-treatment Experiments Because of the tough nature of the chert and chalcedony at SBr-5872, heat treatment would have been necessary to render the ma­ terial more readily workable by pressure flaking. Prior experience in heat treatment of cherts and chalcedonies indicated that slowly increasing the temperature to about 240 ° C. over the course of 10-15 hours, and then slow­ ly cooling it, accomplishes the desired end. Fig. 8. Fragments of block millingstones at Locus 12, SBr-5872. Original dimensions of right speci­ Exactly what happens to stone during heat men, approximately 40 x 30 x 12 cm. treatment is not known, but the fracture quali­ ties are markedly altered (Mandeville 1973). clay was not available. Anasazi inspiration for The stone loses weight as interstitial water is some of the ceramic designs in the northern driven off, myriads of tiny fractures are be­ Lower Colorado area is suggested by designs lieved to be created in the crystal latticework on the type Parker Red-on-Buff. A suggested or in the fibrous intercrystalline matrix, and date for this type is from before A.D. 900 to the strength and toughness of the stone are after A.D. 1900 (Schroeder 1958). substantially reduced. Following heat treat­ An additional undecorated sherd was buff ment, the stone fractures far more readily, on the exterior and black on the interior, and and fresh fracture scars display a waxy texture displays a broken and then weakly ground and and a shiny luster (Rick and Chappell 1983). smoothed edge. It is suggested that a large Experiments in replicative heat treatment jar was broken, and that a large sherd of that and flintknapping were undertaken to better jar served as some form of open container, assess the likelihood that the material could which was then broken. have been used as tool stone in antiquity. For 162 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 9. Decorated ceramic sherds (interior surface) from Locus 12, SBr-5872. Actual size. these experiments, unworked clasts of raw temperature gauge. Stone selected for the stone were collected from the surface of the experiments was placed in a sand bath in an site. In addition, material from the initially iron box. Although it is not possible to recorded Locus 10 (later determined to have achieve good temperature control under a been fractured by natural processes) was used campfire, experience has shown that heat in the experiments. treatment often can be accomplished by The heat treatment was conducted in a burying the stone 10-15 cm. in dry soil and ceramic enameling kiln. Temperature was maintaining a small fire over it for 10-15 controlled with a variable autotransformer and hours. This probably is the way stone was monitored with a thermocouple attached to a heat treated in antiquity (Mandeville 1973). LTTHIC RAW MATERIAL PROSPECTS 163

Flaked stone artifact type

Primary decortication flakes Secondary decortication flakes Interior flakes Biface reduction flakes Shatter Flake fragments Tested raw material Cores Formed artifacts 0 10 20 30 40 50 Percent of flaked stone artifacts

Main site area Locus 12 Fig. 10. Comparison of flaked stone assemblages from the main site area and from Locus 12. If the Stone is heated excessively or too from the heat-treated stone from SBr-5872 rapidly, it is ruined. If not heated enough, it are shown in Figure 12. The figure also can be reheated to a higher temperature, shows one example that failed; not all of the often with satisfactory results. Although there stone improved substantially in quality as a is little timber in the area, creosote bush result of heat treatment. {Larrea tridentata), catclaw {Acacia greggii), The results of this replicative exercise and California juniper {Juniperus califomica, suggest that, for use in tool production that common 3 km. to the north) could have pro­ involved pressure flaking, all of the raw chert vided the necessary fuel for heat treatment. and chalcedony occurring naturally at SBr- Because of the amount of fuel needed to 5872 probably required heat treatment. The sustain a heat-treatment fire for a period of luster and texture of only several specimens hours, stone likely would have been transport­ recovered at the site match the properties of ed to a place where fuel was available. the experimentally heat-treated stone. The Most samples of chert and chalcedony few examples that display these qualities subjected to heat treatment showed pro­ suggest aboriginal heat treatment, but the nounced improvement in flaking quality. process may have taken place elsewhere, with Some (containing more impurities) showed a core of prepared tool stone having been only minor improvement. Figure 11 shows brought to the site already heat-treated. No samples of raw and heat-treated stone, and features indicative of fires or hearths were the increased luster and waxy texture of the noted anywhere at the site. latter are clearly evident. Whereas it was all A final observation is that the presence of but impossible to pressure flake the raw stone, crystal pockets within Mojave Desert crude much of it is readily worked after heat quartzes will sometimes cause the stone to treatment. Examples of replicated tools made explode during heat treatment. One piece of 164 JOURNAL OF CALIFORNLA. AND GREAT BASIN ANTHROPOLOGY

Fig. 11. Raw and heat-treated stone from SBr-5872. Upper row, raw material; lower row, heat-treated material from same clast. Length of lower left specimen, 55 mm. chalcedony with a crystal pocket inclusion did at SBr-5872. These can be discussed with explode during the heat-treatment experi­ respect to what occurred at the site and how ments (Fig. 13). Such flaws are common in those inferred activities may have related to the chert and chalcedony from SBr-5872. activities at the other sites in the region. Examples of cores and tested raw material Data presented above suggest that the from the site show that stone sometimes was original identification of the site as a lithic discarded after discovery of such crystal raw material prospect was correct. All of the pockets when the material had been substan­ evidence from loci 1-9, 11, and 13-17 suggests tially reduced and was nearly ready for nonintensive but probably repeated prospect­ transport from the site. ing for usable tool stone. That the naturally occurring stone was not intensively quarried Interpretations is evident by the sparseness of the surface The data permit interpretations regarding scatter of quarry detritus. That some of the the kinds of prehistoric behavior represented material was of tool quality is evident by the LITHIC RAW MATERIAL PROSPECTS 165

port activities of persons whose primary in­ tention was prospecting for tool stone at the remainder of SBr-5872; or (2) the remainder of SBr-5872 represents ancillary activities by persons drawn for some particular reason to Locus 12. The first of these alternatives would sug­ gest that the possibility of obtaining quality tool stone brought persons to SBr-5872 in prehistory. This alternative would be sup­ ported if no outstanding tool stone source Fig. 12. Experimentallyknapped, heat-treated stone from areas occurred in the immediate region. This SBr-5872. Left, failed dart point preform. The is not the case; several extensively used flaking characteristics of the stone improved aboriginal stone quarries (SBr-5705, -5706, little as a result of heat treatment. Right, three arrow points. The left two are of stone that and -5707) occur about 1.5 km. to the became waxy and lustrous as a result of heat northwest (Fig. 1). All contain tool stone of treatment; the right si>ecimen was not noticeably higher quality and in greater abundance than improved. Length of right specimen, 22 mm. occurs at SBr-5872. It seems highly unlikely, surface assemblage, which includes homoge­ given the proximity to these quarries, that neous chalcedony, and by the results of the anyone in the area seeking tool stone would heat-treatment experiments. Sharp flakes have made special trips to SBr-5872. There­ detached from naturally occurring clasts of fore, this alternative is not supported. raw chert and chalcedony could have been The second alternative is more strongly used to perform a variety of cutting and supported. The evidence at Locus 12 suggests scraping tasks. Such stone cannot, however, that, at some time in the past, persons carried be pressure flaked easily into formed tools large blocks of basalt suitable for use as such as projectile points. Following heat millingstones, or already-made large block treatment to about 240 ° C, most of the stone millingstones, to Locus 12. The millingstones can be pressure flaked easily; other pieces are were left there as site furniture (Binford 1979) not improved by heating. for use whenever people reoccupied the im­ The observations suggest that the limited mediate area. Here they engaged in the volume of quality stone available at the site gathering and milling of some seed resource, precluded the development of an extensive perhaps galleta grass {Hilaria rigida), which quarry. The age of the prospecting activity in grows abundantly just across the drainage to the main site area cannot be determined from the east. If the present distribution of this evidence now available, but it is believed to grass reflects conditions in the past, people date from late prehistoric time. occupying Locus 12 may have collected it. The data from Locus 12 provide some in­ Ethnobotanical information on use of galleta formation on site function and chronology. If grass is almost nonexistent, but one reference the activities represented there are con­ (Weight 1978) suggests that the seeds were temporaneous with those at the remainder of collected and eaten by Indians of the Mojave the site, these data suggest one of two Desert. Given that few other natural seed alternatives: crops occur in any abundance in the region, (1) the locus represents the base of sup­ galleta grass, even if less esteemed, may have 166 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 13. Chalcedony from SBr-5872 destroyed by differential expansion through a crystal pocket during the heat- treatment experiment. Book-matched split; maximum dimension, 6 cm. been collected and ground. Other economi­ (Steward 1938). Persons attached to the cally important species may have been com­ group may have prospected the immediate mon in the past. area for tool stone on a nonintensive basis. The persons that used Locus 12 had The remaining site loci may represent the ceramic containers, at least two of which were residue of this prospecting activity. If this broken there. The ceramic artifacts date at interpretation is correct, the prospecting is least some of the activity at Locus 12 to some best seen as an embedded strategy (cf. Bin­ portion of the last 1,000 years. Occupants of ford 1979) within a direct strategy focused on the locus had heat-treated chalcedony, a seed gathering. All the activity represented at preform of which broke there, and they SBr-5872 may have occurred within a series detached flakes from an apparently heat- of short visits that individually lasted no more treated piece of black jaspagate and from a than one to several days. probably raw piece of red jasper. The latter It probably was necessary to transport materials were not found elsewhere on the water to the site each time it was occupied. site, but had to have been brought there. The activity represented at the site ultimately The seed collecting probably reflects the must have been tethered to one of the very activities of women, to judge from the sexual few water sources in the area, such as Hart division of labor known historically to have Spring, located 4.5 km. to the northeast. existed in arid and semiarid western America Abundant sites occur on the east side of the LITHIC RAW MATERIAL PROSPECTS 167

Castle Mountains not far from this water exposed crude quartz searching for traces of source and appear to be linked to it. Persons gold. Thus, not aU of the broken and spalled using SBr-5872 may have resided for brief material at the site can be attributed to periods at some of these sites. aboriginal stone prospectors. More intensive quarrying of stone at One locus of the site, however, contains nearby quarry sites such as SBr-5705, -5706, evidence of attempts to dislodge a clear in­ and -5707 might best be viewed as a direct clusion or "eye" of blue chalcedony about 15 strategy that involved no use of SBr-5872. cm. in diameter. Part of this outstanding Because of the greater abundance of stone piece of stone is exposed in a clean, naturally and its higher quality, persons seeking quality fractured face of the vein; the remaining por­ tool stone may have made trips to these tion weathered away in geologic time. Indians quarries specifically to procure it. We would, prospecting the vein for quality tool stone therefore, expect some kinds of support ac­ found the inclusion and went to considerable tivities to be present at these sites. This effort to dislodge it. They brought to the site situation would contrast strongly with that a large, elongate, basalt cobble. Using this interpreted here for SBr-5872. cobble as a hammerstone, they pounded away on the surrounding chert matrix in an attempt OTHER EXAMPLES OF PROSPECTS to free the inclusion. All attempts failed, as Other sites in the immediate region doc­ did the hammerstone, and the frustrated ef­ ument patterns of lithic raw material pros­ fort is told clearly by the evidence that pecting in prehistory. The sites vary in terms remains (Fig. 14). A broken ceramic jar of a of behavioral context, yield of useful tool gritty buff ware with a patchy, red, exterior stone, and overall scale. They permit a wash, probably a water container, is represent­ broader understanding of the range of sites ed by a sherd scatter under a large Mojave represented by prospects. yucca 25 m. away. No other artifacts were seen at the site. It would appear that in Blue Eye (CA-SBr-5861) whatever agenda the aborigines were involved Blue Eye is the name we have given to a at the time, considerable effort was expended site along the west base of the Castle in an unsuccessful attempt to obtain an out­ Mountains about 3.5 km. north of SBr-5872. standing piece of stone. The site occurs at an exposed, tilted vein of St. Joe American (CA-SBr-5090) chert and chalcedony, probably precipitated by silicic geothermal waters in a fissure in The Hackberry Mountains are a small volcanic rock that has since weathered away. Tertiary volcanic range about 30 km. south­ The vein is less than a meter in thickness and west of the Castle Mountains. An exposed is exposed on the surface for a linear distance outcrop of gray chert-like material at the St. of perhaps 150 m. The material is of varying Joe American site in a small pass in this range quality but generally rather poor, compro­ was prospected in aboriginal times for tool mised by numerous weathering cracks and stone, but only minor amounts were ever re­ probably a relatively low SiOj content. moved, to judge from the surface scatter of Interpretation of the site is made difficult by debris. Much of this material displayed the fact that a road has been bladed through cortex, and there were no formed artifacts. the exposure, and modern prospectors appar­ Physically, the material appeared to be a ently hammered away here and there at the slightly grainy chert, and samples were col- 168 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Fig. 14. Broken basalt hammerstone next to an inclusion of blue chalcedony (arrow) in chert vein at the Blue Eye site (SBr-5861). The hammerstone is a weathered cobble of material foreign to the locality. It broke during aboriginal attempts to break down the surrounding matrix and free the inclusion. lected and taken to the laboratory for heat- has been alleged by some (e.g., Simpson 1958) treatment experiments. The flakeability of the to represent a Pleistocene lithic industry. material was not improved by heat treatment Attention was first called to the evidence at at 250 ° C. Possibly similar experiments were sites in the area of the Mojave River east of run in antiquity with similar results, and for Barstow, around the shores of ancient lakes that reason the exposure never was exploited Troy and Coyote (together comprising the to any significant degree (Pinto 1987). The Manix Lake Basin). A presumed association site has since been destroyed by construction with Pleistocene Lake Manix, and abundant of a mining road. broken or otherwise discarded bifaces, called handaxes by some, led to definition of the Manix Lake Lithic Industry Sites industry. The sites also contain abundant A widespread occurrence of quarry and tested cobbles of raw chert, jasper, and chal­ prospect detritus in the central Mojave Desert cedony, and extensive scatters of debitage. LITHIC RAW MATERIAL PROSPECTS 169

Examination of tested raw material clasts and shock or otherwise unsuccessful knapping discarded bifaces clearly reveals the imperfec­ efforts. tions or failure breaks that led to their aban­ Third, there is no compelling reason not donment. Pieces of debitage generally are of to conclude that these various stone sources large size, and primary and secondary decorti­ have been used throughout the entire period cation flakes are both abundant and typical. of human occupation of western North Weathering cracks, crystal pockets, and vugs America, including the last century. occur throughout much of the material. Fourth, and most significant, a very There is no question that some of the widespread occurrence of raw material important sites attributed to the Manix Lake prospecting is indicated throughout the entire Lithic Industry are actual quarries; they central Mojave Desert region. Major quarries contain considerable good-quality material. grade imperceptibly into ubiquitous minor raw The surface assemblages at some sites also material prospects with no objectifiable document extensive quarrying activity in change in the nature of the lithic detritus. At prehistory, but the activities of recent the latter sites, overall poor quality and rockhounds make it impossible to determine inability to consistently obtain large pieces of if actual quarry pits are present at most of good stone are clearly evident. Thus, much of them. We think subsurface quarrying general­ the allegedly very ancient Manix Lake Lithic ly was not carried out at these sites, but that Industry would appear attributable to raw material was obtained from the surface. material prospecting throughout the full range Known quarries in the vicinity of Manix Lake of prehistory, whatever that may entail. include, among others, those at Hector Hills, Desert Oasis, Agate HUl, East Rim, and Site 26-CK-2375, Near Lake Mead Toomey Hills. Kamp and Whitaker (1986) discussed what Credibility of the Manix Lake Lithic In­ they termed "unproductive lithic resources" dustry as a record of Pleistocene activity based on an analysis of a surface assemblage around now-extinct lakes can be questioned from the Nevada side of Lake Mead. They on the basis of dating applications (Dorn et interpreted site 26-CK-2375 as a procurement al. 1986, 1987; Bamforth and Dorn 1988). It area where Virgin Branch Anasazi peoples is marred also by several technological consid­ (and possibly earlier groups) gathered tool erations. stone. Based on refitting of flakes to nodules First, aboriginal use of the lithic deposits from which they came, the authors concluded occurred wherever the Miocene marine de­ that usable flakes and occasional cores of posits outcrop in the Mojave Desert, including chalcedony were produced and transported the Jasper Hill and Lavic jasper outcrops, the elsewhere. We believe that the archaeological Sidewinder chalcedony quarry, and extensive situation described by Kamp and Whitaker chalcedony quarries in the Kramer Hills, all closely parallels that described here. How­ localities far distant from Manix Lake. ever, we question their interpretation of the Second, the alleged handaxes are inter- patterning of the behavior implied by the pretable as bifacial cores, or attempts to make surface assemblages. Although no support such cores, and those that occur at quarries facilities nor associated base camps were can be seen to have been abandoned due to reported (they may now be submerged under the presence of checks, vugs, or crystal the nearby waters of Lake Mead), quality pockets, or due to breakage caused by end- stone was rare and probably never was ob- 170 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY tained in quantity. No quarry pits or ham- carved ornaments (Sigstad 1973). The impor­ merstones, and very few formed artifacts were tant property about all quarries is that they found. This suggests that the area was not an offered desired material of a quality and in attractive and reliable place to go to obtain quantities that warranted its exploitation. quality tool stone, but that the assemblage Behavioral Aspects. Quarries are places may instead represent the cumulative result of where people obtained raw stone in quantity, many isolated prospecting episodes embedded presumably on a fairly consistent basis. At within some subsistence-related activity. major quarries, the quality and quantity of Further work emphasizing experimental re­ desired material often warranted intentional duction and heat treatment of stone from this trips to procure it as a direct procurement area may provide information on the behav­ strategy, although stone also may have been ioral context responsible for the assemblage. obtained from quarries as an embedded pro­ curement strategy in conjunction with other DISCUSSION activities, such as hunting, gathering, or travel In the foregoing section, we have pre­ (Binford 1979; Holen 1983). Some quarries sented information and interpretations sug­ were the beginning points of vigorous indus­ gesting that SBr-5872 is a lithic raw material tries that involved the long-distance transport prospect. How do such sites compare behav- and trade of desired commodities. The in­ iorally and archaeologically with quarries tensive activity represented at quarries where where stone was obtained consistently and in people procured stone for flaking can be seen quantity, and with ephemeral stone acquisition as the beginning of a lithic technological and use sites where stone simply was picked continuum. Transport, processing (including up, flaked to a useful configuration, used, and heat treatment where appropriate and neces­ discarded on site? To answer this question, sary), and caching of stone can be seen as it is necessary to characterize the nature and intermediate activities in this continuum. extent of activity represented by these site Archaeological Aspects. Raw stone may types. It also is necessary to identify the role have been transported from quarries, but of each in the prehistoric lithic technology and generally it was spalled or sectioned into the way the activity represented at each site quarry blanks or further reduced into cores at type was integrated into the overall pattern of such sites to ensure quality and decrease aboriginal use of landscapes. weight. Core production permitted the quality control necessary to ensure maximum Quarries yield of useful material. An exception would Most often, quarries are reckoned as be removal of small clasts of obsidian from places that yielded stone for flaking (Bryan areas where this material occurs naturally. 1950). Other kinds of quarries, however, Obsidian requires no heat treatment, and yielded soapstone or steatite for containers, reduction of such pebbles most often required arrowshaft straighteners, and ornaments bipolar sectioning. Quarries typically display (Holmes 1890); turquoise for ornaments the following: (1) large quantities of raw (not (Leonard and Drover 1980); blocks or slabs heat-treated) debitage, much of it of large size of andesite, basalt, or other material for and with cortical surfaces; (2) cores broken in millingstones and handstones (Huckell 1986); the course of production; (3) hammerstones sandstone for abrading tools (Flenniken and used in reduction of raw material; (4) pits dug Ozbun 1988); and pipestone for pipes and to obtain stone not damaged by weathering; LITHIC RAW MATERIAL PROSPECTS 171 and (5) support facilities such as associated reduction of individual clasts reveals attempts camp areas and caches. Minor quarries to eliminate weather checks, vugs, crystal characterized by stone of lesser quality grade pockets, and other imperfections. Heat- imperceptibly into prospects. treated stone is rare or absent. Prospects generally lack quarry pits, abundant hammer- Lithic Raw Material Prospects stones, heat-treatment hearths, or support Prospects, or assay sites, occur any place facilities. We do not believe that support material normally quarried elsewhere was facilities typify prospects. Subsistence-related occasionally obtained. The distinction be­ base camps, to which prospects may be linked, tween these sites and quarries is one of scale; should contain assemblages reflecting the prospects never became quarries because of acquisition, processing, or storage of resources limitations in the quality and quantity of other than stone, but such assemblages are material available. not expected at isolated prospect sites. Behavioral Aspects. Prospects are places Ephemeral Stone Acquisition and Use Sites that were visited by people following a strategy that we believe was in most cases Our ideas on sites of this type are based fully embedded within some direct procure­ on Gould's (1977) characterization of ment strategy; it was this direct strategy that Australian lithic technology discussed at the actually determined the positioning of people beginning of this paper. on the landscape. We expect that raw Behavioral Aspects. Obtaining expedient material prospects often may be recognizably stone tools at ephemeral stone acquisition and linked to some other site in the immediate use sites was casual and dictated solely by the area that reflects a direct procurement needs of the moment. It was not necessarily strategy focused on some important resource, redundant, and it always occurred in the con­ such as a seasonally available seed crop. It text of other activities. Stone used in this may be difficult to identify the primary context was quickly flaked to the desired edge, exploitative strategy of persons or groups used as a tool, and discarded on site. responsible for raw material prospects if Archaeological Aspects. Sites of this type associated camps or exploitation sites or loci are likely to be difficult to characterize cannot be found. without a clear impression of the behavior Intentional stone-gathering trips seldom responsible for them. They will occur where were made to prospect sites because the at least some flakeable stone is present, quality and quantity of raw material did not coupled with where people happened to be in warrant such effort. A probable exception the course of their daily activities. Isolated, discussed above is the Blue Eye site, where expedient tools occur here and there, but the deliberate but unsuccessful attempts were spatial patterning of these artifacts reflects made to extract a single piece of outstanding use of the tools rather than acquisition of chalcedony from an exposed vein of chert. stone for them. Stone obtained on site and Raw material prospects may document single displaying evidence of use as tools should or multiple events. show no evidence of heat treatment. Archaeological Aspects. Quantities of raw CONCLUDING REMARKS waste stone litter the surface of prospect sites to one degree or another. Most of this In areas where high-quality tool stone material is of poor quality. Substantial occurs in abundance, lithic raw material 172 JOURNAL OF CALIFORNLA. AND GREAT BASIN ANTHROPOLOGY prospects are expected to be rare. But in from detached flake blanks, tool resharpening, areas where sources of quality tool stone are and ultimate discard of expended tools also widely dispersed, which includes most of the differ behaviorally and archaeologically. The western United States, raw material prospects commonly applied term "lithic scatter" fails to actually may be the most common site type differentiate any of these site or assemblage documenting lithic procurement and may have types. Proper characterization of lithic tech­ provided much of the tool stone used in nology and site function to explain prehistoric antiquity. behavior requires a more sophisticated analyt­ Only through careful characterization of ical approach to Uthic assemblages than the archaeological assemblages at sites repre­ archaeologists have traditionally employed. senting prehistoric lithic industries can the site NOTES types discussed here ever be identified, differ­ entiated, and functionally interpreted. Sim­ 1. The datum point was established with refer­ ence to a black PVC pipe marking the SE corner of plistic approaches to lithic analysis, such as mining claim Roy 244 and the SW corner of claim that of Sullivan and Rozen (1985), and which Roy 249. From this corner, the datum point is 74° are growing in popularity, add nothing to our east of true north at a distance of 40 m. 2. Our use of the term "split cone" should not understanding of past behavior. Sullivan and be confused with "sheared cone," which describes Rozen's approach, which employs debitage a typical failure of the mass of a pebble during categories that are "interpretation-free," fails bipolar reduction. We use the term "split" to even to recognize the difference between per­ describe cones of force that are "split" dorso- ventrally from the platform toward the distal flake cussion and pressure flakes, between raw and termination. This condition seems to occur when heat-treated stone, or between initial stone the platform is struck too hard, or with too great a acquisition and core prdduction and subse­ velocity, and especially when the flake platform is quent core reduction and stone expenditure. relatively broad. The exact nature of the fracture that causes the split is not known, but one of us It cannot distinguish reduction strategies (PJW) believes it occurs only on rather broad based on different core forms. It can flakes. It is believed to occur in the course of flake characterize certain gross morphological detachment when the center of the bulb is bent differences between assemblages, but it cannot away from the core before the fracture expands lat­ erally to the edges of the bulb area. While this characterize the human behavior responsible situation describes a bending of the flake, it results for assemblages or sites. Instead, it creates a in a perverse fracture that "splits" the cone of force. false impression that lithic analysis is easily The fact that spUt cones, or whatever they are termed, do not appear to have been specifically accomplished by persons not trained in lithic characterized in literature reflects the elementary technology, which it is not. nature of the science of lithic technology. Our discussion of prospects has gone into considerable detail, but this is necessary for a ACKNOWLEDGEMENTS proper characterization of such sites and their This paper has benefited from the continuing role in prehistory. We have tried to show that help of J. Jeffrey Flenniken £md from the comments of anonymous reviewers. We thank Margaret lithic raw material prospects, quarries, and Lyneis for her examination and analysis of ceramic ephemeral stone acquisition and use sites, aU sherds, and Terry Ozbun and Joan Schneider for of which represent early stages in lithic tech­ helpful comments and suggestions on a draft. nology, are very different behaviorally and Jeaime Binning also contributed signifiomtly through helpful discussions, reading and comment­ archaeologically. Other lithic reduction sites ing on a draft, sharing views on concepts and documenting core production, core reduction terminology, and making a number of road trips through flake detachment, tool production with the senior author to numerous Mojave Desert LITHIC RAW MATERL\L PROSPECTS 173 quarries. Carl Hansen clarified certain aspects of carbon Dating of Rock Varnish on geology, and Harold Linder generously shared his Mojave Artifacts and Landforms. extensive knowledge of the geology of the Castle Science 231:830-833. Mountains. Robert Hicks prepared the photo­ Dorn, R., D. Tanner, B. Turrin, and J. Dohrenwend graphs. The work reported here was sponsored by 1987 Cation-Ratio Dating of Quaternary Viceroy Gold Corporation, of which we especially Materials in the East-Central Mojave thank J. Christopher Mitchell. Fieldwork was authorized under Federal Antiquities Permit No. Desert, California. Physical Geography CA-16058. An intramural research grant from the 8:72-81. Academic Senate, University of California, Flenniken, J. J., and J. C. Haggarty Riverside, to the senior author partially defrayed 1979 Trampling as an Agency in the Forma­ research and manuscript costs. An earlier version tion of Edge Damage: An Experiment of this paper was presented 10 December 1988 at in Lithic Technology. Northwest the Third Kelso Conference on Mojave Desert Anthropological Research Notes 13:21- Prehistory, at Nipton, California. 27. Flenniken, J. J., and T. L. Ozbun REFERENCES 1988 Experimental Analysis of Plains Grooved Abraders. Plains Anthropologist 33:37- Ambler, J. R. 52. 1985 Northern Kayenta Ceramic Chronology. In: Archaeological Investigations Near Gould, R. A. Rainbow City, Navajo Mountain, Utah, 1977 Ethno-archaeology; or. Where Do Models Come From? A Closer Look at P. R. Geib, J. R. Ambler, and M. M. Australian Aboriginal Lithic Technology. Callahan, eds., pp. 28-68. Northern In: Stone Tools as Cultural Markers: Arizona University Archaeological Change, Evolution, and Complexity, R. Report No. 576. V. S. Wright, ed., pp. 162-168. Canber­ Bamforth, D. B., and R. I. Dorn ra: Austridian Institute of Aboriginal 1988 On the Nature and Antiquity of the Studies, Prehistory and Material Culture Manix Lake Industry. Journal of Series No. 12. California and Great Basin Anthropol­ Holen, S. R. ogy 10:209-226. 1983 Lower Loup Lithic Procurement Strategy Binford, L. R. at the Gray Site, 25CX1. Master's 1979 Organization and Formation Processes: thesis. University of Nebraska, Lincoln. Looking at Curated Technologies. Hohnes, W. H. Journal of Anthropological Research 1890 Excavations in an Ancient Soapstone 35:255-273. Quarry in the District of Columbia. Bryan, K. American Anthropologist (o.s.) 3:321- 1950 Flint Qu£U'ries-The Sources of Tools, 330. £md at the Same Time, the Factories of HuckeU, B. B. the American Indian. Harvard Univer­ 1986 A Ground Stone Implement Quarry on sity Peabody Museum of American Ar­ the Lower Colorado River. Phoenix: chaeology and Ethnology Papers 17(3). Bureau of Land Management Cultural Colton, H. S. Resource Series Monograph No. 3. 1955 Pottery Types of the Southwest: Wares Kamp, K. A., and J. C. Whitaker 8A 8B, 9A, 9B. Museum of Northern 1986 Unproductive Lithic Resources at Lake Arizona Ceramic Series No. 3. Mead. American Antiquity 51:383-388. Crabtree, D. E. Kelly, I. T., and C. S. Fowler 1982 Introduction to Flintworking (second 1986 Southern Paiute. In: Handbook of edition). Pocatello: Idaho State North American Indians, Vol. 11, Great University Press. Basin, W. L. d'Azevedo, ed., pp. 368-397. Dorn, R., D. Bamforth, T. Cahill, J. Dohrenwend, Washington: Smithsonian Institution. B. Turrin, D. Donahue, A. JuU, A. Long, M. Macko, Laird, C. E. Weil, D. Whitley, and T. Zabel 1976 The Chemehuevis. Banning: Malki 1986 Cation-Ratio and Accelerator Radio­ Museum Press. 174 JOURNAL OF CALIFORNIA AND GREAT BASIN ANTHROPOLOGY

Leonard III, N. N., and C. E. Drover Simpson, R. D. 1980 Prehistoric Turquoise Mining in the 1958 The Manix Lake Archaeological Survey. Halloran Springs District, San Bernar­ The Masterkey 32:4-10. dino County, California. Journal of Steward, J. H. California and Great Basin Anthropolo­ 1938 Basin-Plateau Aboriginal Sociopolitical gy 2:245-256. Groups. Washington: Bureau of Linder, H. American Ethnology Bulletin No. 120. 1989 Castle Mountains Gold Deposits, Hart Sullivan, A. P., Ill, and K. C. Rozen Mining District. California Geology 1985 Debitage Analysis and Archaeological 42(6):134-140, 143. Interpretation. American Antiquity Lithic Analysts 50:755-779. MS Glossary of Lithic Technolo^cal Terms. Warren, C. N. MS on fde at Lithic Analysts, Pullman, 1984 The Desert Region. In: California WA. Archaeology, by M. J. Moratto, pp. 339- Mandeville, M. D. 430. Orlando: Academic Press, 1973 A Consideration of the Thermal Pre- Weight, L. treatment of Chert. Plains Anthro­ 1978 The Spaniards Called it Galleta. Desert pologist 18:177-202. Magazine 41(5):28-31. Pmto, D. G. Wilke, P. J., and J. J. Fleimiken 1987 Archaeological Assessment of Minerals 1988 Experimental Reduction of a Bifacial Exploration Drill Sites m the Hackberry Core. Paper presented at the biennial Mountains, San Bernardino County, meeting of the Great Basin Anthropo­ California. Riverside: University of logical Conference, Park City, Utah. California Archaeological Research Unit Report No. 912. Report on file at WUke, P. J., and A. Schroth the Archaeological Information Center, 1988a CA-SBr-5872: A Prehistoric Lithic Raw San Bernardino County Museum, Red- Material Prospect in the Castle Moun­ lands. tains, San Bernardino County, California. Riverside: University of California Rick, J. W., and S. Chappell Archaeological Research Unit Project 1983 Thermal Alteration of Silica Materials No. 931. Report on fde at the Archaeo­ in Technological and Functional Per­ logical Information Center, San Bernar­ spective. Lithic Technology 12:69-80. dino County Museum, Redlands. Schroeder, A. H. 1988b A Treatment Plan for Significant 1958 Lower Colorado Buff Ware: A De­ Cultural Resource Sites to be Affected scriptive Revision. In: Pottery Types of by the Castle Mountain Project of the Southwest, H. S. Colton, ed. (un- Viceroy Gold Corporation. Riverside: paginated). Museum of Northern Ari­ University of California Archaeological zona Ceramic Series 3D. Research Unit Project No. 931. Report Sigstad, J. S. on fde at the Archaeological Information 1973 The Age and Distribution of Catlinite Center, San Bernardino County Muse­ and Red Pipestone. Ph.D. dissertation. um, Redlands. University of Missouri, Columbia.