7 I CREEK o 3000ft. 0 M. 1' ARCHAEOLOGICAL SITE GROUPS WARM SPRINGS DAM PROJECT Map 1. 9 SITE DESCRIPTION The foregoing geographic and archaeological information we will use in the following pages in the organization and analysis of archaeological data recovered in survey and testing in the Warm Springs Dam project area. For convenience of presentation the area is divided into 7 site groups for convenience of presentation (Map 1). In some ways the 4 northern groups stand as a unit as opposed to the two southern groups on Warm Springs Creek. Fran this point of view the Central Group is a no-mans-laid -- almost no sites of significance occur in most of it while in the eastern part what were once probably important sites have been virtually destroyed by agriculture. Beginning with the Western Group we describe these in order down Dry Creek arn up Warm Springs Creek. Twenty-one maps of individual sites are contained in the following descriptions. It is not possible to show many features on these maps other than symbolically. The symbols are explained in Figure 2. Western Site Group Dry Creek northwest of the mouth of Warm Springs Creek flows through a broad flat valley nowadays well adapted to the cultivation of grapes. Abcwe this the valley narrows progressively until in the Western Group the sides drop off on quite a steep slope with only a narrow terrace above flood level suitable for habitation. This means that there was very little land suitable for a large compact village - any large village would have to have been strung out along the terrace. It turns out that none of the sites has much area, all being snall and evidently specialized. Thus we seen to have an area peripheral to the main activity centers but one which, as we shall see, was of considerable importance. All sites in this Group are on the north side of the creek (except Son-566 which is on a tributary). We see two possible reasons for this. One is the usual one that a north slope is warmer in the winter. The other relates to the phenomena of slumping or land slippage which is camon along this section of the creek and occurs only on the south slope. The occupants of the valley may have avoided the south slope to prevent the slippage coming down on their villages or, alternatively, there may be villages now buried under the talus. It might be thought that the slippage is due to erosion produced subsequent to logging ard in part this may be true but at least part of it is old (see discussion under site Son-593 in Eastern Group below). It would be desirable to auger the talus piles although it would be quite difficult and expensive. The site descriptions of the Western Group begin with Son-265 on Rail Creek and proceed in order downstream. Site Son-565 (Rail Creek) This is a rather promising midden site ard was recaomerKded for excavation in the Phase I report. It turned out to be outside the project area, however, and was therefore not tested. It appears to have some importance as a large site rather far up in the drainage. - *e **v liI Iii H -0 ,1. . - . ' , , . ''I lll ii - i0, 711111111 .... ................... . ......... . .... .. - :. l _ ,,,,, ',,:~~.,'. wN1111llllllll **eeev* ; J I I > * 06 ........................ X ~11 1111' 1i XI w 11~~~~ H._ 11 /) - 11~~~~~1111 2111 qllllllllll[@@@1111111111 I1 I IC , i S);? 0 0~1II1IIllllI 0~~~~~~~~ Map 2. Western Site Group 10 11 Site Son-564 (Slump) Sor-564 is located at the extreme western edge of both Dry Creek Canyon and the Project Area boundaries. It is adjacent to and just north of the bridge where Hot Springs Road crosses Dry Creek. The site is situated on the north bank of Dry Creek, on a second terrace, 10 m above the creek. Two other high terrace sites are located near Son-564, on the north bank of Dry Creek; Sor-565 (midden) is 120 m north of this site, and Sor-567 (midden) is 1 km south. The site is in an open oak/woodland vegetation cammunity on the edge of a pranontory that has been intensively eroded by Dry Creek. Heavy slumping occurs on the bank of this pranontory. The promontory faces east and is cut off fram Hot Springs Road on the west by a modern wire fence. During the survey phase this site was located aix] described as a midden site remnant. It was indicated at that time that most of the site had eroded by the destructive forces of Dry Creek. A test unit was placed in the center of the 20 m by 20 m midden area on the promontory. Cnly 20 an were excavated in this unit. The excavations revealed that no midden now remains on this pranontory. Site Son-566 (Pock Rock) Son-566 is one of four sites found in the extreme northwestern corner of the Project Area. Unlike the other three sites in this area, Son-566 is not situated directly on the banks of Dry Creek, but is located on an intermittent tributary of Dry Creek, about 350 m west of that creek. The site is an open flat on the south bank of this intermittent stream, at the confluence of this stream anx] a second intermittent stream. The second stream forms the eastern boundary of Sor-566, the primary intermittent stream the northern boundary, and the sloping hills form the southern and western limits of the site, caupletely enclosing the flat. This -site measures 50 m by 60 m. The flat has open grassland covered with bay/oak woodland vegetation on both sides of the intermittent streams. The streams themselves support canmon riparian plants, and are actively running nearly year-round. The soil of the site is black clayey silt with talus gravels and cobbles fran the southern and western slopes surrounding the flat. The soil on the other banks of the intermittent streams are dark brown silty clay and contrast with the dark midden soil of the site. Very little historic or modern disturbance is evident on the terrace. Sorr-566 is a midden site with an associated cupule rock. The cupule rock is located on the east central of the flat, near the second intermittent stream. It is a low, flat rock measuring 4.5 meters by 3 meters and bears scores of ground cupules. During surface reconnaisance in Phase I a concentration of chert and obsidian flakes, a grooved hammerstone, hopper mortar fragments, and fire-cracked rocks was found approximately 30 meters north and west of this cupule rock. These Unit I was placed one meter due east of datum in this concentration. 1% Unfortunately, excavation of this unit had just begun when it was found that this site is located just outside the Project Area boundaries on private property. Only three ten centimeter levels had been canpleted at that time and a fourth begun. Excavation immediately ceased arn the fourth level was not finished. Fran the three canpleted levels it was clear that a well-developed midden site existed at Son-566. Chert debitage dcminates, with obsidian, animal bone ard fire-cracked rock being present in moderate quantities. One chert projectile point pre-form was recovered fran the midden canponent. Site Son-566 (Pock Rock) Level Proj. Used Cores Debitage Fired Animal an. Pts. Flakes Rock Bone 0-10 1-c 4-c 1-c 83-c 8 2 l-o 44-o 10-20 2-c 28-c 23 1 3-o 18-o 20-30 50-c 35 1 18-o Site Son-567 (Homestead Pasture) Son-567 is a midden site located on a second terrace on the north bank of Dry Creek, arn] 60 meters above the creek. It is approximately 30 meters south of Hot Springs Road between the road and Dry Creek. Its eastern and western boundaries are both formed by seasonal drainages which outline the circular flat 70 by 90 m. Two dirt roads extend south fran Hot Springs road on either side of the eastern seasonal drainage. The western dirt road extends straight down towards the Dry Creek and ends at an old barn which is still standing. The road is parallel to a row of sheep stake fencing which turns east at the barn and surrounds a pasture. The second, eastern dirt road winds south fran Hot Springs Road to Dry Creek aid crosses the midden area of Son-567 at its southeastern corner. At this point a gate is located on the dirt road. The dirt drive then extends around the southern extent of the midden, crossing the road at its gate arn continuing on to and through the western seasonal drainage. The midden area has open grassland cover presently supporting volunteer hay aid annual grasses of other types. The field encanpassed by the site has had previous cultivation but seens to have lain fallow for a nurber of years. Evidence of burning to clear the meadow for cultivation can be found along the west side of the pasture where several stumps are located. A broadleaf maple and a bay tree are located directly on the site. Both intermittent streans support riparian vegetation including willows, cottonwoods, maples, grape, ard a boxwood elder. Woodwardia and maiden-hair fern are thick in these drainages arnd soap root is in evidence in the iumnuiate area. 13 CA Sons 567 Map N - sss- _ _,.0J> 0 5 10 Meters Map 3. Son-567 (legend p. 5). 0 I I .0 - 0 - - 20 30 40 50 - 60 - 70 - o80 90 - - 100 - -110 Ii ktill- v vQ u I I : l : 10 A~d1!r 11 A 0 0 303 33 0 333 3 03330 ID0 33333 3 33 33 3330 00 30~ 3 33 3 33 3 I 33333 3333333333'K 3 0 a 3 3 333333333 0 333333 3 31 330 33003 033 33333333 0 a 33-a 3 0 0 3 , A. Midden; dry; very dark gray to black; silty loam with roots fran annual grasses. Cultivation zone. Small gravels. B. Midden; black; silty loam with small gravels. Friable. Fire- cracked rock. C. Transitional zone. Dark grayish brown; sand with silt ard gravels. Moderately compact. D. Sub-midden; dark grayish brown to brown; sandy gravels; cobbles. Moderately canpact. E. Yellowish brown; ashy clay lens. Moderately canpact. Figure 3. Son-567 Profile. 14 - - . I. . Ij - j-I - -I) wII-I - . .. .-) ." - . .. ... -q . I *- P.... .... . ---- - - - V-,.l -  I 0 0 0 0 0 0 0 0 0 0 10 0 0 0 0. I I I D a 0 a D D 15 Site Sorn-567 (Hanestead Pasture) Level Proj. Bifaces Reworked Scrapers Retouch Used Cores Debitage cm. Pts. Biface Flakes Flakes 0-10 19-c 1-o 4-o 10-20 l-o 2-o 34-c 8-o 20-30 2-c 1-c 3-c 54-c 4-o 28-o 30-40 1-c 57-c 1-o 1-o 2-o 26-o 40-50 3-c 41-c 1-o 1-o 1-o 10-o 50-60 1-c 1-c 33-c 1-o 8-o 60-70 17-c 1-o 1-o 7-o 70-80 17-c 1-o 1-o 1-o 3-o 8-o 80-90 7-c 1-o 90-10 0 1-c 2-c Total 2-c 1-c 11-c 1-c 281-c 4-o 1-o 2-o 1-o 5-o 12-o 99-o Grand 6 1 2 1 4 23 1 380 Total 16 Site Sor-567 (Homestead Pasture) Level Milling Bone cm. Stone Tool 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 1 Fired Rock 1 86 80 110 56 71 1 40 26 Animal Shell Historic Objects Bone (square nails) 4 15 41 87 75 46 32 22 8 2 3 3 5 1 x x 1 444 332 Total 1 17 The soil of the midden is dark grey brown to black sandy silt, and can easily be distinguished fram the surrounding light brown silts. Augering of the pasture revealed that the midden covers most of the meadow area ard extends under the dirt road and to the east of it. Tess Unit 1 was placed in the darkest part of the midden, 11 meters south and 15 west of site datum, and 15 meters due west of the broadleaf maple tree on the site. Excavation of this midden canponent showed a well-developed midden zone continuing to 80 an. It consisted of black silty loam for the first 40 an, at which point a shallow (5 an) ashy, clay lens intruded into the midden (Feature 1). Below this level, the soil became more clayey but was still midden-like in quality and texture. A shallow sub-midden component follows for twenty centimeters and over-lies sterile soil. The last excavated level (100-110 an) involved only the western half of the unit since it was dug into sterile soil. Artifacts were spread evenly throughout the midden canponent. Three diagnostic projectile points were found, two at 20-30 an, and one at the 70-80 an level, (chert arn obsidian), all representing a late time period. A fragmentary point at 60-70 an may represent an earlier period. Four burins, a biface, a scraper, four retouched flakes arnd a core were also recovered. COe sandstone millingstone was also recovered in the 30-40 an level. One piece of incised bone was recovered. Chert debitage was continually found in greater quantities than obsidian ard both were recovered in moderate amounts. A very large amount of fire-cracked rock arn animal bone also came fran this unit. It is of interest that evidence of historic contact (square nails and a ceramic sherd) continued to the 40 an level. Since the upper madden represents a very late canponent, this site is very important in terms of understanding late prehistoric-early historic contacts. Site Son-575 (Loop) Sor-575 is located on the northwest bank of Dry Creek, on a loop of the creek, about 250 meters southwest of the west entrance to Hot Springs Ranch. Dry Creek Canyon here has very steep and precipitous slopes with this midden site situated on a primary terrace about 6.1 meters above Dry Creek. Son-575 is near a deep pool in Dry Creek at the confluence of Dry Creek and an unnamed intermittent tributary. Originally the site was thought to be bisected by this tributary. However, augering indicated that the midden is limited to the downstream side (with respect to Dry Creek) and does not extend onto the other side of the intermittent stream. Erosion at the mouth of the intermittent stream is quite heavy forming very steep embankments with large boulders on their erosional banks. The main pranontory (west of the tributary) has sheep stake fencing along its northern border, and has a wire fence along the west which encloses the midden. The total extent of the midden encanpasses as area measuring 6 by 10 m. The small pranontory (east of the drainage) has sheep stake fencing along its southern edge. Sorr-575 has open grassland cover dominated by oak/woodland vegetation. Oaks, bays, buckeyes and manzanitas border both the intermittent stream arn a 18 Site Son-575 (Loop) Level Proj. aC. Pts. Biface Reworked Biface Scraper Retouch Flakes Used Debitae Flakes Fired Animal Rock Bone 0-10 1-c 2-o 10-20 1-c 20-30 30-40 1-c 1-c 1-c 6-c 3 2-c 15-o 1-c 1-c 1-c 2-c 33-c 5-o 26-c 5-o 2-c 8-c 6-o 40-50 3-c ToLtal 1-c 1-c 3-c 1-c 5-c 8-c 102-c 73 4 2-o 31-o Grard Total 3 1 3 1 5 8 133 73 4 3 20 2 30 20 2 19 steep sided ephemeral stream on the western boundary of the main terrace. The promontory also has oaks and buckeyes on the primary terrace. The large oak located on the easternmost edge of the promontory is presently covered by wild grape. Heavy slumping occurs along the steep cuts, especially on the banks of the site, the outside curve of this 'loop' in the creek. There is a large amount of sub-angular serpentine rocks naturally occuring in the area and Son- 575 is covered with these rocks. Most of this site has been lost to stream erosion aid at present only an edge of the original area of habitation remains. Test Unit 1 was placed three meters south of the datum stake, on the darkest area of the midden and in the area of least rock coverage. Excavation of this unit revealed a shallow but well-developed midden, followed by a transitional zone aid very shallow sub-midden component. The midden (0-30 an) contained two obsidian projectile points, one a side-notched point of the latest phase (Fig. 23a). Besides these tools, a knife, three reworked bifaces, a scraper aid two retouched flakes, all of chert, were found. Fire-cracked rock was found in moderate quantities, reaching an apex in the 30 an level Very few pieces of burnt animal bone were recovered. The next ten centimeters represent a transitional zone between the midden and sub-midden components. Mottling by clay patches riddled the dark brown, sandy, silty, midden soil, aid at 40 an the bottom of the unit was covered by an extensive root system. What is probably a late chert, straight based projectile point was found in this zone (Fig. 23b), as well as two retouched chert flakes. The amount of debitage, in general, greatly declined from the midden component. The mottling continues in the sub-midden component (40-50 an), aid seems to suggest some type of old disturbance. The major root system found in the transitional zone continues to expand with depth. At this point the soil has become mostly brown silty clay. Only three pieces of unutilized chert were found in this last excavated level. The unit was augered in the center from the 50 an to 140 an level, and continued until auger refusal by roots at 140 an. These 90 an contained brown, sandy silt with charcoal flecks throughout. The last ten centimeters of augering contained charcoal lumps and small water-worn stream gravels in abmdance. Site Son-572 (Banded Rock Pool) Son-572 is located about a mile ard a half west of the east entrance to Hot Springs Road aid 600 meters south of the west entrance to Hot Springs Ranch off Hot Springs Road. It is a second terrace site in a sloping area on the north bank of Dry Creek. The site is 400 meters upstream from Son-571 and is geographically similar to that site. Both sites are located on second terraces on the north stream bank near a deep pool in Dry Creek. Both have first terrace extensions which seem to have been utilized as a grinding areas. The extension at Son- CASon-572 Map 0 20 Meters A A A A A& A 0 A I 0 _0 ~ ~ ~ ~ ~ ta ': % . .*..- Map 4. SSm-572 (1egend p. 5). 0* * 0 II* * . I * * I1* . . * 0 .... II* S . . I * * 5 5 6 I I * S 0 . I * 0 S . lI* * * 5 l * l* 5 l* 0 * 0 0 5 0 * 5 0 0 0 * S S S S * 0 0 5 0 * 0 0 0 0 * S S 0 5 * 5 5 5 0 * 0 0 0 5 * 0 0 5 0 * S S 0 0 * 0 5 0 0 X* X X 0 5 * * 0 0 0 * 0 5 0 6 * S 5 5 4 * * 5 5 * 0 5 0 0 * 0 5 0 0 ** 0 5 0 0 b.* * * 5 0 * * * * * * 5 0 0 0 * 0 5 5 5 5 * 5 0 S S * 0 0 5 0 0 * 0 0 0 0 I * 0 0 5 . I 0@ 5 S 0 . * 0 * * 5 I 0 * 0 0 0 * 0 5 0 * 0 0 * 0 * 5 0 0 0 0 0 * S S * 5 0 0 * 5 5 * * 0 * 0 * 0 * S0 S0 SO0 20 21 572 is located 80 meters upstream fran the second terrace midden in an area of open grassland subject to heavy slumping, directly on the bank of Dry Creek and across fran a deep pool in the creek. Surface finds fran this lower area consisted of chert and obsidian flakes (75% chert; 25% obsidian), many fire- cracked rocks and a large (30 by 15 cm) hopper mortar fragment shaped with pecked cupules on the bottom surface. This surface scatter covers a 60 meter by 20 meter area, surrounded by sheep stake fencing. The midden at Son-572, on the second terrace, is cut off fran the slope to the creek by a modern wire fence which forms the southern boundary of this site. It is located in a deep basin and has sloping hills forming the northern, eastern and western boundaries of the midden area. This basin encompasses approximately 25 meters by 12 meters in areal extent. Sorr-572 has open grassland cover (unlike the sheltered terrace at Sor- 571) with a few isolated oaks, bays and manzanitas in the basin. Scattered oak stands are found on the surrounding slopes. Tree cutting activities have slightly disturbed the surface of this site. The basin is horse-shoe shaped and the midden seems to follow the contour of this area. The midden is very dark brown gravelly sand and can be visually distinguished fran the surrounding alluvial sands. One auger hole was placed in the lighter sands in the center of this horse-shoe shaped midden area to see if buried midden could be distinguished. It was found that no midden occurred here and that the midden does accurately follow the countour of this basin. Test Unit 1 was placed in the definite midden area 1 meter north aid 1 meter east of the site datum stake. Excavation of this unit revealed a well- developed midden canponent extexding to 130 an. This component is followed by a shallow sub-midden zone which overlies sterile alluvium. Stratigraphically, the midden appears to be one continuous deposit of cultural material. It consists of dark brown silt with snall angular gravels, hundreds of madrone anx oak tree roots, and larvae extending to 100 an. At 60-70 an there is an increase in the sand content of the soil an consequently a lightening of soil color. Gravels begin to decrease in quantity at this same level. No soil profile is given for this site since there is no sharp demarcation showing this change. The lithic debitage also slnws a change in the midden component at about this level. The upper levels of the midden were dominated by obsidian debitage with moderate quantities of chert debitage anx] animal bone appearing. Large amounts of fire-cracked rock were also present indicating a well-formed midden deposit. However, chert completely dominates the debitage remains of the lower levels. Obsidian counts begin to decrease abruptly after their peak in the 50-60 an. level. Correspondingly, chert flakes increase in quantity in the same point, reach their peak at 90-100 an, and retain very high counts until the end of the midden canponent at 135 an. Fire-cracked rock continues at high quantities until the end of the midden. Obsidian flakes and animal bone remains, on the other hand, were recovered in very small amounts fran these lower levels. The midden at Son-572 has two separate tool concentration corresponding 22 Site Son-572 (Barded Rock Pool) Level Proj. cm. Pts. Bifaces Reworked Biface Scrapers Retouch Used Cores Flakes Flakes 0-10 1-c 1-o 10-20 2-o 20-30 30-40 40-50 50-60 60-70 1-o 1-c 1-c 1-c 2-o 1-o 1-c 1-c 1-o 1-c 1-o 2-c 1-o 1-c 1-c 1-c 1-o 1-o 5-c 9-o 1-c 2-o 2-c 3-o 6-c 2-o 1-c 6-o 2-c 2-o 1-c 70-80 1-c 1-c 80-90 1-c 90-100 1-c 2-o 1-c 2-o 3-c 1-o 1-c 3-c 5-c 1-c 1-c 16-c 50-o 30-c 45-o 23-c 54-o (56-c) * 59-o (1-c) * 80-o 1-c 24-c 85-o 35-c 68-o 1-c 65-c 58-o 82-c 28-o 168-c 17-o 100-110 110-120 1-o 120-130 1-c 3-c 1-c 120-c 4-o 1-c 1-c 1-c 130-140 8-c 2-c 113-c 3-c 1-c 1-c 1-c 123-c 3-o 76-c ToLtal 4-c 8-c 9-c 5-c 42-c 7-c 932-c 3-o 2-o 5-o l-o 2-o 29-o 551-o 10 5 10 7 71 7 1483 *Chert fran 30-40 an] 40-50 ina3vertantly canbined Debitage Grarnd 7 Total 23 Site Sor-572 (Barded Rock Pool) Level Milling Hamner Net Fired Animal cm. Stone Stone Weight Rock Bone 0-10 7 1 10-20 134 9 20-30 100 9 30-40 98 7 40-50 1 65 9 50-60 70 4 60-70 100 5 70-80 1 97 7 80-90 76 9 90-100 77 2 100-110 57 1 110-120 1 1 58 - 120-130 92 1 130-140 24 2 1 1 1075 Total 2 66 24 to the two midden levels. Within the upper levels, two small, side-notched projectile points were found (obsidian and chert) which are diagnostic of the latest phase. Eleven reworked bifaces (7 obsidian), four bifaces, three scrapers, arn one chert core were found with these and two other projectile point fragments within the first half meter of excavations. One millingstone fragment was also recovered in this concentration. The second tool concentration extends fran 70 an to the end of the midden component and also contained two diagnostic projectile points. Both points are large chert side notched points and correspond to Meighan's Mendocino Ccmplex, Type 11, (Meighan, 1955), and thus to the Borax Lake aspect. These points are associated with six biface (five chert), two buins, five scrapers, and four cores, all of chert. Also, a millingstone, a basalt hammrstone, and a possible net weight came fran the lower levels. these tool concentrations correlate nicely with the debitage change noted above, and suggest that at least two separate cultural canponents existed within the midden zone. We discuss the chronological implications in more detail in the group summary. The shallow sub-midden zone at Son-572 is characterized by a great decrease in cultural material of all types. Both animal bone and fire- cracked rock disappeared in the lower levels of the midden and do not re- appear in this lower zone. Very few pieces of chert or obsidian debitae were recovered. This component overlies sterile soil. Site Son-571 (Poolside) Son-571 is located approximately 1 mile west of the east entrance to Hot Springs Ranch off Hot Springs Road ard approximately 250 meters south of that road. It is on a small secondary terrace on the north bank of Dry Creek in a deep canyon. This terrace is approximately 12 meters above Dry Creek in a small flat clearing in an otherwise boulder-strewn wooded slope. The site is tightly delineated on the north, east and west by talus slope covered with large boulders. The southern boundary of this site is the steep cut bank above Dry Creek. Burnt logs are found on the site and many madrone, bay and oak trees canpletely shelter this small terrace. The entire terrace is covered with black midden and measures 35 m by 15 meters in its widest parts. Historic activity on this terrace is indicated by a possible firepit (ca. 5 m south and 1/2 m west of datum) and by sheep stake fencing which runs parallel to the creek, ca. 6 meters north of datum, bisecting the terrace. Surface finds on this terrace include an obsidian mid-section of a projectile point and an obsidian leaf-shaped point base fragment. About a meter northwest of this second terrace is a very small clearing among the large boulders that fringe Son-571. This clearing has yellow-brown silty talus soil but contains a surface scatter of chert and obsidian debitage. Soap root, wild white and purple irises, and bunch grass cover this area. Mariposa lilies are also present in this small northern area and on the northern edge of the midden covered terrace. Poison oak is abundant among the boulders between these two open sections. Approximately 20 meters east of and below this second terrace are the remnants of a first terrace. This lower area has two sections, a grassy slope and a clay/cobble area next to the gravel of Dry Creek. The CA Son 571 Second Terrace Map 0 5 10 a a -. Meters I7 O . Map 5. Son-571, Second Terrace (legend p. 5). 25 CA-Son* 571 First Terrace Map 0 5 10 I1 I - 7 4 4 - .0 V Map 6. Son-571, First Terrace (leend p. 5). 26 27 clay/cobble zone has numerous ground stone tools still lying on the surface including milling stones arid a maul. The northerrinost section of this zone has the heaviest concentration of artifacts. These artifacts may have eroded out of the grassy slope or may have been lowered as soil fran the area was washed away by Dry Creek. On the first terrace are madrones, oaks, bays, art two alders. The clay/cobble zone is located approximately 1 1/2 meters above Dry Creek. 20 meters south and 20 meters west of this zone, Dry Creek backs up against a rock spur on the opposing bank ard forms a deep pool of water, after which this site was named. Test Unit 1 was placed 2.5 meters due east of the site datum in the center of the black midden on the second terrace. Excavation of this unit revealed a well-developed midden extending to approximately 85 an. A transitional zone (85-110 cm) followed the midden canponent ard was characterized by mottling due to increased clay content of the soil. Finally, a shallow sub-midden canponent exterded to 130 an. The midden component was characterized by very black, greasy, slightly sardy, silty, loam. This canponent contained huge quantities of obsidian, unparalleled by any other site investigated. The lowest count of obsidian debitage in the entire 85 an of midden was 364 pieces, ard the peak was reached in 50-60 an at 713 pieces. Chert debitage consistently renained under 40 pieces per level until obsidian began to decline (70-80 an), at which point the chert count jumped to 64 pieces. Larger quantities of animal bone than recovered fran the usual site were found at Son-571 (up to 71 burnt pieces per level), an] animal bone count seem to rise and fall with obsidian. Unfortunately, the area of this site is so littered with angular rocks that difficulties were encountered in trying to separate out fire-cracked rock. Only unquestionable pieces were recorded in the level records. Thus, the quantities of fire-cracked rock recorded (unusually small for a midden site) may be misleading. Twenty-one projectile point fragments were recovered fran the midden canponent, three of which are canplete. Two small, possibly late specimens were found in tne 0-10 and 10-20 cm levels of the midden zone. The third canplete point is a side notched chert projectile point we date in the Borax Lake aspect. This point was recovered fran the 60-70 an level, just at the point when the high obsidian counts are declining ard chert debitage is increasing. A bowl mortar rim fragment was found with the late points; a possible handstone below the late points; and a definite hardstone one level below the side notched projectile point. It is important to note that preference for certain types of lithic material can be seen at Son-571. This site has a unusually high quantity of retouched flakes, (26 were recovered fran the midden canponent), only four of which were chert. 20 reworked bifaces came fran the midden zone ard all of these were manufactured fran obsidian and eleven bifaces (ten obsidian) were recovered. However, despite all of the obsidian debitage, retouched flakes, and tools, no obsidian cores were found. Four chert cores and two chert scrapers were recovered implying that chert was still preferred for the heavy-duty tools. The transitional zone fran Unit 1 was characterized by an increase in the clay content of the soil producing subtle mottling and lighter soil 28 A. Midden; dark brown to black; silty loam with sand. Root zone. Friable. Rock intrusion: mainly angular. B. Greasy midden; black to very black; silty loam with sand. Very friable. Rock as in "A". C. Transitional zone. Very dark grayish brown to dark brown; silt with sard ard clay. Moderately friable. Rock as above. D. Sub-midden; brown to dark yellowish brown; gravels with sand and clay mix. Compact. Rock as above. E. Rock. Figure 4. Son-571 Profile. Proj. Pts . 2-o 1-o 4-o 1-o 1-c 2-o 3-o 1-c 4-o Site Son-571 (Poolside) Bifaces Reworked Scrapers Retouch Biface Flakes 1-o 2-o 1-o 1-o 2-o 2-o 1-o 1-c 1-o 90-100 1-c 100-110 l-o 110-120 120-130 Totals 18-o 3-c Grand 22 Total 10-o 1-c 11 4-o 4-o 3-o 4-o 3-o 1-o 1-c 2-o 1-o 1-c 1-o 1-o 23-o 23 3-o 2-c 5 1-o 5-o 2-c 1-o 9-o 3-o 1-o 2-o 2-c 22-o 4-c 26 Used Cores Flakes 14-o 4-c 1-o 1-c 4-o ll-o 7-o 2-c 1-c 7-o 2-c 7-o 2-c 19-o 3-c 5-o 5-o 3-o 2-c 1-o 84-o 13-c 4-c 97 4 29 Level cms. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 Debitage 785-o 39-c 506-o 14-c 366-o 26-c 544-o 20-c 364-o 32-c 713-o 13-c 516-o 69-c 416-o 31-c 251-o 64-c 265-o 36-c 95-o 23-c 42-o 10-c 2-o 3-c 4865-o 380-c 5245 30 Site Son-571 (Poolside) Level Hard Mortar Fired Animal an. Stones Pock Bone 0-10 4 44 10-20 1 17 14 20-30 14 43 30-40 1 11 47 40-50 11 71 50-60 14 66 60-70 6 57 70-80 1 7 34 80-90 4 30 90-100 3 20 100-110 2 5 110-120 1 2 120-130 1 95 433 Tobtal 2 1 31 colors. With this clay increase large rock concentrations of cobbles and boulders began to fill the unit. The transitional component continued until approximately 110 an when a definite stratigraphic change to a lighter soil color became apparent. Lithic material fran this camponent included 2 undiagnostic obsidian projectile point fragments, 1 obsidian burin on a snapped biface, anx] two retouched pieces of chert. Animal bone, fire-cracked rock, anx] both obsidian and chert debitage counts declined. The shallow sub-midden canponent at Son-571 consists of flakes only; no tools being apparent. Further, only two fire-cracked rocks and two pieces of burnt animal bone were recovered. It is highly probable that these pieces are only partially representative of the material fran this component, since heavy rock concentrations prohibited excavations. Both the transitional and sub-midden zones were deposited on talus slope soil with boulders fran a rock slide that may have been contemporaneous with the large boulder concentrations that now surround Son-571 on three sides. It should be noted that a large quantity of obsidian fell through the 1/4" screen used to dry screen the site. Only small bits of charcoal were found in while excavating. This latter may correlate with the small number of fire-cracked rocks found in the midden or may be influenced by the fact that roots and rootlets continued to the 130 an level. Site Son-570 (Flat Iron) Son-570 is a petroglyph site located 280 meters upstream fran site Son- 568. It consists of two large boulders 30 meters apart with hundreds of cupules in them. Detailed descriptions of these petroglyphs, as with most others in the project have not yet been made. Site Son-568 (Smiley) Son-568 is located on a first terrace on the north bank of Dry Creek, approximately 200 m south of the Hot Springs Road. It is 3/4 of a mile west of the east entrance to Hot Springs Ranch and approximately 200 m west of where Hot Springs Road climbs away fran Dry Creek. Son-568 has three serpentine rocks with cupules apparently associated with the midden. They are located 91.70 m southeast fran the Son-568 site datum. Sort-568 is a well-developed midden site bordered on the north by a seasonally intermittent stream. A second intermittent stream, 15-30 m south of the first, bisects the midden area of Son-568. The area of this site continues approximately 15 m south of the second intermittent stream on its first terrace. Both of these seasonal streams empty into Dry Creek, which is located 10 m to the west of the midden, down an abrupt cut bank. This abrupt cut bank forms the eastern boundary of Son-568. The western section of the midden has been disturbed by an historical roadbed which runs the full extent of both first terraces and crossed the two intermittent streams by two wooden bridges. The north bridge that crosses the first intermittent stream is more or less intact. The south bridge across the southern, second intermittent stream has collapsed sections. The area of Son-568 passes under this old roadbed arn at least 5 m west of it. Son-568 has open grassland vegetation with oak/woodlarnd cover. The site 32 CA Son * 568 Map 0 5 ID meters Slump ; CU ule Rock A 91. 1M From% Datum Map 7. Son-568 (legend p. 5). 111% -% , 9 Ec P 9 9 no Eg 9 9 I 9 I 33 is heavily populated wih oaks arn bay laurels, vegetation especially interesting because the species of oak present is black oak, which has not been noticed on other sites in the Project area. The small terrace between the two intermittent streams is campletely enclosed by these oaks, screening it fram the surrounding slopes arK terraces. A large fallen oak lies across the western edge of this terrace ard extends onto the historic roadbed. The midden is canposed of black, silty, gravelly, alluvial sand. A series of seven auger holes was dug into the site. These test holes revealed that the darkest part of the madden is located within the highest contour, between the two intermittent streams, ani between the steep cut bank ard the old historic roadbed. The midden continues *west of the roadbed, east beyond the 400' contour, and south of the second stream, but is slightly lighter in color and seems to be fanning out fram the central area. Lithic material was recovered fran the two auger holes near the site datum, just east ard just west of the roadbed. Test Unit 1 was placed on the darkest part of the midden area, 13 meters east of the site datum stake. Excavation of Text Unit 1 revealed a well-developed midden, followed by four more components consisting of transitional, sub-midden, gravels ard clay. The unit was excavated to 170 an where the few flakes recovered were fran a disturbed area. These camponents should be described as stratigraphic changes and not cultural components. The 0-90 cm level is broken into three zones on the basis of rodent disturbance which produced increasing mottling arn increased clay content in successively lower levels. The soil appears to get one chrana lighter approximately every 30 cm. However, the cultural layer continues to approoximately 90-100 cm where it gives way to relatively sterile clay gravels. The midden zone is characterized by a high artifactual content which remains without marked deviation in quantity, at least down to 110 or 120 an Fire-cracked rock alone rapidly decreases fram the top of the midden to its lowest points. Chert is by far the most canmon material found in the debitage, although numerous artifacts were maie fran obsidian as well. Of the ten projectile points recovered in the midden, two are diagnostic: a small obsidian Excelsior point (40-50 an) and a red chert Gunther barbed projectile point (60-70 an). Animal bone fragments consistently stayed below 20 pieces recovered per level. Two pestle fragments were found. At a meter depth, animal bone remains and fire-cracked rock cease; utilized pieces of either chert or obsidian are rare, arn] only one actual tool were recovered in the full 70 an excavated in the gravel zone. Debitage counts remain at less than 10 pieces per level. Stratigraphically, this zone is divided into two canponents. The first consists of clay with a high gravel content (perhaps suggesting stream flooding), ard the lower canponent is relatively gravel-free. No diagnostic chipped or ground stone tools are found below 1 meter. Test Unit 1 was dry screened until 30 an when both dry and wet screening were introduced to facilitate sorting. At 90 an the clay content of the soil began to increase rapidly ard wet screening was solely employed. 34 - 0 140 _ 3 0 1-40 1 507 _ 4_ _ 160 _ a _ - I _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ A. Midden; very dark gray to black; silty loam with sand. Root zone. Friable to very friable. Fire cracked rock. B. Transitional zone. Very dark gray; silt with sand and clay. Moderately compact. C. Sub-Midden; very dark grayish brow to grayish brown silty clay with sand. -Decrease in armont of fire-cracked rock. D. Dark brown clay with sand and gravels. No fire-cracked rock. Compact. E. Brown to dark yellowish brown clay; sticky. Comact to very compact. K. Krotovina andi/or heavy disturbance. Figure 5. Son-568 Profile. Site Son-568 Bif aces Reworked BiLfaces 1-c (Smiley) Retouch Flakes 1-c 1-o 1-o 1-o 1-o 1-c 1-c Used Cores Flakes 3-c 1-c 1-c 1-o 2-c 1-o 3-c 1-c 1-o 2-o 4-c 2-c 1-c 1-o 2-o 2-c 35 Proj . Pts. 1-c 2-o 2-o 1-c 1-c 1-o Level cm- 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 1-c Debitage 60-c 15-o 50-c 16-o 41-c 13-o 54-c 17-o 63-c l1-o 82-c 16-o 50-c 12-o 75-c l1-o 47-c 5-o 42-c 4-o 1-o 36 Site Son-568 (Smiley) Level Proj. Bifaces Reworked Retouch Used Cores Debitage cm. Pts. Biface Flakes Flakes 100-110 2-c 16-c 110-120 9-c 1-o 120-130 6-c 130-140 3-c 1-o 140-150 7-c 150-160 1-c 2-c 160-170 1-c 3-c Level Pestle Fired Animal an. Rock Bone 0-10 121 1 10-20 1 251 3 20-30 300 6 30-40 95 12 40-50 175 14 50-60 105 18 60-70 1 66 1 70-80 90 4 80-90 32 1 90-100 10 2 (None below 100 an) 37 (Summary) Site Son-568 (Smiley) Level Proj. Bifaces Rework Retouch Biface Flakes Used Core Flakes Debitage Fired Animal Rock Bone Totals 4-c 2-c 2-c 22-c 2-c 610-c 1245 62 6-o 2-o 2-o 6-o 122-o 10 2 2 4 28 2 732 1245 62 Grand Totals 38 Besides . the midden area, two cultural features are present at Son-568. Three housepits are located on the midden; one within the 400' contour, approximately 5 meters northwest of Test Unit 1; a secord, 20 m southeast of the unit on a lower contour; ard the largest housepit (ca. 5 m diameter) is located approximately 20 m south of Test Unit 1 on the terrace south of the second intermittent stream. Three cupule rocks are found approximately 92 m southeast of site datum arn 100 m east of an old shed. These rocks are 5 m by 5 m ard 4 m high ard have numerous cupules and at least one grove. Test Unit 2 was placed 20 cm east of the northerrinost petroglyph in an attempt to find associated cultural material. Excavation of Unit 2 revealed no midden ard only one artifact in the full 40 an. Stratigraphically, two canponents were distinguished. The first (0-25 an) was characterized by dark brownish yellow alluvial sard with disturbed areas by possibly pig rooting activity. No cultural constituents were noted. The final zone (25-40 an) consisted of darker brown soil and produced a reworked obsidian biface. Excavation ceased at 40 an because of lack of cultural material. All material excavated fran this unit was dry screened. Western Site Group Chronology They are 8 sites in this group and 6 of these have enough information to attempt chronological placement. There are noted as follows. Son-566 There are hopper mortar fragments here so the site probably belongs to the Houx aspect but whether early or late we cannot tell. Son-567 There is a possible milling stone fragment at 30-40 an but the projectile point fragments are so small and delicate arn] the milling stone fragment so doubtful that we are inclined to attribute the entire deposit to the Houx aspect. There is a slight break in the chert-obsidian ratio at 20 an so we assign the top 20 an to late Houx and the remainder to early Houx. Square nails were found as in the top 40 an but these must be due to early white settlers rather than to historic Indian occupation. Son-575 This is probably at least partially late Houx because of the small side notched point (Fig. 23a) and may also be early Houx as well. We assign it to Houx generally. Son-572 This site presents some stratigraphic difficulty. We are certain that the upper portions are Houx aspect because of small side notched points on the surface and the 0-10 an level (Fig. 23d, e). We are also sure that there are Borax Lake components here because of the presence of 2 milling stone fragments and the large side notched points in the lower levels (Fig. 23g, 39 i). The first obvious break in the chippage ratios canes level but there is a milling stone fragment at 40-50 cm. conclude that the Houx components go down to the 70 an level milling stone fragment is out of place. at the 70-80 an Reluctantly we ard the (small) We divide the deposit as follows: 0-20 late Houx (arbitrary division), 20-70 early Houx (chippage break), 70-100 late Borax Lake (chippage break), 100-140 early Borax Lake. Son-571 There is no definite evidence of late Houx but the upper midden must be Houx because of the small Excelsior point at 10-20 an (Fig. 22e). We are assigning the deposit fran 0-80 an, where there is a break in the chippage ratio, to early Houx. Levels below that we attribute to late Borax Lake. This leaves a large side notched projectile point (Fig. 22h) out of position in Houx aspect context although not by very much. Son-568 There are two pestles (one at 10-20 this site must be Houx Aspect. There is a 70-80 an and we are therefore assigning the the rest to early Houx. an andx the other at 60-70 an) so break in chert-obsidian ratio at levels down to 80 an to late and western Site Group Chert-Obsidian Debitage Ratios by Site ard Level Site CA-Son 575 2.13(47) 6.60(38) 5.20(31) 1.33(14) * ( 3) 572 0.32(66) 0.67(75) 0.43(77) 0.41(87)# 0.41(109) 0.28(109) 0.49(103) 1.12(123) 2.93(110) 9.88(185) 30.00(124) * (131) 41.00(126) * ( 76) 571 0.05(824) 0.03(520) 0.07(392) 0.04(564) 0.09(396) 0.02(726) 0.13(585) 0.07(447) 0.28(315) 0.14(301) 0.24(118) 0.24( 53) 1.50( 5) 568 4.00(75) 3.13(66) 3.15(54) 3.18 (71) 5.73(74) 5.13(98) 4.17 (62) 6.82 (86) 9.40(52) 10.50(48) * (16) 9.00(10) * ( 6) 3.00( 4) * ( 7) * ( 2) * ( 3) debitage frequency infinity, or all chert 30-40 ard 40-50 levels inadvertantly canbined in this site 40 Level 566 1.89(127) 1.56( 46) 2.78( 68) 567 4. 75(23) 6.75(42) 1.93(82) 2.19(83) 4.10(51) 4.13(41) 2. 43(24) 2.13 (25) * ( 7) * ( 2) CM. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 ( ) : * 41 Western Site Group Chert-Obsidian Ratios by Site ard Phase 567 0-20 4.42 (65) 575 Site CA-Son 572 571 0-20 .48 (141) 568 0-80 4.28 (586) 0-50 3.29 (133) 20-100 Ha 2.62 (315) Blb BLa 20-70 .40 (485) 70-100 3.06 (418) 100-40 61.71 (439) 0-80 .06 (4454) 80-130 .21 (791) 80-130 12.27 (146) 2.84 3.29 1.69 (380) (133) (1483) .08 5.00 (5245) (732) : Houx Aspect : Early Houx Aspect : Late Houx Aspect : Early Borax Lake Aspect : Late Borax Lake Aspect Hb H H Ha Hb BLa BLb - | X.0~~~~~~~~~0 rn~ ~ ~ ~ ~~~c z N~ ~ ~ ~ ~ ~ w 01.1 ~~~~~~~~~~~~~-J 4 0 00 06 ~ ~ ~ ~ > ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I O x 0 ~ ~~~~~~ O ~ ~ w (0 C~ ~~~~~~~~~~~C Map 8. Eastern Site Group 42 43 Eastern Site Group The valley of Dry Creek is wider along this stretch than it is either above or just below Where it is pinched between two low mountain ranges. It widens especially on the NW around site Son-593 and again toward the SE at the mouth of Cherry Creek. Perhaps because of the availability of this flat bottom lard we have the densest accumulation of sites and also the two largest sites in our area. We argue elsewhere that this area was a tribelet center and had been for many centuries. Perhaps the reason was topographical; they simply had more roam for large winter groups to gather. It is also thanks to topography that this area has occasionally been thought suitable for cultivation and as a consequence one of the two largest sites (Son-593) has been partially destroyed. Another peculiarity of this area is that it is virtually all surrounded by soil types which have been determined as having been mixed evergreen and oak forest (Map 8). This contrasts with other parts of the project where different supposed vegetation patterns prevail. We indicated earlier that we considered mixed oak-evergreen land as slightly less valuable than woodlard- grasslard territory. In view of the apparently high population density here it may be that our assessment was wrong aix that the oak-evergreen environment is the most favorable. Alternatively it may have been the case that immediate environment was of less importance than that of the tribelet territory a whole and that location of principal villages deperded more upon other factors such as centrality or on topography as previously mentioned. A total of 13 sites was recorded for this group beginning with Son-593 in the northwest anx] continuing downstream to Son-597 at the confluence of Brush Creek and Yorty Creek. Site Son-593 (Treganza 3 and 4) As irdicated by the title this site was recorded as two separate sites in Treganza's survey. The separate midden deposits were originally one site with the central portion having been disturbed by agricultural acitvity as shown on Map 9. The north midden area is south of Hot Springs Road at the entrance to Hot Springs Ranch. Datum stake 2 was placed on the western edge of this midden. The south midden is about 150 m southwest of this anx] lies 15 or 20 m north of Dry Creek. This midden was deposited on a low mound a meter or two high. Datum stake 1 is at the highest point of the mound. Both middens are on a flat extensive stream terrace exterding about 150 m north- south aix] 300 m east-west. It is bounded by steepening hillsides to the north and west, by Dry Creek on the south ard by a spring fed stream on the east. Directly opposite this flat, across Dry Creek, is an extensive area of lard-sliding on a steep slope made up of a part of the Franciscan formation referred to as "melange," an easily weathered, serpentine derived metamorphic rock. Mr. Richard Thompson, a Corps of Engineers geologist, theorized that the mound on which the south midden is located was deposited by an older, more massive landslide which at the same time dammed Dry Creek, forcing it north of this mound. After a short time, perhaps 10 to 20 years, Dry Creek eroded through the landslide ard re-assumed its present channel. The CA-Son-593 1 4II 11111 I I I I ? I ~ I O fig*~~~~~~ ...~ \ | | ..._"...... Possible Former Mildden Boundary )0 *0 * * b. II9 ** a'** L* A Map 9. Son-593 (legend p. 5). 44 : - g l J A: :' 1 ::1 A; S :' >.# :: !: 1 ) ] X r I 45 landslide left behind the low, uneven mound on an otherwise flat terrace and an old stream channel that has since becane filled in and plowed over and is now poorly defined. As previously mentioned, agricultural land leveling has removed much of the midden, especially just west of the north midden. Plowing of the terrace flats has further buried or obscured the midden that still exists. Wien the land leveling took place and where the midden was taken could not be determined. There is a possibility that the midden sown in the southeast on Map 9 is re-deposited midden fran the area north of it. The vegetation in the flat around the site is now open grassland with a few isolated Oregon oaks, live oaks, and buckeyes. The surface must have been more heavily wooded in aboriginal times and may have been an open parkland with the usual riparian suite of laurel, alder, and willow along the fringes of Dry Creek. The slopes north and south of the flat are heavily wooded in oak and madrone with Douglas fir and laurel more prevalent south of Dry Creek. The soil sequence on the terrace around the south midden is 1.0 to 1.5 m of fine grained alluvial deposit underlain by about a meter of sandy alluvial gravel containing many boulders. Shale bedrock nrderlies this bed. The midden is very dark brown soil with considerable fire-cracked rock and some animal bone. It generally runs to a depth of 30-40 an, but goes as deep as 50 an in one place. A single small side-notched projectile found in the 0-10 cm level, the low chert-obsidian ratio of debitage, and the well developed housepits on the surface identify this as a late midden; we believe a single canponent midden. There are 11 well defined housepits on this south midden. They measure between 2.2 and 6.0 m in diameter and were 40 an to 1.5 m deep. Two are oblong and may be sweat houses while the rest are circular. There are at least 36 petroglyph rocks on or close to the south midden. All but one of these has cupules as the only design element. The exception is a wedge shaped rock with cupules on its south face and apex and some shallow grooves on its north face. The grooves are not well defined but are similar to other, definitely man-made examples found elsewhere in the North Coast Range. The petroglyphs are on medium sized boulders ranging fran 50 an to 2.0 m (avg. 1 m) in largest dimension. Almost all these boulders are found along the front edge of the terrace just along Dry Creek. They are scattered fran the south midden about 300 m downstream. There are also two cupule rocks on the south midden itself and an isolated one on a higher terrace 150 m to the northwest. The north midden is located on a secondary terrace 80 to 100 m fram Dry Creek. A large amount of midden has been removed by land leveling so that the northern section is now represented by two small remnants. In the north midden the excavated deposit can be divided into four units which constitute at least three cultural components. The topmost (A and B in the profile) consist of dark grey to dark brown midden with a large amount of fire-cracked rock and animal bone. It extended fran 0 to 70 cm in depth. There are square nails and bottle glass near the surface. Diagnostic artifacts include 46 CA Sons 593 Test Unit 1 Soil Profile North Wall -o 10 - i20 30 -40 50 60 cm A. Midden; very dark gray; sardy silt. Root zone. Friable. B. Decaposing shale. Friable. C. Decacposing shale and gravels; brown to dark brown silt with sand. Friable. D. Rock. Figure 6. Son-593, Unit 1 Profile. 47 ,oOOO-* *.,o 0 ,. . o . . 0 . ~ ~ ~ ~ ~ ~ ~ ~ 0 ' *o;"oo~oo"'OO) .- - *.. -. **.* * . ~~~~~~6 lb..* C)OOO.OC) OOOOC)CC(C)ooooO()()oo OOOC)OO00000 o0CCOO C)0000 c U C) C)) ) (C ) )( (C U )) ) 0 )) ) 0 0 0 ) ) C) C( ) (Co O CCC 0 0 0 0 C C) 0 0 0 C 0 (O Oo C) 0 0 0 0 0 C C ) (O O C C 0)()(()o~o()o( C ooo 0) ))C))))) C)C)C)CCC)C ,),ooo C()oooooloooooooC lo(Coo o 0)() ( o o(CCC C C ) ooC ooooooUCocC0C U O C)CCoooo 0CCo 0CCO Co 0000000 CC)() C)00)C)()( C)C)0)C)C) O0C)) C)i )OOCC )CO CC OOOOOOO)OOOO' C O~nOOOOOOO)OOOOC)OOOC 0C () )( ())CC)C)C)CCCC )C C)C)C C ( C)C C) OOOOO O oC)oo CCoooo o 000000000 OCC)O) O CCC)COCOOCOC O)OCCOCC)CC)C 0 CCCOCCOOO )00 0 00000 0CC)00O0 0 000 0 00 0 0 0 0 0 0 OCCCI )() ()O()oC)0)o0)()C(C)C()() oooooo oooooof o00CC o00000000o0000o )o (oo() ()o(o o o oooo() 0o oo0o()oc0 o oooot0 ooo()ooo()00 0 0of0 ooooooc 00ooo '(o()oo o()oo()o()O())()o((((()()oooooc)oooooo()ooooooooooooooo )o()(()o(o()o()oooooooooooooo(ooOo(')()oooooooooooooo0 ) U C C C C C ) U ) C) 0 0 C ) C ) C C C 0 0 0 C ) C ) C CO OO 0 OO C ) O C C C C Cr0 n nn)0 O O O O O O O O O CCO )C)C CO)) CO O CO ))O)O C C))) (CO00 O 00C C) 0 CCC C)0 0 0 C) CC 0CC0000000000000 A. Midden; dark gray to very dark grayish brown; silty loam. Annual grass roots. Fr iable. B. Midden; very dark brown to very dark gray; silt with sand. Friable to moderately canpact. C. Burnt living surface; very dark grayish brown; silt with clay and sand; canpact to very canpact. D. Transition Zone; silt with clay and sand; dark brown; friable to moderately compact. E. Sub-midden; dark brown to brown; silt with clay and sand; small gravels. Compact to moderately compact. F. Sand and gravels; brown; very loose and friable. G. Fire hearth; brown, sandy silt with char- coal and gravels. Friable. Figure 7. Son-593, Unit 2 Profile. 10 20 30 - 40 - 50 60 -70 90 - loo - 110 - 120- -130w -140- -150- 160- -170- -180- -190- -200 -210 -220 cm ( () ( ( () () () () () () 00 () C) ( ( () () 0 0 0 () 0 0 0 0 0 0 0 0 O 0 0 0 0 0 0 0 0 0 0 0 0 00 0 0 Oa O 00 ( L 4 p 11 a I I 4 I c F -d I I L 48 Site Son-593 (Treganza 3 & 4), Unit 1 Scrapers Used Flakes 1-c 1-c 1 3-o 1-c 2-o 1-c 1-c 5-o 6 Debitage Fired Rock 12-c 50 12-o 14-c 100 22-o 2-c 100 12-o 1-c 120 3-o 1-c 80 30-c 450 49-o 79 450 Level an. 0-10 10-20 20-30 30-40 40-50 Total Grand Total Proj. Ptso 1-c 1-c 2-c 2 Animal Bone 36 57 32 10 5 130 130 - - 49 Site Son-593 (Treganza 3 & 4), Unit 2 Level Proj. Biface cm. Pts. Rework Biface Scraper Retouch Used Flakes Flakes Cores Debitage 0-10 10-20 20-30 30-40 40-50 2-c 3-c 1-c 4-o 1-c 3-o 1-o 1-o 2-c 1-o 1-o 1-o 1-o 1-o 50-60 2-o 60-70 1-c 70-80 5-o 3-c 4-o 2-c 8-o 4-o 1-c 80-90 1-o 1-o 90-100 100-110 5-o 3-c 2-o 1-c 12-o 1-c 11-c 71-o 13-c 71-o 20-c 122-o 1-c 13-c 125-o 1-c 14-c 196-o 22-c 191-o 1-c 21-c 153-o 25-c 70-o 30-c 40-o 14-c 37-o 17-c 24-o Level Used Debitage Level Debitage Flakes c 0 c 0 110-120 120-130 130-140 140-150 150-160 1-c 12 8 8 2 3 1 3 1 160-170 170-180 180-190 190-200 200-210 4 8 8 4 9 1 3 50 Tlotals Pr. Pts Biface Rework Scraper Biface Retouch Used Cores Flakes Flakes Debitage 1-c 2-c 15-c 3-c 266-c 5-o 3-o 1-o 2-o 50-o 1109-o 6 2 3 1 2 65 3 1375 51 Site Son-593 (Treganza 3 & 4), Unit 2 Animal Bone 2 28 93 67 109 135 91 41 12 8 7 3 HI storical Material 145 nails 4 nails 2 5 1 3 Total 2 Level cmn Pestles 2 Fired Rock 50 90 80 40 20 25 75 30 38 11 3 1 14 22 7 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 190-200 200-210 210-220 2 508 607 149 52 two pestles at 10-20 an aid a Gunther Barbed projectile point at 60-70 an. The latter indicates a very late time period (perhaps 1200 to 1300 AD) so that it may well be out of place at that depth. This represents the latest phase of this part of the project aid correspords to the entire deposit of the south midden. The next unit (C anx] D of the profile) is fran 70 to 110 an arn consists of dark brown silty midden with much clay anx] sand and is sealed at the top by a layer of burned and compact soil which may have been living surface. There are no diagnostic artifacts in this level but the chert-obsidian ratio increases markedly. We feel this is a component of an earlier phase. The third level exterds fram 110 to 180 an (level E of profile) and is like the previous level but is more canpact and has a larger component of gravel. Again there are no diagnostic artifacts but the chert-obsidian ratio increases to the point that obsidian nearly disappears altogether. This, and possibly the fourth level below, we assign to yet earlier phases. The lowest level, fran 180 to 210 cm (level E on profile) consists of brown sandy gravel. Cultural renains consist of a few chert flakes and some animal bone. Augering of this level irdicated only 10 an more of this material before refusal in very coarse yellow brown gravel. In addition to the above we should point out that the area around Site Son-593 has a considerable history of Euro-American habitation. According to the Historical Atlas Map of Sonana County published by H. Thanpson aid Co. in 1877 there were at least three homesteads in the immediate area. The Scott District School was located very close to Son-593 ard may have been the source of the nails and other historic materials in the north midden. Site Son-579 (Henry Moore) Son-579 is a site with a very sparse surface scatter or artifacts over a 20 by 80 meter area. Associated with this site aid 50 meters down Dry Creek fran it, are two petroglyph rocks. During the survey of this area four auger holes were made in the area of the flake scatter aid close to the terrace bank where dark soil was seen under 15 to 20 an of overburden. The augerings showed 10 to 25 an of very dark grey brown silt which was considered midden. A unit was excavated here to determine if this buried dark soil was midden, but no artifacts were found below the plow zone. The site is located on the north side of Dry Creek on a secordary terrace that is 2 m above and 5 m north of the gravel strean bed. The strean bed is 50 m. wide at this point; the unusual width is probably the result of recent erosion that washed away the primary terrace that was here, as shown by a remnant of the primary terrace remaining on the stream bed. Local residents report that a site rich in artifacts was in this general area and was located on or next to a "sand bar". The sani bar referred to may have been the primary terrace now eroded away anx the surface scatter above may be a peripheral remnant of the site. The soil of the site is a 2 to 3 m alluvial deposit of silt anx sand underlain by a .5 to 1 m boulder bed (i.e. very course strean gravel with many boulders) ard finally by shale bedrock. The site is now in open 53 grassland with riperian vegetation along the front edge of the terrace. This vegetation includes laurel, live oak, Oregon oak, willow, ard wild grape. There are also two stands of sedges, the kind used by the Pano in basketry, along the front edge of the terrace near the site. The petroglyphs, which are on two sandstone boulders now located on the gravel stream bed, probably were once on the now-vanished terrace. These boulders have been subjected to heavy stream erosion and some of the petroglyphs are now very worn arKn faint. The petroglyph motifs are unique for the project area ard consist of pecked circular grooves and cupules--at least one is a cupule within a pecked circle. Site Son-578 (Sniffles) Son-578 is located on the south bank of Dry Creek arnd 300 m across Dry Creek from Hot Springs Road. It is approximately 900 m east of the east entrance of Hot Springs Ranch on Hot Springs Road. It is a weakly formed midden site on the west edge of a large flat. This flat is bordered on the north by Dry Creek ard on the south by an unnamed intermittent tributary. The flat is triangular in shape ard is canpletely surrounded by sheep stake fencing. The flat and most of the site have been disced. Directly south of the intermittent stream is a steep hillside covered with mixed oak/conifer vegetation. Directly north of the flat (between Dry Creek and Hot Springs Road) is a first terrace remnant arn] Son-579 (Henry Moore), a surface site with associated petroglyph. The area of the site is open grasslarnd with riparian vegetation along both the intermittent stream ard Dry Creek. A large bay laurel is located in the middle of the south edge of the midden area ard live oaks are found bordering both water sources. An area of sedge, the type used by the Pano for basketry, is located along the eastern boundry of this flat and is bisected by the stake fencing. This sedge growth encompasses a 30 by 5 m area, 80 m east of the midden. The soil of Son-578 is very dark brown sandy gravel. The limits of the midden are defined by the dark brown/brown alluvial sand of the surrounding areas. A series of four auger holes within arK outside the midden area confirmed the extent of the midden as an area measuring 60 m by 40 m ar] limited to the south west edge of the flat. Test Unit 1 was placed in what was thought to be the deepest section of the midden, 21 m south and 20 m west of the site datum stake. This unit is south of the sheep stake fencing arn north of the intermittent stream. Excavation of Test Unit 1 revealed a shallow, weakly formed midden overlying sterile gravels. The midden exterded in depth fran 0 to 30 m arn consisted of dark grayish brown silt. No projectile points or ground stone tools were recovered. Among the few artifacts fran the midden levels were a reworked biface of chert and one retouched chert flake. No animal bone was found ard there was little fire-cracked rock. Charcoal flecks ard ash were scattered lightly throughout the midden. Gravels began to intrude in the 10- 20 cm level arK] became more abundant with increased depth. The final excavated level, 30-40 an, consisted canpletely of gravels ard roots. Upon 54 canpietion of this level, the center of the unit was augered to a depth of 90 an and found to be devoid of midden. Site Son-578 (Sniffles) Level Rework Retouch Used Debitage Fired an. Biface Flakes Flakes Fock 0-10 l-c 5-c 14 4-o 10-20 1-c 3-c 32 1-c 20-30 1-c 1-c Site Son-609 (2 Cups) Son-609 is located approximately 1.4 miles northwest of Cherry Creek Bridge, and approximately 25 meters south of Hot Springs Road. It is on the edge of the north bank of Dry Creek ten meters east of a garbage dump. The site is a petroglyph site consisting of one rock with 2 large cupules on its upper surface and is about 300 meters upstrean fran Son-607, a petroglyph site of several boulders with associated surface finds. Son-609 is located in a riparian envirorunent on brown alluvial sards. Augering was not possible at this site. It is of interest that this rock is located near an historic dump. It is suggested that this dump be excavated in the future for possible proto-historic material. This site as well as Son-607 (petroglyph) and the nearby habitation sites may be important sites for understarding the purpose and location of petroglyph rocks anx may also help suggest dates for petroglyph decoration. Site Son-581 (Madrone Point) Son-581 is located on a first terrace on the south bank of Dry Creek at the confluence of an intermittent stream and that Creek. This terrace is approximately 300 m south of Hot Springs Road anx 150 m south and upslope of Dry Creek. It is canpletely sheltered on the north, south aid west sides by madrones, oaks, anx conifers. An old road bed runs across the entire east side of the promontory at the base of a grassy slope that has been heavily logged anx] has tree stumps scattered on its surface. The site itself has been logged anx at present has a fallen log on its northern edge. A chert scatter, including a chert scraper, was found during the survey phase so the site was augered to test for possible midden. The soil of this pramontory consists of red-brown gravely loam. Five auger holes were placed at the tip of the pranontory and on its neck and no buried middens or other cultural deposits of any kind were found. Thus this is evidently a strictly surface deposit anx future work may be confined to surface collection. 55 Site Son-582 (Cherry Creek) Site Son-582 lies on a promontory just back fran the confluence of Cherry Creek arnd Dry Creek, about 100 m NW of the former ard 80 m NE of the latter. The site consists of two separate midden deposits lying on either side of a gully which carries water only in wet weather. The midden deposit on the east measures 80 m by 30 m while the smaller one on the west is about 30 m in diameter. The site lies on the secornd terrace above Dry Creek. The second terrace here does not necessarily correspond to that on which other sites nearby are located--most of these occur 2-5 m above the stream bed while this one is at about 15 m. The soil is typical terrace soil with fine textured sediments on top grading into coarser sediments below. The fine textured sediments run 50 to 100 an in depth before reaching gravel. Thus the soil is well drained. Two test units were excavated in Son-582. Unit 1 was located on the SE edge of the site and was dug to a depth of 90 an. The soil there is dark brown, friable loam only marginally classifiable as midden. The evidence of human occupation consists of a single projectile point plus a few chert ard obsidian flakes. Unit 2 was placed near the center of the site ard proved considerably richer. It was excavated to a depth of 110 an. Down to 80 an it was black, sandy midden with considerable quantities of pebbles, cobbles, arnd angular rock. As irndicated in the table there were large quantities of cultural material in this level. Two of the projectile points recovered are of the type called Excelsior (see Fig. 24 a, b). One of them was in the 20-30 an layer while the other was fram 50 to 60 an. Below 80 cm a change is indicated by an increase in alrge, percussion flaked artifacts and debitage. It is also notable that the ratio of chert to obsidian changed markedly at that level. In addition to this it is the case that animal bone ard fired rock no longer occur below that level arn it therefore seems likely that the deposit below 80 an represents a different cultural component fran that above. Map 8 shows Son-582 to be a central site with smaller sites arn isolated housepits surrounding it. This, together with suggestions of its suitability as a winter village, raises the possibility that it was a headquarters site for the surrounding territory. In that case one would expect that it contained many houses, one or more sweat houses, arn a dance house during the period of its prime occupancy. This clearly is an important site ard future excavation should have as its goal (1) an estimate of the size ard number of dwellings at any given time and (2) identification of ceremonial structures. Let us suppose we take a sample of 5% of the site. There is a total of 1700 sq. m so the sample would be 85 one meter units. For the dance house let us assume a diameter of 10 m, giving an area of 79 sq. m or 4.6% of the total site area. If the 85 units were located at randmn then the probability of at least one being inside the dance house area would be the same as one minus the pro bility of all of them being located outside the dance house or 1-(.954) . This 56 CA* Son 582 Map 0 10 20 Meters 9m~ 0 I ~ o ; - A-, "7vAt h ',,Al .&. Map 10. Son-582. 0 I / 0 JN 57 10 -20- -30 -40- - 50~ - 60 ... -10-~~~~~~~~~~~~~~~~~~~~~~~~~~~ C ., OOO() C 0000 (C 00)0000000 (0000 0 0000000rC 00 00(COO 0 0000000000 (0 -)()() 8() (!0( r) ) ())()(()_)(OCO 00C CCOO 0000 COC O( 000(0O OC ) 00 0 00 000 00((C )() () OOCOOOO OOC (, C (O)() (0 C 00 (C 0( 0 0( C( 0 0 0000CoO00000000 ()) (0CC ((( () COC) 0( () 00(C) ()(000C C 00 C0 C O C OCC ( 00000 0 CCOC C) CO ) _fl 000000() )CC! OOC)OCC C.)OC()000CCCCOOOCOCCOOCOO CO - 00000CCvOCC)0000000 00000000 00o)O( O ( C'( ')( (;)( (( ( )~ C ( )C C C (C CO O OOO O O C O O O 0 () C C O O 000 OO )0000 (C C CC 000000~~~()CO 000C ()C oC00( C) C 0 00000C(C 00000 ( )000000000000(00 0000 C ( 0 CC 0 0()() ( )() () ( O (0 (C C ()C 0 C f (C C ) () (CC OC) C 00000 o ()(0)0 O 90 000000 00(JOOOOO;)0CC C0C0(C 0000C0 00 ( C 0000OC)CC)OOO CCC )(0)OO(0OO0)0000000000000000000 )t7__ 00()0 0) I () 0 M7Roo C(C C C)oCoo C C) OOOn OOCCC(CCOOOOC C 00 O o000CC 0000000000)000(0 D ))(f(()()()()0( (,()()(()()()()L()(()()()()((O ()()O(O~i ()OOCOCOfr(C)O >)(O(())00(~)(C)O)OfDOOCOOOOOOC)OCvOf))f O()f 0 0 0 0 0 C )((1)()O)(O((,(() C) C0(" ()00 000 (C00(0CC()( CCC(CCCC0 C(0000 C,,) 1flfl CCCCCCCCCCCCCC0CC)CC ((((CrOCOC ()CCCC O) (000CC OCOOOOOOOC OC)00COCCOOCOOOOCOCOOCCOfD00(0 CC) CCC C'((C( CCC (CCC000000CC( C( C)( CCC 0 000( CCC0(()OO)000000COO CC) oC, CC( C CCC(C(C'C CCC O C)C(CCCCCo C 00C C ((00C C (O0000 0 0000C 0 0)oC )000( 00000 00C(OrooocC )O(( -((CCC 7CCC0()0CC(C' ((\(CCCC(0f )(0 000f0000 (00( _ 00()(00o -80 ) ( I0 ) 0 ,(0( C) C )(0 r ) 0C) r0()0 COO C 0 000 ( 0)000C)O0of) 0 0000() 000000o000 )Oo()(C((o r)Cof ) o)() O( 0 cm A. Midden; dark brown; granular loam with gravels; annual grass roots; heavy grazing caused surface compaction. Moderately friable. B. Midden; silty loam; dark brown; friable. C. Gravels and sand with cobbles; dark yellowish brown; friable. Figure 9. Son-608 Profile. qw f r q '-I qw -i 0n (N 0 N 10 %D %O r- Ln lr _I O0 1 1 NO O co C O N co 0) 0 Un o% II4 0 0 CN' u00 1 1 r-q D LO CN -4 0) 0 Q 0 0 0 00 0 .4 rI r1 -4L 0 0) (N4 0 0 u- 0 v-I 0 '-I 0 U- 0 (N FI Q r- 0 m 0 (N 0 4- 0 Co 1n' Pb 66 co 0 0 ~ a r8 .9 E4- a; DW U1) ._lI U) .rN; 8 Tq ? ? 0 co aO 0) O rl, I 0 10 0 i- I 49 m I 0 U) CO (a r1f u ? LO CM Ct) N (C U-4 U-Il v-I (N qwj Lrn (N N Ltn rfn COD UN o 0 F- u00 (N c)o tQ w -I w *,-l 0 r;. 0 (I *u- 67 erdemic to this area. Rather, it has been noted as beirn common in the Clear Lake Region of Lake County, one of the sources of obsidian used by the local Pono, suggesting that this basalt was obtained at the same time as the obsidian fran that area. The basal canponent at Sot-608 is nearly sterile in artifactual content. For this reason, only 10 am were excavated into the sandy clay. The bottom of this unit was then augered to a final depth of 150 an. The 70 an core consisted of water-worn gravelly sand and contained no artifacts of any type. No diagnostic artifacts were recovered fran the entire depth of Test Unit 1. However, several surface finds were previously collected by Alf and Ann Young, owners. These artifacts have been drawn and photographed aid include five Excelsior and leaf-shaped points aid point fragments, and a steatite bead. It should be noted that a small unworked piece of steatite was also recovered in the first level excavated in Test Unit 1. Site Son-597 (Treganza 7) Son-597 is located on the first terrace at the confluence of Yorty and Brush Creeks. The site is divided into two sections. The main part of the site is bordered by Hot Springs Road aid Yorty Creek to the north aid by Brush Creek to the south. The secoid, smaller section of the site is delimited by Brush Creek to the north aid by the old stream bed to Brush Creek aid Kelly Road to the south. The present course of Brush Creek has destroyed at least a five meter wide strip through the original site. The area of the main section measures approximately 50 by 30 meters aid includes most of the promontory left by the convergence of the two creeks. This promontory has open grassland and coverage, with oaks aid bay laurels bordering Brush Creek. Wild irises are found on both the eastern aid western extremities of the site. A faint jeep trail passes through the center of the main section. The midden consists of a lower section of a knoll one aid a half meters high. On the knoll, two housepit depressions are visible. The larger definite housepit is located on the western part of the knoll, just south of the site datum, aid measures approximately 5.5 meters in diameter. The smaller, less distinct housepit is located two meters east of the large housepit aid site datum aid measures approximately 1.6 meters in diameter. Test Unit 1 was placed 15 meters south of the site datum stake to investigate the area near the housepits. Excavation of Test Unit 1 revealed three stratigraphic components consisting of dark midden, a transitional zone, aid sub-midden levels. The soil from 0-50 an ranged from black to dark brown and contained the greatest amount of cultural material. Animal bone, fire-cracked rock aid obsidian counts reached their peak in the 30-40 an level, aid chert steadily increased to reach its highest quantity in the 40-50 an level. At their highest points, at least five times as much chert was recovered as obsidian. In this stratigraphic component two possible Excelsior bases (obsidian aid quartz) were recovered as well as one triangular straight base chert projectile point. 68 CA* Son * 597 Map Mp 12. Son-597 (legend p. 5). 444- 1 / , N& -o- -10 - -20- 30 "40- 10 -,00 -so- - go- H10- -12ir -13- -150 c1m1- cm A. Midden; annual grass roots; ard clay; very dark gray to brown. Moderately compact. silt with sand very dark grayish B. Midden; brown to dark brown; clayey silt. Moderately compact. C. Transitional zone. Dark yellowish brown; silt with sand and clay; small gravels. Moderately compact. D. Sub-midden. Yellowish brown; sand and silt; small gravels. clay with some Compact. E. Dark yellowish brown; clay; no gravels. Compact to very campact. F. Gravels and shale. Friable . Son-597 Profile. 69 Figure 10 - Level Proj can. Pts. 0-10 10-20 1-o 20-30 30-40 1-c 40-50 1-c 50-60 60-70 1-c 70-80 80-90 90-100 100-110 1-c . B iface 1-c 110-120 120-130 130-140 140-150 150-160. Site Son-597 (Treganza Scraper Used Debitage Flakes 1-c 1-c 142-c 2-o 41-o 1-c 63-c 2-o 26-o 1-c 37-c 19-o 3-c 87-c 1-o 24-o 2-c 131-c 15-o 2-c 121-c 5-o 3-c 81-c 5-o 1-c 43-c 1-o 2-o 2-c 59-c 2-o 1-c 33-c 1-o 2-c 22-c 1-c 2-c 1-o 6-c 2-o 9-c 1-o 17-c 11-c 1-o 5-c 70 Animal Bone 4 4 2 5 7) Fired Rock 165 77 20 40 24 10 12 5 3 6 - Totals 4-c 1-o Grard 5 Total 1-c 1 1-c 1 22-c 867-c 6-o 145-o 28 1012 71 362 15 362 15 72 In the 50-100 an levels the soil gradually changed fran brown to dark yellow brown and cultural material decreased in quantity. A few gravels began to permeate the otherwise clayey soil in the 80-100 an levels. The only tools found in these levels were projectile point tip and one bifacially retouched chert knife tip. The 100-150 an levels again showed a lightening in soil color, changing fram dark yellow brown to brown. In this sub-midden canponent a heavy gravel lens intruded at the 120-140 an levels, after which gravels nearly disappear. Also, at 120-130 an the clay content of the soil greatly decreased perhaps indicating a division within the component. Animal bone, fire-cracked rock and obsidian all but disappear in the gravel intrusion. Chert, on the other hand, seems to peak briefly at the end of the gravels and continues to predominate down to the 160 an. level. Unfortunately, diagnostic tools of any type are absent fran this sub-midden component. The last excavated level, 160-170 an, was sterile. Site Son-592 (Treganza 2) Son-592 is located on a primary terrace on the east bank of Dry Creek, approximately 220 meters south of the confluence of Yorty and Dry Creeks. It is a surface site located approximately 300 m upstream fran Son-541, another surface site on the west bank. This site was relocated using Treganza's San Francisco State College Survey notes during the survey phase of the Warm Springs Archaeology Project. The site was listed as a surface site located on a terrace with open grassland cover and consisting of a sparse scatter of chert and obsidian flakes covering an area measuring 40 by 20 meters. The alluvial silt covering this terrace remained consistent in the area of surface scatter and in the surrounding areas. A series of five auger holes were placed on this terrace to test the site for possible subsurface midden. The holes were placed 20 meters south of site datum; 20, 40, and 80 meters north of datum and 20 meters west of datum. None of the five auger holes revealed sub-surface midden of any kind. This site remains then, a surface site which has been greatly disturbed by both modern activity and erosional activities of Dry Creek. The site still has evidence of discing and five apple trees are located approximately 60 meters due east of site datum. A modern wire fence runs perpendicular to Hot Springs Road 2 meters west of these apple trees and forms the eastern boundary of this site. Just west of the apple trees and the wire fence are sheep pens enclosed by wire fencing. A gate leading onto the terrace is located approximately 10 meters north of the sheep pens. A second wire fence runs parallel to the creek and is located on the cut bank of that creek. This fence forms the western boundary of this site. A small intermittent tributary runs parallel to Hot Springs Road, connecting the two wire fences and forming the northern boundary of Son-592. Oak trees line both this intermittent stream and Dry Creek, just west of the second wire fence. Site Son-541 (Terry's Terrace) Son-541 is located on a second terrace on the west side of Dry Creek, approximately 120 meters downstream and southwest of the confluence of Yorty 73 and Dry Creeks. It is located on the part of Dry Creek that connects the Upper Dry Creek drainage with the Lower Dry Creek drainage and is just downstream fran Son-592, another surface site on the east bank of this same central section of Dry Creek. This site has open grassland cover with isolated oaks and willows both on the site and on the bank of the terrace above Dry Creek. Dry Creek curves around the site and forms the northern and eastern boundaries of this terrace. A jeep trail runs parallel to Dry Creek (eastern part), bisecting this terrace. This jeep trail forms the western boundary of Son-541, and probably did some damage to the site during its construction. A further indication of modern disturbance is evidenced by a small corral located approximately 5 m northwest of site datum. During survey the site was located and found to contain both midden indications and a housepit, approximately 2 meters in diameter. A scatter of artifacts enclosed an area 10 meters square around the housepit a]d midden area. A 20 meter square area on the bank of this terrace seemed to have a concentration of chert flakes. A series of seven auger holes were placed on this terrace to test the extent and depth of the midden canponent. It was found that only one hole (C) contained cultural material and flakes fran this auger testing only extended in depth to 20 cm. The rest of the terrace seens to have only surface scatter. Eastern Group Chronology We follow procedures similar to that of the Western Group here and in addition we have two radiocarbon determinations here which we evaluate in discussing the following site. Site Son-593 We evaluate the two sections of this site separately. The south, actually southwest, midden to judge fran the one small side notched projectile point (Fig. 24c) and fram the well-preserved housepits on the surface is very late and we attribute it all to the late Houx Aspect. The north or northeast midden is more canplicated. There are two breaks in the chert-obsidian ratios, one at 70 an and the other at 110 cm depths. These two points also coincide with changes in the character of the midden (see profile). The top 70 cm has the following artifacts: a Gunther point at 50-60 (Fig. 24f), a large Excelsior point at 30-40 (Fig. 24e), and 2 large pestle fragments in the 10-20 an level. The Gunther point, which indicates late Houx, is thus below the large Excelsior point which would be early Houx at latest; thus one of the two is out of place. We are inclined to attribute the top 70 cm to late Houx (we were so inclined even before our radiocarbon determinations came in). This puts the 70 to 110 cm level in early Houx and material below that in the Borax Lake Aspect. We have two radiocarbon dates fram this site as follows: UCR-352 Charcoal lump fran the 90-100 an level 1710 + 150 AD 200 UCR-351 Charcoal fran hearth in 160-170 an level 4720 + 240 2770 BC 74 Thus the 160-170 cm hearth is on the borderline between early and late Borax Lake and we therefore feel the very bottom of the site (but we don' t know how much) represents early Borax Lake Aspect. If this analysis is correct then all four of Fredrickson's late phases are present in this site. Incidentally these two radiocarbon dates corroborate or at least do not conflict with, Fredrickson's scheme. Site Son-582 Both the chert-obsidian ratios anx] other changes in chipping waste (see site description) indicate a possible cultural change at the 80 cm level. The upper 80 cm seems definitely attributable to the late Borax Lake Aspect on the basis of the two large Excelsior points (Fig. 24a, b) and the two possible milling stone fragments. We therefore attribute the bottom portion to early Borax Lake. Site Son-608 There is no noticeable break in chert-obsidian ratios nor any in the midden at this site. The excavated portion therefore seems to be one component which, on the basis of large Excelsior points fran the surface we attribute to late Borax Lake. The surface finds mentioned in the site description suggest that the site may have other components distinct from this. Site Son-597 There is a break in debitage ratio and also in the midden at 50 an depth. On the basis of the single large Excelsior point in the 40-50 an level (Fig. 24i) we attribute the top 50 an to late Borax Lake and therefore the bottom to early Borax Lake. 75 Eastern Site Group Chert to Obsidian Ratios of Debitage (by period) Period 593-1 593-2 Site CA-Son 582 608 597 0-50 0-70 Hb 0.61 0.12 (79) (1043) 70-110 Ha 0.50 (257) 110-180 0-80 0-80 0-50 BLb 7.50 1.58 1.40 4.47 (51) (867) (1122) (711) 180-220 80-110 50-160 BLa 7.00 4.00 19.07 (24) (60) (301) Eastern Site Group Chert - Obsidian Debitage Ratios by Site and Level Site CA-Son 593-1 593-2 582 608 597 0.15(82) 0.18(84) 0.16(142) 0.10(138) 0.07(190) 0.12(213) 0.13(174 0.36(95) 0.70(73) 0.38(51) 0.71(41) 12.00(13) 2.67(11) 8.00(9) * (2) * (3) 4.00(5) * (8) 2.67(11) * (4) * (9) * (5) 1.03(59) 1.43(114) 1.25(79) 1.30(115) 2.14(166) 1.17(117) 1.95(121) 2.00(75) 3.09 (45) 1.29 (135) 1.59(149) 1.15(172) 1.44(212) 1.62(254) 1.64(116) 1.96(77) 0.17(7) 13.00(14) * (1) 3.46(183) 2.42(89) 1.95(56) 3.63(111) 8.73(146) 24.20(126) 16.20(86) 21.50(45) 29.50(61) 33.00(34) 22.00(23) 3.00(8) 9.00(10) * (17) 11.00(12) * (5) 76 Level an. 1.00(24) 0.64(36) 0.17(14) 0.33(4) * (1) 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 190-200 200-210 210-220 77 ~~~~~~~I.. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ a ( 0 ? .. ......w U) (0 I ~ ~~ ~.. .J... CD.gll .::::~'l * ..... ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ........ * ... ~~~~~~~~~~~~~CD ................. U) ........... .. ....... . I- U) aCL~~~~~~~~~~~U C~~~~~~~~~C 0 00 a~~~~~~ 0 0~~~~~~~~ Map 13. m0 t I 79 Yorty Creek Group Description This small group of seven sites is one of considerable importance - it has deep gravel canponents which potentially are of great antiquity. Unfortunately we have been unable to date them as yet; there is no charcoal fran them arx] geologists have not studied them intensively. Yorty Creek is a minor tributary of Dry Creek and therefore the valley it has excavated is narrower arn less suitable to habitation. It widens slightly where site Son- 576 is located arn] then again upstream at Son-584 and 583. The reconstructed vegetation here is virtually all woodlard-grass or woodland- chaparral, mostly the latter (Map 13). This presumably makes this prime lard for deer but no specialization toward hunting is evident in the archaeology. We again find the situation that the total area is so small that topography may be the most important element of the immediate environment. Site Son-585 (Yorty Cupule Rock) Son-585 is located in the creek bed of Yorty Creek on a gravel bar, approximately 20 meters north of Hot Springs Road. It is about 150 meters west of the sheep shearing station and pens near the midden site of Son-576. Both these sites seem to be isolated in the central section of the Yorty Creek drainage that is encanpassed in the Project Area. Son-585 is a petroglyph site consisting of a single large rock with approximately 98 pecked cupules on its surface ard sides. A great amount of stream erosion has taken place in the area of this boulder. It is located on the very tip of a small gravel bar in the center of Yorty Creek, which reaches a maximum width of 4-5 meters arK externds upstream approximately 25 meters. Yorty Creek runs to the north of this bar. The terrace to the south arK east of the cupule rock has been eroded to three separate levels. No midden is apparent in the cut banks. During the field reconnaissance phase, a chert biface fragment was found just west of the petroglyph in the stream gravels. Two large chert cores were also found both upstream and downstream fran the petroglyph. At that time a series of five auger holes were placed on each of the first terraces around the petroglyph. No midden or cultural material of any type was recovered fran this testing. Site Son-576 (Sheep Shearing Station) Son-576 is located on the north bank of Yorty Creek, 120 meters north of Hot Springs Road. It covers the lower terrace of a pranontory formed by the erosive and depositional patterns of Yorty Creek. The site is bisected by an unnamed intermittent tributary which drains west of this main pranontory. The midden is on a first terrace which slopes steeply down to a gravel bed at the foot of the promontory. A steep stream cut bank is located on the south side of Yorty Creek arK is bordered by a wire fence. Just west of the promontory Yorty Creek runs sharply south, towards Hot Springs Road. On its western bank a wooden fence encloses a sheep shearing station with several corrals arK out buildings (50 meters south of the site). Between this sheep station ard Hot Springs Road, CA-Son- 576 Map pq 9 0 0 0 0 * 120 Meters To Hot Springs Road 0 0 a 0 0 5 10 M er - Meters (I Map 14. Son-576 (legend p. 5). 80 ShepBarn 92 a I--- 92 81 Yorty Creek swings along an east/west line again. A bridge crosses the creek at this point and allows a dirt road to connect the sheep station and Hot Springs Road. The area of Son-576 has open parkland vegetation with scattered clusters of live oaks. It is-> notable that in this area there is a preponderance of valley oaks found around the central section of the Yorty Creek drainage. The midden consists of black gravelly loam and can be visually distinguished fran the dark tan gravelly loam of the surrounding areas. A series of 19 auger holes were placed both near Son-576 and in the neighboring fields to test the extent of the known midden and to explore for areas of buried midden. Only one of these borings definitely produced midden; an auger hole located just west of the intermittent drainage and approximately 2 meters upslope fran Yorty Creek had evidence of midden. Fran 32-102 an fire-cracked rocks and heavy charcoal concentrations were noted iidicatirg well-formed midden deposits within those depths. However, those placed west and north of this proved to contain only alluvium with slight traces of charcoal. Thus, it is known that the intermittent stream bisects the midden, but that this midden does not extend as far west as a second seasonal drainage (15 to 30 m west of the first drainage). Test Unit 1 was placed 3 meters south of the site datum stake, in the center of the promontory. Excavation of this unit revealed that 20-30 cm of recent overburden covers this promontory. The first 20 cm of soil is red/brown and shows signs of disturbance. At the 20 an level a transitional zone is apparent. The next ten centimeters seem to be basically midden-like with some mottling aid rodent activity. Fran the recent aid transitional levels only one tool was recovered, a small leaf-shaped green chert projectile point dated to a late period. Moderate amounts of lithic debitage, animal bone aid fire-cracked rock were also found. A well-developed midden deposit underlies the recent overburden (30-140 cm). This midden has been stratigraphically divided into two different zones on the basis of the clay content of the soil, which greatly increased with depth. However, both zones consisted of black, greasy silt and differed in color only by one half of a chrana. Heavy concentrations of gravelly clay began to appear at 130 Ac; by the next level the entire eastern half of the unit consisted of this material; aid by 150-160 an gravelly clays dominate the whole unit, with only small pockets of midden. Thus, 140-180 an has been called a transitional zone aid represents the bottom of the second midden component. No diagnostic tools were recovered fram these levels, but two separate hearths of some size were found. Feature I measured 25 a by 32 an, was located in the southwest corner of the unit, and was surrounded by several fire-cracked rocks. The lowest depth of this fire. pit was 165 an. Feature II, the larger of the two hearths, measured 33 an by 38 mn. It was found in the center of the east wall and consisted of many fire cracked rocks although it was not surrounded by them. The bottom of this pit was at a depth of 175 an. Fire unutilized chert aid obsidian pieces came fran this second hearth as well as six pieces of burnt bone. A charcoal sample fran this hearth was submitted for radiocarbon dating. The last component at Son-576 consists of gravelly clays aid is nearly 82 ~0 - -, -30- -40- - 50 -60- -so - ~90 o- A. 0-wa aidn aap s deois dar reddish -170 - monsoon aI brown; granular loam. Friable. B. Midden I; black, greasy loam with small angular gravels. Friable. C. Midden II; very dark grey; clayey loam with small angular gravels. Moderately friable to moderately campact,. D. Transitional zone. Brown to dark brown; clay with subangular gravels and sand; shale and pebbles. Compact. E. Rcdent disturbance; krotovina. Figure 11. Son-576 Profile. 83 Site Son-576 (Sheep Shearing Station) Level Proj. Bifaces Retouch Scrapers Retouch Used Cores Debitage cm. Pts. Biface Flakes Flakes 0-10 2-c 47-c 1-o 7-o 10-20 1-c 1-c 17-c 1-o 10-o 20-30 1-c 44-c 1-o 12-o 30-40 1-c 3-c 60-c 1-o 14-o 40-50 3-c 3-c 57-c 12-o 50-60 1-c 1-c 1-c 47-c 23-o 60-70 2-c 1-c 1-c 2-c 31-c 1-o 2-o 16-o 70-80 1-c 20-c 1-o 1-o 23-o 80-90 1-c 27-c 27-o 90-100 2-c 53-c 3-o 7-o Level Proj. CM. Pts. 100-110 1-o 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 Site Son-576 (Sheep Shearing Station) cont. Bifaces Rework Scrapers Retouch Used Cores Biface Flakes Flakes 1-c 2-c 1-o 2-c 2-c 2-c 1-o 1-o 2-o 1-o 1-o 3-o 1-c 1-o 1-o 1-c Debitage 5-c 1-o 18-c 37-o 36-c 51-o 20-c 26-o 10-c 29-o 9-c 9-o 8-c 3-o 2-c 1-c 3-o Total 1-c 1-c 2-c 3-c 24-c 8-c 512-c 3-o 2-o 1-o 19-o 310-o Grard 4 1 2 2 4 43 8 822 Total 84 85 Site Son-576 (Sheep Shearing Station) Level Pestle Fired Animal Historical cm. Rock Bone Material 0-10 30 1 shell casing 10-20 40 10 20-30 18 30-40 1 200 21 40-50 200 31 50-60 70 15 60-70 70 45 70-80 100 21 80-90 130 41 90-100 130 61 100-110 145 78 110-120 90 45 120-130 110 76 130-140 40 72 140-150 20 36 150-160 15 20 160-170 20 37 170-180 7 8 180-190 1 2 86 sterile. There were no tools among the few specimens recovered here, only debitage, fire rock, animal bone, an] one used flake. Heavy rodent activity in the two levels excavated into the clay obscured the unit and made it impossible to tell if the debitage and other items were intrusive. At two meters excavation of this unit ceased. An auger boring was made to the water table at 270 cm. The auger showed that charcoal flecks were present at least to a depth of 265 an. Site Son-588 (Live Oak Housepit) Son-588 is located on a low secondary terrace on the south bank of Yorty Creek, approximately 1/2 mile from where the eastern edge of the Project Area boundary crosses Yorty Creek. It is on a promontory formed by the confluence of Yorty Creek and an unnamed intermittent stream. It is also approximately 150 meters southwest of site Son-577. The site is a housepit site consisting of a well-defined housepit with a large live oak growing fram the back wall. It is located on a gently sloping grassy slope that has evidence of deforestation. This housepit measures 8 meters by 7 meters by 1.5 meters in depth, and has a distinct square entrance off its northwestern side. A chert scraper, an obsidian flake, a chert flake, 3 square nails, window glass fragments, 1 wire nail, 44 caliber shell casings, an old ceramic bottle cap and window glass pieces were all found on the surface of the housepit. Trowel exploration of this surface and augering around the housepit revealed no evidence of midden. This site is extremely interesting because of its close proximity to Son- 577 which has no housepits but is a deep midden site. It undoubtedly represents a house of very late, probably historic or even late historic (1850's?) date. It is a possible indication of Indian presence in the North Section into the American period, a presence for which there is no historic or ethnographic evidence. Site Son-577 (Touchdown) Son-577 is located on the first alluvial terrace on the west bank of Yorty Creek, east of Hot Springs Road. A large turnout area adjacent to Hot Springs Road is about equal in size to the area of the site and entirely covers the western section of Son-577. This site has also suffered fran intense stream erosion along its entire eastern perimeter. During survey, chert debitage was found approximately 15 m northeast of the main site area on the opposite stream bank in the gravels. It was thought that this cultural material may have eroded out of the eastern bank of Yorty Creek and that the creek actually bisected Son-577. Erosion by Yorty Creek in this area is so intense that these artifacts have now disappeared. The site has open grassland cover with a sparse scattering of oaks and bay laurels on its eastern edge, along Yorty Creek. The midden of Son-577 is a black sandy loam. The extent of this midden is delineated by the dark brown alluvium of the surrounding areas. Two auger holes placed north and south of the midden area revealed no buried midden components, thus, delimiting the area of midden to a 20 m by 40 m section of the terrace. 87 Map 15. Son-577 (legend p. 5). 88 A. Midden; very dark gray; loam with angular pebbles and small cobbles. Friable. B. Dark grayish brown; coarse sandy gravels with clay; rounded pebbles and cobbles. Friable to ccipact. C. Dark brown to dark yellowish brown; sandy clay; friable. D. Krotovina disturbance. E. Ash and charcoal deposit. F. Dark grayish brown; sand without gravels; friable. Figure 12. Son-577 Profile. 89 Site Son-577 (Touchdown) Level Biface' Scraper Retouch cm. Flakes Used Core Flakes Debitage Fired Animal Rock Bone 0-10 1-o 10-20 1-o 2-c 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-10 0 1-c 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 1-c 2-c 2-c 2-c 3-c 1-c 1-c 5-c 1-c 1-c 1-c 4-c 7-c 4-c 4-c 1-c 33-c 6-o 2-c 47-c 3-o 47-c 2-o 21-c 1-o 15-c 13-c 5-c 6-c 5-c 1-c 7-c 1-c 3-c 9-c 9-c 6-c 1-c 12-c 3-c 1-c 1-c Totals 1-c 2-c 7-c 33-c 5-c 243-c 325 35 1-o 1-o 12-o 2 Grard 1 Totals 8 34 5 255 325 35 50 1 75 3 90 20 ? 7 4 60 30 10 10 90 No housepits or other cultural features are evident on this terrace. The test Unit was, therefore, placed in the center of the midden concentration 2 m east of the site datum stake, on the north side of the east/west transect passing through the datum. Excavation of Test Unit 1 revealed a well-developed but shallow midden overlying more than a meter of gravels with cultural material. The soil to about 30 an was a dark grayish-brown midden with subtle lighter mottling. Animal bone, fire-cracked rocks and chert all reached their highest quantities in this canponent. Obsidian artifacts were rare. No projectile points or ground stone tools were recovered fran Test Unit 1. In levels 30-50 an mottling by lighter soils became much more distinct and prevalent. Since midden pockets were still continuing into the mottled soil, these two 10 an levels were called transitional. However, it is now clear fran the stratigraphic profile that much of the mottling in this canponent and below 50 cm was due to krotovinas and an isolated sari intrusion. Thus, 30-80 an should be considered as transitional canponent between the midden and the gravels. This transitional zone, then, contains mottled brown soil with midden pockets and a high proportion of gravels. In this stratigraphic zone all artifacts decrease in quantity. At 50 an obsidian disappears aid at 60 an animal bones do the same. Chert renaims continuous in small amounts, as does fire-cracked rock. At the 80 an level, all intrusions described above are absent and it is safe to say that a pure gravel canponent begins. Also at this level, the soil becames slightly lighter and the clay content noticably increases. The gravel canponent at Son-577 is quite extensive, continuing at 210 ans. This gravel zone is characterized by a distinct change in artifacts. Nearly every piece of chert recovered is large scale, percussion flaked, aid exhibits some form of retouch or utilization. Tools tend to be quite informal consisting mostly of heavily retouched flakes; two scrapers aid seven retouched flakes were recovered fran this gravel component. 18 flakes were heavily utilized aid three of the four cores recovered had at least one edge retouched. Soil fran the 0-20 an levels was dry screened next to the unit. At this level, however, the clay content of the soil made sorting difficult. Therefore, fran 30 an to the eid of the unit, the soil was wet screened. The material was washed in 1/4" screens next to the unit by using water pumped fran Yorty Creek aid spray nozzles attached to the hoses. The 170-180 an level consisted of excavation only fran the eastern half of the unit. At this point excavation became difficult because of ground water percolation. The unit was then augered to 210 ans. Gravels continued at least to that depth. Site Son-590 (Grindstone) Son-590 is a petroglyph site consisting of two rocks with approximately 20 shal low, 5-10 an long grooves. These rocks are located in the stream bed of Yorty Creek, 10 meters west of, aid 7 meters below Hot Springs Road. They are 70 meters north of the point where Hot Springs Road first crosses Yorty Creek, and are located in a v-shaped canyon, almost in the actual flowing creek. The rocks cover an area measuring approximately 3 meters square, aid is on alluvial saidy silt. 91 This site is about 300 m downstream fran Son-584 (quarry, midden). It's actual association with either of these sites is questionable at present. No artifacts were found with these rocks aid no augering of the imediate area was possible due to the position of Son-590. These petroglyphs are extremely interesting because of their decorations. Ethnographic information has indicated that petroglyphs with cupules ground on their surface and sides were used by the Pono wamen as fertility rocks. Site Son-584 (Yorpy Quarry) Son-584 is a midden site associated with a quarry. It is located on a primary terrace on the north bank of Yorty Creek and bisected by an intermittent tributary of Yorty Creek. Even though there are two midden areas separated by the tributary, and also a quarry, only one unit was excavated at this site due to the pressure of time. The artifact bearing strata, as exemplified by the excavated unit aid stream cuts, are 80-90 an deep. As shown by the excavation, there are two definite components at this site, and there may be three. Fram 0-40 cm, there is a late ccnponent that is defined by two small corner notched points, a pestle fragment, and a much higher ratio of chert artifacts to obsidian artifacts. There is also in this late canponent a possible living surface on which were found many chert tools. This surface was at a depth of 23 cm to 32 cm in the east wall of the unit. This reading corresponds to the present slope of the ground. There may be a middle component at Son-584 but it was very poorly defined in Test Unit 1. It may extend framn 40 cm to as much as 80 an and is defined at Unit 1 only by an increase in the use of obsidian so that there is nearly a 1 to 1 ratio between chert and obsidian. The lowest cultural camponent was also poorly developed in this unit. We would have expected a gravel layer here to judge fran nearby sites but the gravel was found only in the west third of the unit ard was replaced elsewhere by nearly sterile clay. The industry in this gravel canponent characteristically consists of large, percussion flaked chert artifacts. Elsewhere in the immediate area of Son-584, the gravel layer may be seen nearby in stream cuts where it averages 50 an in thickness. There was an especially rich concentration of chert artifacts in the gravel layer on the NW side of the confluence of Yorty Creek and the tributary which cuts through Sor-584, including two cores, two flakes, one of which was retouched, and a bifacially worked tool. None of the artifacts found in this gravel layer is water worn and there is no evidence of their having been redeposited. Auqering in the area just S of the quarry showed the aravels at 70-110 an indicating the gravel canponent is well represented at the site, although there were only traces in the stratigraphic sequence in Test Unit 1. Son-584 has a geographic setting similar to Son-583 and Son-576, i.e., it is on a primary terrace on the N bank of Yorty Creek and situated right on the confluence of Yorty Creek and an intermittent tributary. Son-584 is located on the last wide, flat expanse of land along Yorty Creek, before that creek narrows and steepens up to its headwaters. It is then similar to Son-576 in being adjacent to a large, flat area of floodplain 92 CA-Son * 584 Map 0 10 20 Meters I I I I I I9 I \ 1 1 0 92 A 4/ * To Hot Springs Road Map 16. Son-584 (legend p. 5). N '9 no 20 5 0 _. _ O -30- -40- 50- -10- - 70-~ ~ ~ ~ ~ ~~~~~- ~ ' 100~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 A. Brown to dark brow; sl cl a _ A. a gr Ps s _ A root zone. Compac~~~~~t.A4 <>L-1& VA A4 C. Dark brown; clAy g comp at . _ _ A, _ _ D_ Dar gra to ver dar grys brwn silty_+_ _ _._ cla wit gravels__ _ . Compact _ _.- silt. Very compact._ _ _ _ _ _ _ _ _ _ _ _ _ _ ___A_ __AA____ F. Bedrock;to decomposing shaly. c a grass root~~~~~~~~~~~~~~~i zone. Ca A actA clay~~~~~~~~~ ~~~~~~ wit 4rves Cadet. E. Very dark gray; sandy gr~~~~~~ ~~~avl wit cla an silts~~~~~~~~~~~~~~~~~A 4A1 Ver aat F. Bedrock; decanposirlg shale.~~~~~~~~~~A A V Figure 13. Son-584 Profile. r-q % o ao L r-I q q 0) Ll 0D 0 0D 0D Nl 0 %O N N CN4 Cy o CN m v? 1? ? %O % O CO C4 r-i r-- tIIA A I 0 0N C0 o CO u ao 00 00 I C II qw (%~~ 0 o a' CO 0 0 0 r- 1 46 aN 0 1-I 94 0 m *- .0.4 CM a) n CO a1) ,14 a) w 4-I .0 U U) oY U) Co) -4 X a} ; m) )00 I o oCN o V00 u(0 0 U00 0 >1 1- 4J 0 >4 a) >-I .,,I Un 0 A I 1-I40 0 m 00 0 0 0 - 04 Pi 4U) 04- U 0 1-I a) 0 0 0 C4 0 CN ON y s Lo L en)-I 40 0 0 0 0 0 0 I Anr- N (0e r- N) 0n (N c-o w O aDE 95 I (N I N- 0 0 C) CN 0 96 The stratigraphic sequence for this area consists of 1-2 m of fine textured sands and silts which include midden deposits, underlain by less than a meter of sandyy, clayey, coarse well-rounded stream gravels, then shale bedrock. The quarry associated with Son-584 is found on the hillslope just 20 m north of the surface midden area. The greatest concentration of cores, flakes and chunks in the quarry are found in an area of recent slumping where these buried artifacts are now exposed and appear to be emerging fram the hillside. The chert is a locally derived red and green banded chert but the actual source of the quarry has not been found yet, and may also, along with most of the artifacts, be buried. The area of Son-584 is one of open oak woodland, with Oregon oak and live oak being the dominant trees. Recent deforestation has reduced the number of trees which would naturally occur here. A hundred meters down Yorty Creek fran the site, the valley of Yorty Creek steepens and narrows for the next 400 meters. In this steep valley the vegetation is open of dense oak-madrone forest with a few Douglas fir present. Yorty Creek and the unnamed tributary running through the site provide easily accessible water sources. However, both these streams probably dry up during some sumner months. There is very little historic impact on the area of Son-584. The area of this site has been grazed fairly heavily but does not appear to have been plowed. It also has not been disturbed by roads or buildings and local residents did not know of the existence of an Indian site at this location. However, recent heavy stream erosion has cut away a portion of the midden along both Yorty Creek and the intermittent tributary, and recent slumping in the quarry area has not only exposed the artifacts but has moved them. Son-584 is an important site for three reasons: one, the extent of the midden which contains the late and middle components indicate that it may be the central habitation site for upper Yorty Creek since it is larger than any of the other nearby sites; two, the gravel component, an important and unexpected addition to the archaeology of the project area, is found in all stream cuts around the site with an especially rich concentration just 60 m fran the site. A deep and thick buried midden, which may be analogous to the gravel, has been found on the site - all of which indicates this is the prime area for exporing the gravel canponet; and thirdly, the presence of the quarry makes Son-584 an important site since it is the only quarry found in the northern portion of the project area. Site Son-583 (Yorty Hollow) Son-583 is located on a small primary terrace on the north bank of Yorty Creek. It is approximately 150 m north of Hot Springs Road and can be approached by the first dirt road upon entering the Yorty Creek drainage. An abandoned cabin and several farm buildings are located at the northern end of this dirt road on the south bank of Yorty Creek. Son-583 is 250 m upstream fran these buildings. Geographically, this site is very similar to Son-576 in that they are 97 CA- Son - 583 Map 'iil il ti'i i'i!jipi ji ji jii ji ji ji'ji ji jij, I j I sj I j|j I j I j I j I j I j' I j I 'j I j I !j I j I j ,j j j Ij Sj | ffi ffi 18181?1|1 1 1 181g1?181|1?1|1 1118181g1810181| I 1 | 1 ,1.1,1, 1 1 1 1 | I 11, | I I 1 | 1 1 I I | I | * 1 l 1 1 1 1 * I @ I | I | I | I ?1 I | I | I | I w I | I | I | I | I | I | kN , | | | *l | * I | | * | *, | I * I * I * | I | * | ?, l, ?, * | | | | l, l, l, l, | F_: |ll - |||1 .1.|.||8 l l lB llll le1||||||s | 1wll lll 1sU _l | l, | lllllllllllllW ,|,gllllllllllllW , 1||l 11llle1|1|1|l lll llllll 1|1|1||||l lllllI 'S M W 1sls l|"1"|s1ffi|Xl l 1s 1 llla &1s1"11g11 1|l s1 l R I |1|1|1|11"|911|1|1|1|"|1|1|1|||||1|1|1|1|||M1 lw lllllllllllllllllllllllullllllll&||lllllllul 9|||. 1||l ll"191||"1,111 l l l sl|& 1||||||||s 11l, l' - ||l 1|1,l@|sl|sl.1ffi1s1s1,1,|,|,1,1,1,11111 1|1|l 11|1 j ||111121@lslxl@glllgllllulllulsllllBlululwl|lll . !i!i!,!,!ifi'l'i'i'i'i'i'i'i',','ili'i'i!i!,''; /I /II rI, I I I, I I 'II I I 'II I I I, II I I I I I'I I I I'I I I I I I' I I 0 5 10 Meters Map 17. Scn-583 (legend p. 5). I I i I 98 207 -10- 40 -0 compact. Annualgrassrootzo .. Q~~~~~~~~~~~~. gravels; moeatl omat - 20~~~~~~~~~~~~ 100 C. Grayish brown to dark brown; sandy clay with gravels and cobbles. Compact. D. Bedrock. Son-583 Profile. Figure 14. 99 Site Son-583 Proj. Scraper Retouch Pts. Flakes 1-c 1-o 1-o 1-o 1-o 1-c 1-c 1-c 1-c 1-c 1-c 1-2 2 1-c 1-2 2 1-c 2-o 3 (Yorty Hollow) Used Cores Debitage Flakes 1-c 93-c 1-o 16-o 1-c 166-c 1-c 5-o 1-c 40-c 9-o 2-c 42-c 12-o 3-c 85-c 1-c 17-o 2-c 62-c 2-o 21-o 2-c 24-c 2-c 3-c 3-c 1-c 15-c 3-c 4-o 1-o 19 4 8-o 13-c 2-o 3-c 17-c 24-c 31-c 600-c 90-o 690 Fired Animal Rock Bone 50 150 10 215 5 230 4 large quant ity n 2 2 small quant ity 23 23 Level an. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 Gravel Gravel Gravel Totals 1 2 3 Grard Totals - - - 100 both located on the confluence of an intermittent tributary and Yorty Creek, and they both have a deep, well-developed midden limited to a relatively small area, Both sites seem to have been inhabited by a small group of people on a seasonal basis for a long period of time. The area of Son-583 shows evidence of historic use. An old redwood sheep stake fence passes through the extrene eastern edge of the pranontory and an old road cut can be found 50 m south of the site, running parallel to and 50 m north of Hot Springs Road. Recent disturbance is indicated by a powerline located just south of the old road cut. Son-583 is in an open oak/woodland canmunity consisting of live oak, Oregon oak, and a scattering of buckeyes and madrones. The surrounding slopes and terraces sIhw signs of heavy deforestation in the form of numerous stumps and fallen trees. Intensive sheep grazing and rapid stream erosion have produced slumping in the area and have otherwise disturbed the site. The soil sequence in the immediate area consists of 10 an of overburden, then approximately 1 m of fine-textured alluvial soil, and finally 20-50 ac of coarse gravel overlying shale bedrock. The midden of Yorty Hollow encompasses nearly all of its pranontory, measuring 8 m by 15 m, arxn averages 50-60 an in depth. An associated flake scatter extends across Yorty Creek, south to the historic road and includes a possible housepit near this road, (50 m southeast of the midden). Test Unit 1 was placed in the area where the stream cut indicated the deepest part of the midden: 8 m south, 450 east of the site datum. Excavation of the unit indicated that this area was the edge of an ancient stream cut that eroded onto the shale bedrock, was subsequently overlain by gravels and then filled in by midden and recent overburden. There are at least two separate components at this site: midden and gravel. The midden is black clayey soil with some angular gravel extending to a depth of 50 cm at the west edge of the unit and 89 cm on the east. This midden overlies at least 40 an of grey-brown clayey gravels. Within this zone animal bone, fire-cracked rock and obsidian are not present. Only chert debitage survives and increases in quantity with continuing depth. This debitage is characterized by large, informal, percussion flaked tools and flakes, and represents a distinct change fran the midden component. These gravels continued to bedrock at 80 cm in the southwest corner of the unit and 102 ams in the middle of the east wall. The midden was first cleared off the top of these sloping gravels in less than canplete levels, and then the gravels were excavated in three separate levels. The clay content of this lower zone necessitated wet screening fran 50 cm to the end of the unit. Son-583 will not be subject to inundation. This site, however, has a high potential for destruction by natural erosional processes, and will be located near a proposed recreation area. 101 Yorty Creek Group Chronology Our chronology here is based primarily upon typology but again we have two radiocarbon dates (Son-576) for cross checking. Son-576 This site presents some difficulty in dating. It will be seen fran the table of chert-obsidian ratios that there is very little consistency in this statistic throughout the nearly two meters of depth. Even adjacent levels are inconsistent in most cases so there are no natural breaks to be discovered here. In our earlier report we distinguished three ccmponents in the site as follows: 0-30 an. Late Houx 30-50 an. Early Houx 50-190 an. Late Borax Lake Since then we have obtained two radiocarbon determinations fran this site. UCR-353 Charcoal fran hearth at 40-50 cm. 675+ 140 AD 1325 UCR-354 Charcoal fran midden at 170-180 an. 2300+ 160 350 BC There are only three diagnostic artifacts from the site: a very small Excelsior point fran 10-20 an (Fig. 25g), a large Excelsior point fran 70-80 an (Fig. 25i), and a pestle fran 30-40 an. The large Excelsior point could date Late Borax Lake but it could also be fran Early Houx. The small point probably canes fran Late Houx while the pestle is most probably fran Houx Aspect. These items together with radiocarbon dates suggest that the entire midden should be attributed to the Houx Aspect, part early and part late. But which part. The later radiocarbon date indicates that the top 50 an. should be Late Houx so we will tentatively assign the remainder to Early Houx. Son-577, 584 and 583 The chronology of these sites must be dealt with together because at the base of each of them is an artifact bearing layer of gravel which appears to be the same both chronologically and culturally aid must be interpreted before the superimposed middens are placed in context. The gravel layer in question is found also away fran the numbered sites being exposed by stream cutting in Yorty Creek fran about Son-576 at least to the edge of the project. The stream cutting is presumed to be recent and due to changed stream flow conditions since denudation of hillside vegetation in historic times. In any case the gravel banks of Yorty Creek occasionally included flakes aid cores apparently modified by man with a few undeniable artifacts such as scrapers aid bifaces. Although a few of these items seem a little waterworn none of them could be called "rolled" so we cannot believe they are redeposited fran some midden site upstream. The flakes and cores are heavier aid cruder than anything fran the middens aid all are of chert - there is no obsidian at all. We illustrate in Fig. 30 the unquestionable pieces. It will be seen that there is nothing there of diagnostic value fran the point of view of either North Coast Range chronology or that of any nearby area. Thus we are faced with a situation (like that of the Farmington canplex) of a crude chert industry in an undated gravel deposit. We feel certain that 102 the industry is old but we are unable to give it a more precise date. Nearly all the diagnostic material fran superimposed middens seems to date fran the Houx Aspect. The principal exception is the single large side notched projectile point shown in Fig. 25c. This was fran Son-583 and was found embedded in the stream bank 45 an below surface and well above the gravels. This type we indicate as being (most probably) part of Early Borax Lake. If this is accepted then the gravel layers must be earlier than the Borax Lake Aspect. At the same time there is no evidence of any relationship to the Post Pattern. We therefore are canpelled to place them in sane pre-5000 BC limbo. We will call them Yorty gravels for present purposes. Clearly a great deal more work needs to be done on these gravels fran a geological as well as an archaeological standpoint. For the midden components of these sites we will make the following chronological placements. Son-577 has nothing diagnostic - we attribute it to the Borax Lake Aspect since it immediately overlies the gravels. Son-584 has 3 stratigraphic levels, the top one (0-40 an) being attributed to Late Houx because of small barbed projectile point (Fig. 25d, e) and the pestle at 20-30 -an while the bottan layer is Yorty gravels. This leaves the middle level (40- 80 an) either as Early Houx or Borax Lake. Son-583 must also be late Houx in its upper levels (0-30 an) because of the small point shown in Fig. 25b while the middle levels 30-90 an must be Borax Lake for reasons discussed above. 103 Yorty Creek Site Group Chert-Obsidian Debitage Ratios by Site and Level Site CA-Son 584 * (1) 7.67(52) 5.05(127) 1.74(107) 1.52(53) 0.88(62) 1.68(42) 2.83(27) 1.75(22) 1.50(10) 2.00(3) 0.00(1) 583 5.81(109) 33.20(171) 4.44(49) 3.50(54) 5.00(102) 2.95(83) 3.00(32) 6.50(15) * (3) * (17) * (29) * (31) ( ) : debitage frequency * : infinity, or all chert Level cm. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 576 6.71(54) 1.07(27) 3.67(56) 4.29(74) 4.75(69) 2.04(70) 1.94(47) 0.87(43) 1.00(54) 7.56(60) 5.00(6) 0.49 (55) 0.71(87) 0.77(46) 0.34(39) 1.00(18) 2.67(11) * (2) 0.33(4) 577 5.50(39) 15.67(50) 23.50(49) 21.00(22) * (15) * (13) * (5) * (6) * (5) * (7) * (3) * (9) * (9) * (6) * (12) * (3) * (1) * (1) 104 Yorty Creek Site Group Chert-Obsidian Debitage Ratios by Site ard Period Period 576 Site CA-Son 577 584 583 0-50 4.09 (280) 50-190 1.13 (542) 0-90 16.00 (204) 90-180 all chert (51) 0-40 3.35 (287) 40-80 1.36 (184) 80-125 1.77 (36) 0-30 * 9.97 (329) 30-90 3.81 (289) Gravels all daert (72) Late Houx Early Houx Early Borax Lake Late Borax Lake Hb BLb BLb BLa Hb: Ha: BLb: BLa: - 105 Cherry Creek Group Description This small group does not seen to be of great importance. Aside fram the surface site Son-552 the sites of the group occur well up Cherry Creek where the valley becomes narrow and steep-sided. These sites are separated some distance fram what would appear to be the main activity centers on Dry Creek and Yorty Creek and we speculate that they may be given over to some specialized activity, perhaps hunting. The supposed prehistoric vegetation cover here is woodland-grass or woodland-chaparral either of which would suggest deer habitat. Site Son-552 (Cherry Creek Crossing) Son-552 is located on the east bank of Cherry Creek, approximately 30 meters south of the point where an old jeep trail crosses Cherry Creek. It extends NE-SW along a terrace located on the 400' contour. This site is in the center of the Cherry Creek drainage, all of which is encompassed by the Project Area boundaries, and is more than a mile away fran the nearest site to the north or south. Son-552 is a surface site located beneath a large spreading oak tree with a marsh area east of it. It appears to be an early homestead site with evidence of still earlier Indian occupation. The jeep trail runs a half mile north of Hot Springs Road and bisects the terrace on the north. South of this trail, just above Cherry Creek, is an historic dump with many glass and china sherds, (ca. 25 meters north of datum). A brick lined well is located approximately 30 meters south of the dump and an old rusty plow disc was found 15 meters north of the jeep trail. A basalt pestle fragment was found on the eastern edge of the dump and next to the road. A second pestle was located approximately 8 meters due south of the western part of the dump area, and an obsidian flake was located on the edge of the road north of this same dump section. The terrace has open oak woodland and grassland cover. The soil seems to be a brown sandy, silty loam with no obvious dark midden showing. Two auger holes were sunk on this terrace, one in the center of the historic dump and a second, located 2 meters west of datum stake. The one in the dump recovered clear and opalized glass fragments and charcoal chunks; the second augering was sterile. Site Son-587 (Two Housepits) Son-587 is located in the northern section of Cherry Creek approximately 225 m south of Son-604 and 270 m south of Son-589. It is a housepit site consisting of two circular housepits approximately 2 meters in diameter, 40 cm deep and 10 meters apart. The terrace on which these are located has two seasonal drainages. The northern drainage is located about 90 meters north of the housepits, and the southern drainage, 30 meters south, bisects the terrace. The housepits are approximately 10 meters west of the base of a hill, and about 50 meters east of Cherry Creek. The terrace seens to have supported an old homestead, witnessed by old fruit trees (just north and south of the southern drainage) 12-18" in 106 diameter. Sheep stake fencing is found along the eastern edge of this terrace at the base of the hill. An iron pipe was found approximately 15 meters east of the housepits. This terrace has open grasslard cover which has been cleared for the cultivation of the nearby fruit trees. The soil of the terrace and the surrounding area consists of brown gravelly silt and shows no signs of midden deposits or cultural debris. 'No auger holes were placed on this terrace, one between the two housepits, arn one 45 meters south of the housepits, under the fruit trees. No sub-surface midden was indicated in these holes. Site Son-604 (Skunk Creek) Son-604 is located in the northern section of Cherry Creek approximately 50 meters north of the confluence of Skunk arn Cherry Creeks. It is located on a narrow flat that lies parallel to Skunk Creek and is approximately 2 meters wide ard 3-4 meters long. This flat is on the east bank of Cherry Creek arn lies on top of ard is surrounded by serpentine outcrcppings. It is in a deep canyon with steep slopes, many waterfalls and year-round deep pools. The terrace on which this site is located has much chapparal, scrub oak, chamise, etc., on its south-facing slopes and has Oregon oak woodland grassland cover on its north-facing slopes. This is a midden site and the area of the midden (2 by 3.5 m) is dark brown friable silt with much angular rock and can easily be distinguished fran the surrounding soils. Two auger holes were placed north and south of datum in the midden area to test extent and depth of this small midden. In both holes the midden exterds to 20 an at the most, with auger refusal by the serpentine bedrock. Most of the extent of the original midden has been washed away ard only this small, shallow midden area presently remains. Site Son-589 (Rake-In-Tree) Son-589 is located in the northern section of Cherry Creek, approximately 100 meters north of the confluence of Skunk and Cherry Creeks. It is on the east bank of Cherry Creek about 50 meters north of Son-604. It is a housepit site consisting of one housepit located on a rise in an area that has signs of homesteading. The area has old wire fencing around its perimeter aid old boards are scattered on the surface of the knoll. A rake head is buried in an oak tree on this terrace. The housepit is located approximately 50 meters east of Cherry Creek and about 22 meters west of the base of a hill. Site datum is located in the center of this housepit. The housepit measures approximately 2 meters in diameter aid about 30 an in depth. This terrace has open grassland cover with Oregon and black oaks scattered on the surface. No midden deposit was evident on the surface of this knoll. A series of six auger holes was placed around the housepit to test for sub-surface midden deposits. Only one of these six revealed any evidence of cultural habitation, producing small flecks of charcoal and 2 small round nails, (auger hole 5; 3 m west of datum). 107 Site Son-603 (Salmon Hole) Son-603 is located on the west bank of Cherry Creek on the northern part of its drainage. It is approximately 750 meters south of the confluence of Snow and Cherry Creeks and a half mile north of the confluence of Skunk and Cherry Creeks. The site is on a terrace measuring about 35 meters by 15 meters which runs north-south, parallel to Cherry Creek. This terrace is one of only a few openings in a heavily wooded, steep sided canyon. Son-603 is located approximately 250 meters south of Son-601 and about a half mile north of Son-589. This site is a shallow midden encanpassing most of the open terrace. The terrace is bordered on the north aid west by steep sided, heavily wooded slopes covered with wild iris. The entire eastern edge of the terrace is delimited by the cut bank formed through erosion by Cherry Creek. The terrace itself has semi-dense oak woodlard cover consistent of Oregon, canyon, anxd coast live oaks with grassland beneath. Some buckeyes, bays, madrones and manzanitas are scattered around the perimeter of the site. The midden is easily distinguishable from the surrounding lighter soils on this terrace. The entire site is covered with naturally formed augular rocks, making identification of fire-cracked rocks difficult. Test Unit 1 was placed N 30? W, 10.55 meters fran site datum in the area which seemed to have darkest midden and the least amount of rocks. The shallow midden component at Son-603 exteixded to 50 an with snall midden pockets continuing into the sterile sub-midden gravels below. Obsidian dominated the lithic debitage counts throughout the entire midden depth. A few pieces of chert debitage aid animal bone fragments were also recovered. Three obsidian projectile point fragments (all tips) came from this midden component. Singley these are undiagnostic but their general shape and size indicate that they are not of a late phase. We assign the site to the Borax Lake Aspect. This site is very similar in geographic position to the other four sites located in the northern extreme of the Cherry Creek drainage, (Son-601, 604, 589, 587). This cluster of sites is important geographically because it may be a link to sites of another tribelet in the next drainage. There is a distinct hiatus between these sites and the rest of the sites located in the northern section of the Project Area. This cluster is approximately three miles north of the Yorty Creek/Dry Creek confluence on which creeks the northern sites are located. Two sites located further north, (outside the Project Area on this drainage), probably part of this cluster, have produced interesting surface finds which are part of the Moretti collection. Among these finds is the only charm stone known to exist from this area. Site Son-601 (Eel Pit) Son-601 is one of five sites located in the northern section of the Cherry Creek drainage within the boundaries of the Project Area. All five sites are found on the only available open terraces within this heavily wooded, steep canyon. The two northerrinost sites, Son-601, and Son-603, are located on the west bank cf Cherry Creek and the three southern sites are on the east bank of this creek. 108 Site Son-603 (Salmon Hole) Level Proj. Reworked Retouch Used -Debitage Animal cM. Pts. Biface Flakes Flakes Bone 0-10 6-c 7 1-c 38-o 10-20 1-c 7 2-o 30-o 20-30 2-c 6 1-o 35-o 30-40 11 2-o 46-o 40-50 l-o 14-o 4 50-60 3-o Totals 2 2 2 1 175 35 109 -o~sit anua .ras root zon. Friable 10 -20 40 -~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 60~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ silt anualras roo zon. Friable B. Midden; very dark grayish brown; gravelly silt; friable to very friable. C. D~ark yellowish brown; gravels with sari aid silt; crumbly. Figure 15. Son-603 Profile. a~ 0 5 10 L a I Meters Map 18. Son-601 (legend p. 5). 110 - joi lo m 20 - 30 40 50 cm I L I * A. Midden; dark brown, silt and sard; Friable to moderately canpact. B. Midden; dark canpact. annual grass root zone. brown; silty gravels; friable to moderately C. Brown; silty clay; canpact. Figure 16. Son-601 Profile. ll 112 Son-601 is located approximately 500 meters south of the confluence of Snow and Cherry Creeks arn approximately 1000 meters north-northwest of the confluence of Skunk arn Cherry Creeks. This terrace is oriented north-south arn runs parallel to Cherry Creek, approximately 3 meters above the creek. The terrace measures 30 meters by 15 meters with the midden concentration located within the northern 20 meters of the terrace. The site has semi-open oak/woodlarn] cover with black oaks, canyon live oaks, bays, buckeyes, and broadleaf maples. Annual grasses and Brodeia are found over the entire terrace. The northern perimeter of the site is delimited by a large bedrock outcrop; the western by stew oak/grassland slopes; anx] the eastern by the cut bank eroded by Cherry Creek. The area of the midden is easily distinguishable fran the surrounding lighter soils on this terrace. Test Unit 1 was placed in the darkest midden area, 2' meters east of site datum. The area is heavily covered with rocks. The unit was placed in the area of least rock cover, but despite this, the unit was very difficult to excavate. Areas of rock concentrations began to occur in the first level excavated aix continued to 30 an. Roots were equally heavy in these initial levels. Fire-cracked rock could not be visually separated fran the other rocks in the unit aix] was, therefore, not recorded. At 30 an large areas of krotovina dominated a major portion of the unit. Excavations ceased at 50 an when the 12 flakes recovered all came fram disturbed areas. The shal low midden component at Son-601 extended to 40 an with small midden pockets continuing into the sterile suknidden clays below. Obsidian dominated the lithic debitage counts throughout the entire midden depth. A few pieces of chert debitage ard animal bone fragments were also recovered. One chert projectile point, a rounded base fragment came fran this midden component, (Fig. 24j), suggesting that this site was not late in date. Cherry Creek Group Chronology There is insufficient evidence here to warrent extensive chronological treatment but a couple of observations about the two excavated sites seem in line. Firstly there is no evidence in either of these sites of material fram Late Houx so on this basis they must be attributed to Early Houx or Borax Lake. This suggests that the specialized hunting or local plant gathering activities hypothesized for these sites was abandoned in this latest period. We are inclined to doubt this aix] attribute our lack of late evidence to small sample size. The other point worthy of note is the very small chert-obsidian ratios, 0.05 for Son-603 aix] 0.23 for Son-601. The other sites showing similar obsidian domination are Son-571 in the Western Group and Son-556 in the Lower Warm Springs group. Both these sites have large parts of their deposits attributed to the Early Houx. It may be then that this is also true of the two sites on Cherry Creek. 0 0 (a~~ 0).0- a b-. b- e 0 1~0 0 0 0 x ~~0 I- 0 0 C.) Map 19. Central Group West. 113 (I) w a. I- z CD 4 w z n I w -j 0 0 0 w X: 4 0 4 0 I- z w C) 06 4- 0 0 0 N . . . . *b** ................................... -*- ***@@*X* *@@ ............................. .......................... .................................. -@ *@*X**@******* * **** ..... * ........................... ........ ....... _! .. i ..... ... a, .... s . v > *@@ @@-sss ***X***X@* *@@ *@@@ * *** > ...................... *********@@@@***s* *-.* .. XX. *X?****-*** .................... ********X *X * *** v ..................... ........................... ..... ............................ ............................ *v v@@@* *@@@* * ***v @e*** ........................... ........................... ....................... ....................... *^ ........................ ............................... *. * *@o .......................... ......................... .................... .................... ***. * ** ............... * *.** .......................... .......................... ......................... ................................. * vv *v* ... * ** ****@@ ..... @- .......................... ..*...................... ........................ ....................... .............................. *e * * *- ......................... *e e .... * ***- *....................... .* *-X-X X**- ........................ ....................... ................... . *.*. ....................... s ^, . .. .78s , . , ................ % _, ..... .. s , o ., s ...... _ z , z . . . . . . . . Di. . . . . . . * * , ........ \ ,., * ' : ......... ....... , ,, . .......... ........ _ . , , ............ .............. _ o . ........... - , .@ ........... o# S . ............. ............. ,o 1 ........... ., * w.s . X, ..... .. f***- * t @b ............. ................ . . ............ ........ ,%.,5 > .............. , *,^, ,, ............ ............... , .o -\ .............. , . # 4 4 t @ O ............. ............... . .* * ............. ............ ,4., c , .............. ................. * r .............. .......... : , ? . . . ? @ @ - @ @ @ v * * * . ! 2 :::::::::::::::M' IL - l -::: -'5?S;: ;s: > # '.tv '''' 1i X S;'g , :' . N , * .. ,.......... _: , : . l- ,. . U) 0.~~~~~~~0 4.- a 1 1 4,e 4-L O Yt n ~ r m ~ n w ~ r 1 ~ T T T T ~ r T T : Tn Map 20. Central Group East. 114 'J. , , ... ....................... . ..* 0. r.............. j4+ i.. cn) Cl) w a- I- z 4 cn w LI w 0 z Cl) w I- CD 0 0 w 0.. 0 CD -J Ir- z w C- / 2V-10 4 CY 115 Central Group Description The Central Group is on the stretch of Dry Creek fram Prichett Peaks on the west to the mouth of Warm Springs Creek on the east. The first half of this stretch is rather narrow ard steep sided while the lower half widens out considerably. The first part contains only a few small sites, as one might expect; the second part, which one would have expected would be more favorable to native settlement, was also favorable to orchards and vineyards so that it seems apparent that at least two excellent archaeological sites were almost totally destroyed by agricultural activity. On the whole, though, art even assuming that there were formerly more sites here the total amount of archaeological remains in this group seems much sparser than elsewhere in the project. As discussed in the conclusions we are hypothesizing that this was a boundary area between tribelets and may have been less densely settled for that reason. There is no obvious ecological reason for the lack of settlement in the distribution of vegetation types as shown on Maps 19 aid 20. Site Son-562 (Oak Point) Son-562 is in the extreme western part of the Central Section. It consists of a single housepit on a small first terrace on the north bank of Dry Creek. The tergace is five meters upslope from Dry Creek, has open grassland cover, aid is encircled by heavy riparian vegetation. Several large stands of oak, bay, manzanita aid madrone, with an occasional fir, caopletely shelter the terrace, which measures 50 meters by 30 meters. The housepit, two meters in diameter aid 80 an in depth, is in the center of the eastern part of the terrace. During survey the area around this housepit was searched for petroglyphs aid cultural evidence, but none was found. A series of 12 auger holes were placed randcmly over the entire surface of this terrace, aid although charcoal flecks were in evidence in many of the holes, only one obsidian flake was found. This flake was recovered frcn auger hole 10 at a 55 an depth aid was associated with a few charcoal chunks. However, auger hole ll, two meters north of 10, contained a few flecks of charcoal on the surface and did not reveal buried midden or cultural material. Both of these auger holes were located at the western end of the terrace, at the point furthest away fran the housepit. Site Son-574 (Purple Blind) Son-574 is a probable hunting blind found on the south bank of Dry Creek, a mile and a half upstream from the Old Rockpile Road Bridge, aid directly below Pritchett Peaks. The blind is on a small terrace, 110 meters south of aid 15 meters above Dry Creek. It overlooks a gently sloping, grass-covered clearing in a steep, densely-wooded terrain. The clearing is bordered to the north by a barbed wire fence that is on the cut bank of Dry Creek, to the east by a barbed wire aid stake fence, aid to the south aid west by a steepening hillslope. Above the terrace. with the blind is another oen, grassy clearing. The vegetation of the site is open grassland with a few scattered staids of oak and manzanita. The surrounding area is a dense oak/madrone forest with some redwood aid fir. 116 The hunting blind consists of a circular depression measuring 2.5 meters in diameter and 60 centimeters in depth. The grassy clearings above ard below this blirx] form a natural path to Dry Creek, making the position of this blind excellent for ambushing game. No midden or cultural artifacts were found in or around the blind. Site Son-598 (Treganza 10) Son-598 is a non-midden site with 14 well-defined housepits and a cupule petroglyph rock. It is situated on the south bank of Dry Creek directly below Pritchett Peaks. The site is on a primary terrace, measuring 150 meters by 80 meters, arn is two meters above the stream bed of Dry Creek. The terrace is bordered to the north by the Dry Creek stream bed, to the east by an intermittent drainage, and to the south arn west by the increasing hillslope. The vegetation of the terrace consists of open grasslarxn cover with scattered stards of oak, buckeye arnd manzanita. There is riparian vegetation of laurel and willow along the drainages. The surrounding slopes are covered with a dense oak/bay/mairone forest with sone redwood an] fir. The soil around the housepits is the same as the soil of the rest of the terrace; a moderately canpact, brown to dark brown, alluvial gravelly sand silt. No midden is in evidence on the surface nor was any uncovered in the five auger holes made in the vicinity of the housepits. The housepits are clustered on the eastern half of this extensive terrace and cover an area measuring 70 meters by 30 meters. The 14 housepits are very well defined, range fran three meters to six meters in diameter, and are as deep as one meter. The cupule rock is located on the northwest edge of the terrace and the streambed. The rock is a small, low-lying serpentine boulder that was partially buried in the terrace slope. It contains six shallow cupules four to six centimeters in diameter. There were no artifacts found during an intensive search of the surface of the site during the initial phase of the Warm Springs Archaeology Project. A one meter square test excavation unit was placed in the area of the housepits, between the largest of the 14 housepits and a small one 10 meters to the west, 15 meters due west of site datum. This unit ended at 30 an after three levels of sterile alluvium. A five centimeter thick burnt layer of soil was uncovered at ten centimeters below surface, which contained no cultural material. This carbonized layer is probably due to natural causes. Site Son-548 (Turtle Stink) This is a shallow, weakly developed midden site on the north bank of Dry Creek, three quarters of a mile upstream fran the Old Rockpile Road Bridge, at the south base of Pritchett Peaks. The site is located on the western side of a primary terrace, measuring 70 meters by 25 meters, in a steep, fairly narrow valley. The vegetation of the terrace is moderately dense oak/bay woodland, with some buckeye and madrone. There is a dense grass groundcover on the site. The slopes around the site have roughly the same vegetation type, with the lower slopes on the south bank of Dry Creek being more densely wooded and 117 having occasional redwoods. The soil of the site is the same as the surrounding terrace soils; a moderately caopact, brown, alluvial sandy silt to silt. Three auger holes were made on this terrace in the area of the site but no sub-surface midden deposits were found. The area of the site measures 10 meters by 15 meters aid is defined by a sparse scatter of artifacts aid fire-cracted rock. The first five to ten centimeters of the surface of the site is disturbed by the rooting of feral pigs, who feed on the acorns fran the many oaks on the site. This was the only disturbance of this site; erosion by Dry Creek was minimal. A one meter square test excavation unit was placed in the area of heaviest artifact scatter, 90 ac due east of site datum. The unit ended at 30 ac in a nearly sterile layer of brown, alluvial silt. An auger hole in the bottom of this level exteided to 65 cm aid revealed no midden. Excavation revealed that this is a weakly developed midden site, with only one zone of occupation, and few artifacts, none of them diagnostic: Level Scraper Debitage Shell Fire Rock Bone 0-10 an 3-c 1 7 5 6-o 10-20 an 6 - 1-o 6-o 20-30 1 - 2-o Site Son-542 (possibly "Oak Ball Drifts Out") This is a disturbed site with a shallow, weakly developed midden located on the north bank of Dry Creek about a mile upstrean fran the confluence of Dry and Warm Springs Creeks. The site is on the edge of a 9-12 m high first terrace, 150 meters south aid east of Old Rockpile Road. The weakly developed midden covers an area measuring 50 meters by 25 meters. It is bordered on the south by Dry Creek, on the east by an intermittent tributary, aid it grades into non-midden soil to the west aid north. There is an associated flake scatter which is confined to the front of the terrace aid measures 120 meters by 40 meters. Dry Creek is now 300 meters south of this site and is separated fran the site by a wide gravel streambed which has been used in the past as a gravel borrow area. Directly below the site there is a year-round pool which is probably the result of mining operations. The original vegetation on the site was, according to knowledgeable local residents, open grasslaid with scattered staids of Oregon aid valley oak. Riparian vegetation was once quite abundant in the seasonal drainages, aid redwoods, which are still found in the uper parts of these drainages, once exterded over the terraces to Dry Creek. The natural vegetation has long 118 since been replaced by cultivated crops starting with hops, plums aid finally the present vineyard. This site has been cultivated for at least the past 80 years aid the midden is now very disturbed. Artifacts found in the larger area of the flake scatter may have originally been in the midden area but became more widely distributed by agricultural activities. The soil of the midden is only slightly darker than the surrounding soil deposits on the terrace. It is a loose, dark gray to dark brown, silty gravel with many pebbles aid cobbles. The rock content aid looseness of the soil did not permit augering. Two one meter square test excavation units were dug on this site. The first unit was located 450 south aid 8 meters west of site datum in the central part of the midden. This unit ended at 20 cm in a sterile, yellow brown, sandy coarse gravel. No artifacts were found in either of the two levels. The second unit was excavated at 590 west aid 35 meters south of the site datum, where four chert aid obsidian flakes had been seen on the surface. This unit also extended to 20 an aid ended in a sterile, brown, sandy gravel. One chert side-notched point aid two utilized obsidian flakes were the only tools recovered in excavation. Surface finds were also few, but included an obsidian corner-notched projectile point fragment, two chert scrapers and a chert core, as well as 20-30 flakes and chips. Fran this inventory, it is clear that the weak midden at Son-543 has been so disturbed as to reider intensive excavation of limited value. Site Son-599 (Jackson's Blind) This site is an isolated, well-defined huntirn blind located on a minor tributary of Dry Creek, 200 meters upstream from its mouth on Dry Creek. The blind is situated at the southeast dege of an open, flat clearing in a very steep, moderately forested terrain. It is 50 meters west of aid 35 meters above the steep-banked tributary. The blind overlooks both the tributary aid the clearing. The clearing is bordered on the east by the intermittent stream, on the north by a densely forested slope continuing down to Dry Creek, aid on the south and west by a moderately forested steep hillslope. The vegetation of the clearing is open grasslaid with very dense grass groundcover. The dense forest north of the blind aid clearing consists of predominately oak aid madrone with some bay present. The forest to the south aid west includes chaparral/oak woodlaid with some pine on a thin, serpentine- derived soil. This is the best preserved hunting blind in the Project Area. It is a stone-lined, rectangular depression which measures 2 meters by 1.2 meters aid is 80 an deep. Approximately 40 stones line this pit, some of which still remain in courses. The highest remaining row of courses is oriented towards the southwest. No midden or cultural material was found at this site. However, Son-600, a late midden site, lies on the north bank of Dry Creek approximately 400 meters below this site. The sharply defined nature of this blind indicates that it is a relatively late feature aid may be contemporaneous with the late parts of site, Son-600. 119 Site Son-600 (possibly "Takoton") This is a disturbed midden site with a shallow, weakly developed midden located on the north bank of Dry Creek 750 m upstream fran the confluence of Dry and Warm Springs Creeks. The site is on a high first terrace approximately 40 meters north of Dry Creek. This site, like Son-542, is a low mound on the edge of the terrace. The vegetation of the terrace is now vineyard. Prior to cultivation, this area was in open oak woodland with thick groves of bay, willow, and cottonwood along the fringes of Dry Creek. The slopes tQ the north are thickly forested with madrone, oak, and pine and remnants of large redwood stands are in evidence in the drainages. The soil of the site is loosely compacted, dark gray to dark gray brown, gravelly silt. The surrounding soil is brown to dark brown, gravelly silt. Five backhoe trenches were excavated in the vicinity of Son-600 before testing by the Warm Springs Archaeologgy Project, but none showed any signs of midden, (see Map 21 T-18, T-19, T-20, T-21 and T-29). Nine auger holes were bored throughout the entire area of possible midden. These borings demonstrated that the extent of the midden covered the same area as indicated by the surface finds. The midden in the auger holes on the perimeter bottomed out at 20 cm and towards the center at 50 ca. Those outside the darker, midden area were sterile. Although the flake sqatter measures 140 by 60 meters, the midden only encaopasses an area 100 by 40 meters. The midden at Son-600, like that of Son-542, is nQt ntact. The previous owner of the land, Mr. Edward Mauritson, stated that there has been much agricultural activity that would have disturbed the site. In the late 1880's all of the oak trees that were growing on the site were pulled out. Grapes were planted, then replaced by hops, which were in turn replaced by plum trees. The plum trees remained for many years until, in 1947, Mr. Mauritson personally bulldozed them out and planted grapes which are still present. While removing the plum trees, four foot deep trenches were dug every 25 feet and new topsoil was deposited in these trenches, Nine units were excavated at Son-600 to test the depth of the midden, to determine the amount of disturbance, and to delineate the boundaries of the site. Their provenience is as follows: Unit Grid Location Excavator 1 N9 E20 Midden Larry Banks; Richard Lerner (1974) 2 N10 E19 Outside ZSAP (1975) Midden 3 NlI E19 Outside WSAP (1975) Midden Midden %SAP (1975 4 N8 E19 120 CA-Son 600 -a Map O ZV_~~~~~~~~~~~~~~~~~~~~~ 0 O ~4 0 0~~~~~~~I .00 ~ ~ 1~/ /1f lit i //Jill I I I I 1111111lii IIIIJI I I11III itlir) 0 2 liii IaI ---- Meters C= T-18 C= T-19 Map 21. Son-600 (legend p. 5). 4 iSAP (1975) WAP (1975) EAP (1975) WSAP (1975) USAP (1975) There was no evidence the above units. There is in each unit excavated. caused by tree roots either burned in situ. of the trenching described by Mauritson in any of ample evidence, however, of extensive disturbance The disturbance in most cases seens to have been ^ having been renoved fran the soil or having been In unit 4 soil disturbance associated with tree removal externis frqn 20 an to over 100 an and is fran 50 an to over a meter in width. The burning of trees is evident in units 7 and 9. Unit 7 contained a charcoal filled pit, 50 an in depth which extended over half the unit. Unit 9 also had large areas where burning was evident, involving over 60% of the entire unit. The soil of all nine units was disturbed even though the disturbance was not always clearly illustrated in the unit wall profiles. The earth was splotchy in color, differentially canpacted, aid the disturbed areas, unlike krotovinas, were discontinuous. Many of the units showed extensive rodent activity, as evidenced by the sharply defined krotovinas. The surface indications of the midden extent were borne opt by the test excavations. The three units within the midden, units 4, 5, 6 averaged 20-25 artifacts in the first four levels while the other units averaged ten or less per level. Almost all of the tools came fran the units within the midden. Site Son-560 (New Bridge Flat) This is a shallow midden site on the east bank of Warm Springs Creek, 175 miles upstream fran its confluence with Dry Creek. The site is 120 meters west aid 22 meters below the Stewart's Point-Skaggs Springs country road. It is approximately a quarter mile upstream fran the New Rockpile Road Bridge which spans Warm Springs Creek canyon. The site is on an meters, in a very steep, Warm Springs Creek. It either direction. open primary terrace, measuring 120 meters by 40 constricted valley approximately 50 meters west of is the only inhabitable area for nearly a mile in The terrace vegetation includes open wyethia, wild iris, lupine, aid much poison are covered with a stunted, moderately dense aid poison oak. oak woodland,, sane buckeye, oak. The slopes on both sides forest of madrone, Oregon oak, The midden covers a small area, 15 meters by 20 meters on the eastern 5 6 7 8 9 121 N7 E19 N8 E22 N8 E23 N8 E25 N10 E17 Midden Midden Outside Midden Outside Midden Outside Midden 122 edge of the terrace at the base of the hillslope. It is south of an intermittent stream which bisects the terrace. The midden is weakly developed, being only a shade darker than the surrounding soil. Both the midden and terrace soil are dark brown, alluvial gravelly silt. Four auger holes were placed on this terrace but no sign of buried midden was found. A one meter square test excavation unit was placed in the midden three meters due north of site datum and two meters south of the cut bank of the intermittent stream. The unit ended at 20 an in a sterile yellow brown, sardy gravel. During the survey a scraper and several chert flakes were found in the bank of the intermittent stream, and a few flakes on the weakly developed midden. Test excavations showed that there was only 10 to 18 an of midden. This single occupation zone contained six chert flakes and one obsidian chip. No diagnostic artifacts were found at Son-560. Central Group Chronology It is possible to deal (partially) with the chronology of three sites in this group. Site Son-598 has 14 housepits so well defined that they must date to a very late period. Since there is no midden it is impossible that it can have any time depth. Therefore we date it entirely Late Houx. Site Son-542, now destroyed, at one time had a Late Houx component arK] also a Borax Lake canponent to judge by the projectile points shown in Figure 26a an: 26b. Site Son-600 apparently also had both Houx and Borax Lake components (Fig. 26c ard 26e). The chert-obsidian ratio in the 9 units excavated there is 1.50 (based on 620 flakes) and variation in this was not great fran unit to unit. We will not tabulate the material excavated at this site since it was so evidently mixed. 75 o~ a C 0 0 0~ ~~~~~~~~0 *0~~~7 LUL m S 0 0 o . LiCSLlBE Map 22. Iter Warm Springs Group. 123 Cf) w 0a z 0 w CD) w 0 z w -J Cf) 4 C.) 0 0 so w 4 a- 0 C,) z 4r -J 0 .0 0 N 125 Lower Warm Springs Group Warm Springs Creek forms a narrower, more restricted canyon, by and large, than does the larger Dry Creek. The exception to this is around the mouth of Little Warm Springs Creek (or Hot Creek) where it opens out a bit for about a half mile along Warm Springs Creek and it is here that we find the important sites. There seems no doubt that ethnographically both lower and upper Warm Springs Creek areas were auxiliary to the larger, more important towns which were further down Dry Creek (Barrett 1908) the sites of which are now under cultivation. The villages in this area may thus have had some specialized function relative to those below. We will discuss this where appropriate but no overall specialization is apparent. Probably this region was more densely wooded and forested than was the open valley of Dry Creek although even this is not certain. In any case a substantial part of lower Warm Springs Creek was apparently given over to woodland grass vegetation. The sites of this group are shown on Map 22 and the eastern part of Map 26. Site Son-594 (Skaggs Springs) Son-594 is at present a surface scatter site in the vicinity of the site of the old Skaggs Springs Resort. It is in an area of several naturally occurring hot springs that flow fran the west bank of Little Warm Springs Creek (also known as Hot Springs Creek), about 300 m up the creek fran its confluence with Warm Springs Creek. Foundations of the hotel, several cabins, and a concrete swimming pool remain to define the site of the old resort. The hot springs are still used by many people and access roads and campsites are nunerous. The surface scatter covers a 200 m N-S by 50 m E-W area on the north edge of the extensive, open valley that is 150 m south of the present hot springs. Most artifacts were found along the dirt road that passes over this site. Only half a dozen chert and obsidian flakes were found here. The surface of this area is covered with a very dense grass ground-cover. The area of Skaggs Springs in general was in open oak woodland but is now dominated by such introduced species as eucalyptus and locust trees. -The slopes on both sides of the valley in which Skaggs Springs is located are covered with dense oak madrone forest. Near this site and about 250 m downstream fran the confluence of the two creeks already mentioned, is one of the few gathering areas for Angelica, a plant held in high esteem by the Pcnos and used both as a spice and a medicine. There is no midden to be seen anywhere on the surface around Skaggs Springs, the prolonged exploitation of the area by those who use the hot springs (the resort dates fran the 1860's), as well as historic cultivation and recent .innabar mining having totally obscured any midden that is here. Both local residents and ethnographic information collected by Barrett indicated that the area of Skaggs Springs was once a major occupation site. Barrett (1908:220) gives the following account. "Kaho'wani, fran ka or aba, water, ho, hot or fire, and wa'ni, ?, at Skaggs SpIirsi on the east bank of Hot Springs Creek, an affluent of Warm Springs Creek. Mr. Mulgrew, the proprietor of Skaggs Springs, has found on this site a number of mortars, pestles, and other large stone implements as well as many arrowheads and smaller implements." 126 Additional ethnographic evidence was collected by David Peri and Scott Patterson. There were told by a Pono (Kashia) informant that different bands used to meet on the terrace on the SW side of the confluence of Little Warm Springs Creek and Warm Springs Creek. As a result of this information, a unit was excavated in this area and 5 m south of the concrete steps that mark the location of the old resort hotel. This unit ended after 20 an because the extensive disturbance caused ..by the hotel had obliterated most evidence of Indian occupation. Seventeen augerholes were made in. the area of Skaggs Springs. Four were made on the east side of the confluence of the two creeks. Thirteen were made in the extensive valley in which the surface scatter was found and extending fran the hot springs up or south to Son-558, a seasonal camp site at the end of the valley. However, in none of these augerholes was found evidence of any midden. Site Son-558 (Bob-l) Son-558 is a shallow midden site found on the West bank of Little Warm Springs Creek about a half mile up the creek form its confluence with Warm Springs Creek and at the SW end of the extensive, open valley in which Skaggs Springs is located. The site is located approximately 50 m up Little Warm Springs fran the confluence of this creek and the first major intermittent stream that enters it. It is on an open, flat primary terrace that is immediately west of and 2.5 meters above Little Warm Springs Creek. There are dirt roads on both sides of Little Warm Springs Creek which continue up the creek. A secondary dirt road passes over the midden deposit at the site and crosses the creek, connecting the two larger dirt roads. The midden is approximately 40 meters north south by 15 meters east west and south west, and the madden extends approximately halfway to the hillslope and grades into non-midden alluvial soil in other directions. The midden is found on both sides of the connecting dirt road that passes over the site, although most is south of this road. The soil of Son-588 is a very dark gray brown, loose gravelly silt. The surrounding soil is a dark gray brown to gray brown gravelly silt. There are several areas in the vicinity of this site that contain dark-colored soil, but only at Son-588 were artifacts found. These other dark soils may be caused by there being more organic constituents, such as leaf litter, in their soils than the surrounding soils. Auger holes were made in these darker soils but they indicated only 10-20 an of dark color, underlain by the general alluvium. Auger holes were also put in the extensive terraces to the north and east, but no midden was found. The vegetation of the site is a mixed riparian/open oak woodland'. Daminant trees are bay, Oregon oak, alder, and maple with wild grape an buckeye along the creek. Little Warm Springs Creek probably maintains a year round flow. The hot springs of Skaggs Springs are, approximately 500 meters downstream. This site has been disturbed recently only by the aforementioned road 127 which passes over its northern edge. The flat terrace upon which this site is located doesn't appear to have been disturbed by plowing. The site is on the outer periphery of the major historic site of Skaggs Springs Resort, a hotel and spa which was active fran the 1860's on. However, the site does not appear to have been disturbed by any resort activities. Erosion is fairly severe along the cut bank of Little Warm Springs Creek. A 1 x 1 meter test unit was located at 5 m east of the site datum stake in what appeared to be the center of the midden. The unit ended at 50 ans in a sterile, loose sandy gravel. No augering could be done in the floor of this unit because of the looseness of the soil. The nearby cutbank showed no buried midden below 50 cms. Son-558 contained only 20 an of a loose, dark gray brown, gravelly silty midden. The first level of this midden was richer than the following levels and included a small obsidian side notched, wide stemme-d projectile point, a chert biface, a chert retouched flake, an obsidian retouched flake and a quartz core. The debitage count, after a high of 12 chert and 15 obsidian in the first level, rapidly diminished with depth, and chert and obsidian debitage were found in roughly equal amounts in all levels. Son-558 also contained a relatively large amount of firecracked rock in the first two levels, i.e., 31 lbs. and 23 lbs. respectively. None was found below 20 cm. Animal bone was found in a moderate amount in the first level, 26 fragments, after which only a few pieces occured in each level. The only diagnostic artifacts from the excavation is the small side notched proj. point fran the 0-10 an level. This was inadvertently anitted fran the illustrations but undoubtedly is a late Houx specimen. Site Son-561 (Buzzard Flap Blind) Son-561 is an isolated hunting blind site that is situated approximately 300 m south of, and approximately 60 m upslope fran, Skaggs Springs. It is on a sall ridge between two well watered clearings about half-way up the hills on the west side of Skaggs Springs valley. Son-561 is on the north facing brow of a small transverse ridge between two gently sloping clearings that are watered by apparently year-round springs. The blind is about 40 m fran the west clearing, and about 60 m from the east clearing. Two intermittent drainages trend north fran these clearings for approximately 250 m to the area of the Skaggs Springs thermal pools. The area of the site, which is approximately 6 x 6 m is surrounded on the north, east, and west sides by mixed, open oak/madrone woodland, with some manzanita, bay laurels, and buckeye. T[o the south, the slope becanes more steep and much more heavily wooded with predominantly Oregon oak and madrone, and with Douglas fir increasing upslope. The soil of this area, including that of the site, is a brown to red- brown silty sand with much colluvial gravel. Logging roads have made some changes in thae original land contours in the general area of Son-561, but the Immediate area of the site is undisturbed. There was, however, much disturbance throughout the blind because of a thick matting of bay laurel rootlets fran the nearby tree. 128 The pit of the hunting blind at Son-561 was measured at 1.5 m N-S by 1.05 m E-W, and was as deep as 42 cm. There were many large and mediun sized rocks, some of which were placed on others, found along the northern ard western, or downslope, perimeter of the blind. There were many other rocks that were either within the pit or downslope, apparently having fallen fran the rocks that once lined the north and west sides of the blind. These rocks need not have formed a continuous wall, but may have been used to anchor brush screening. At the present time there are game trails on the east side and on the west side of the blind, both being within 7 m of the blind. There is also a game trail below, and 20 m north of, the blind that goes between the two clearings. The alignment of rocks on the north and west isde of the blind may indicate that aboriginal hunters expected deer and other game would approach the blind fran these directions where even now game still passes. The general procedure used in testing a blind was to first clear the grass ground cover. After the grass was removed, the ground was scraped with a trowel to a hard, undisturbed surface and this detritus was screened in a quarter inch mesh screen. All large rocks were left in situ. This clearing method was continued outside the actual pit to encompass the rocks that had scattered downslope fran the rocks lining the perimeter of the blind. The purpose at this time was to remove the soil and leaf litter that had filled in and obscured the original pit contours, and to find any artifacts that might occur in this soil. The clearing of this blind expanded and deepened the dimensions observed earlier so that the pit now measured 3.8 by 3.1 by 54 cm deep. The thick matting of roots throughout the pit loosened the soil and made it difficult to be fully certain of the hunting blind's original dimensions. No artifacts were found and no other large rocks fran the perimeter of the blind were found.- Son-561 will not be subject to innundation. However, because so little is known about hunting blinds, more excavation should take place. The blind could perhaps be trenched to better determine Ltratigraphically the original dimensions and contours of the pit. In addition, intensive surface collections should be done on the surrounding slopes in order to recover any possible artifacts. Son-573 (Blue Blind Site) Son-573 is a possible blind site situated on the north bank of Warm Springs Creek. It is on a flat shoulder that is 250 m north of and 45 m above Warm Springs Creek. It is up the second intermittent drainage below the confluence of Little Soda Creek and Warm Springs Creek at Son-556. The blind is on a gently sloping, 30 x 40 m open, flat clearing in an-otherwise steep, moderately wooded terrain. The blind is on the southern edge of this clearing and is oriented towards the rest of the clearing. The blind is a rectangularly shaped, rock-lined depression that measures 2.4 m by 1 m and is 30 cm deep. There are several rocks lining its north and east sides which may have been used to anchor brush screening. The vegetation of the site is open grassland with a thick ground-cover of 129 annual grasses and scattered live oak, madrone, and a few digger pines. Along the intermittent drainage are scattered bay, Oregon oak, and live oak. No artifacts and no midden were found at this site. Site Son-556 (Mrs. Walker's Mortar) Son-556 is a deep midden site with a housepit ard an associated surface scatter on the north bank of Warm Springs Creek and on the northwest isde of the confluence of Warm Springs Creek and Little Soda Creek, the latter being a local name for a major intermittent tributary a half mile up stream fran Skaggs Springs. Son-556 is located on an open, extensive primary terrace that is 100 m north-south by 40 m east-west. Across the intermittent creek is another large primary terrace of 40 m north south by 100 m east west. The site is adjacent to and 3-4 m above Warm Springs Creek. It is north of, and across Warm Springs Creek fran the Stewarts Point - Skaggs Springs county road. Three dirt roads pass directly over the midden. The first and largest continues up the north bank of Warm Springs Creek, while the dirt road caning down fran Little Soda Creek form a Y about 35 m before its junction with the largest of these roads. The midden covers an area of 40 by 35 m. It is bordered to the south by the cutbank of Warm Springs Creek, to the east by the bank of Little Soda Creek, to the west by the steepening hillslope, and grades into non-midden soil just past a point where two dirt roads meet and continue up Little Soda Creek. The housepit is within the midden area and about 5 m north west of the site datum stake. There is a surface scatter of chert, obsidian, and homestead artifacts covering nearly the whole surface of the primary terrace upon which the site is found. This scatter measures about 100 by 35 m. The soil of this site is a dark gray brown, loose gravelly silt. The surrounding soil is a brown gravelly silt. Six augerholes were made in the vicinity of the site. Three were put in the extensive terrace across Little Soda Creek fran Son-556. There was no midden found. Two were put in-the terrace north of the site. The first, about 15 m north of Unit 1, indicated only weak midden until 75 an, then darker colored probable midden until at elast 110 cm where the auger bottomed out on rocks. The second of these two, located 70 m north of Unit 1, also found what may be weakly developed, buried midden fran 30-110 an. No artifacts were found in this buried midden, although a firecracked rock fragment was found in the first of these augerholes at 100 an. The sixth augerhole, 100 m north of Unit 1, found no midden. The vegetation of this site is now open grassland with riparian vegetation along both drainages. The dominant trees around the site are coast live oak, bay laurel, and Oregon oak, with some wild grape. On the slopes to the north, there is open oak woodland with Oregon oak, and live oak. On the opposite bank of Warm Springs Creek is a dense oak-bay forest. The surface of Son-556 has been subject to much disturbance. As mentioned, three roads pass over the midden. The largest has been recently bulldozed 10-20 an into the midden. There has also been a 1 m diameter fire 130 I 0 0 0 .0 . 0 . .00 **0 00 *0 . 0 * *0 00 0 0 .0 * 00 0 0 0 0 0 0 0 00 * 0 1J00 0. *0 .0 0 * 00 g 0 0 0 0 0 ** * 0 0 .0 00 .0 * 0 . 0. 00** o o o * . 0 .0 * * 0 0 . 0 00 0 0 0 * 0 0.0 0 . 0000 * o . . 0 * .0 . 0 0 0 * 0 o* *0 0 0 .0 .0 o*0 * 0 0 .0 * .0 *00 .0 . * . 0 00 0 0 . 0 0 0t 0 0 0 0 0 0 0 0 0 00 . 0 0.00 0000 0 00 * .0 ..0 * 00*0 0 * . .0 * *0 O 0 0 0 0 .0 0 .0 *00 0 & . 0 * 0 .0 0 00 * 0~~~~ * 0 0 0 a * @0 * *o0 .0 00 0*0 *0 0 * 0000 0 *e 0.0 0 .0 0* 0 ** . 0 0 0 0 ~~~00 0.0 O - 0 * 0 0 00 * 0 0 0 0 .0 0 *0 . 0 0 O O O00 00 O 0 0 00 * *0 *aC 0 00 * \% .00 .0 0 o e~~~~~~~~ I'0 O 0 00 * 0 0 0 0 *0 *0 . 0 0 .0* 0 0 NZ W 0 0 000 0 00 0.0 000 *.o 0 0 0 0 0 0 .0 II0 O 0 O I 0 1/0 0 *0 /10 CA Son 556 MAP 4, 0 10 20 Meters 0 /I I 1 I I'I I I I I 0 Map 23. Son-556 (legend p. 5). I. 131 A. Midden; dark grayish brown; dry silt with gravels; friable. Annual grass root zone. B. Yellowish brown; sardy clay lens; ccznpact. C. Midden; dark grayish brown to very dark gray; dry silt with gravels; moderately friable. D. Kortovina disturbance. E. Midden; very dark gray to very dark brown; sandy silt with gravels arn cobbles; moderately caopact to friable; slightly moist. F. Rock ard charcoal concen tration. G. Transitional zone; dark grayish brown to dark yellowish brown; silt with sand and clay; moist; campact. H. Sub-midden; dark brown; clayey silt; moist; campact. -220- Figure 17. Son-556 Profile. 132 Site Son-556 (Mrs. Walker's Mortar) Level Proj. Bifaces Reworked Scraper Retouch Used Debitage an. Pts. Bifaces Flakes Flakes 0-10 1-c 1-c 6-c 12-o 10-20 1-c 19-c 26-o 20-30 1-c 5-c l-o 1-o l-o 20-o 30-40 1-c 6-c 1-o 10-o 40-50 1-c 8-c 1-o 21-o 50-60 1-c 8-c l-o l-o 33-o 60-70 1-c 2-c 10-c 1-o 1-o 2-o 33-o 70-80 4-c l-o l-o 30-o 80-90 1 drill 6-c tip-o l-o l-o 37-o 90-100 4-c 2-o 2-o 1-o 2-o 55-o 100-110 1-c 9-c 2-o 2-o 1-o 1-o 65-o 110-120 1-c 3-c 14-c 1-o 1-o 90-o 120-130 1-c 1-c 20-c 5-o 64-o 130-140 1-c 17-c 3-o 2-o 1-o 63-o 140-150 1-o l-o 1-o 40-o 133 Proj . Pts. Site Son-556 (Mrs. Walker's Mortar) Bifaces Reworked Scraper Retouch Bifaces Flakes 1-o 1-o 1-c 1-o 1-o 1-c 1-o Totals 3-c 12-o Graix] 15 Totals 1-c 1 l-o 11 3 3-c 3-c 4-o 7 3-c 13-o 16 Used Debitage Flakes 16-c 2-o 55-o 12-c 5-o 54-o 1-c 21-c 1-o 27-o 14-c 18-o 14-c 17-o 18-c 10-o 7-c 1-a 5-o 11-c 238-c 19--o 785-o 30 1623 'Level an. 150-160 160-170 170-180 180-190 190-200 200-210 X210-220 - - 134 Site Son-566 (Mrs. Walker's Mortar) 1-c 1-o 1-c 1-o 1 Level Cores Hand an. Stone 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 190-200 200-210 Total 2-c 2-o 1 Mil liwg Stone Mortar Fired Rock 15 31 5 11 1 1 1 1 1 3 2 25 10 13 108 36 37 16 24 23 33 65 45 22 22 23 5 6 575 Animal Bone 9 24 2 4 11 14 13 20 17 15 45 31 26 50 10 29 19 8 4 7 6 364 135 pit recently dug into the midden about 3 m north of Unit 1. Additionally, there was probably a homestead located in the vicinity and the extensive terraces have certainly been plowed, although not recently. Homestead activity on the site is indicated by numerous ceramic and bottle glass sherds found scattered over the terrace. Also, an historic dump was found in a cutbank of Little Soda Creek about 100 m north of the midden. Erosion on the site is fairly severe at the point of confluence of the two creeks and moderate along the cut bank of Warm Springs Creek. A 1 x 1 m test unit was located on the least disturbed, darkest part of the surface midden at south 650 west, 9 m. from the site datum stake. No wet screening was done. The unit ended at 220 cm because the thpe allotted for the test excavation phase of the Warm Springs Project had elapsed. The artifact count had greatly decreased by 220 an, but the dozen artifacts found in the last level gave reason to believe that this unit would have gone deeper. An augerhole was made in the floor of the unit. It continued for 50 more cm in the same soil as found in the last level, i.e., a dark brown clayey silt. There followed a lighter colored, coarse silt until the end of the auger at 35 cm. No artifacts were recovered from the augered soil. Son-556 at 220 an was the deepest site excavated in the south section uf the project. The midden was rich in artifacts and especially so in the strata containing Houx Aspect artifacts. Another significant fact was that obsidian artifacts consistently, and sometimes greatly, outnumbered chert artifacts. Tentatively, there were five canponents identified at Son-556 after test excavation. There was an upper midden component extending from 0-50 an canposed of a moderately loose, dark gray brown, gravelly silt. In the north quarter of the unit, it ended on a yellow brown, canpact, sandy clay lens. This lens may have been the floor of a dwelling. This stratum contained few tools and only two diagnostic artifacts. The latter included a corner notched opsidian projectile point fran the surface, aid a fragment of another projectile point from 40-50 cm (probably a Gunther barb). Other tools included a chert biface fran 0-10 an, a nosed chert scraper from 10-20 ant an obsidian scraper fran 20-30 an, and four retouched and four utilized flakes. The debitage in this component generally numbered under 40 artifacts per level and chert to obsidian ratios generally were.60 or less. There was a sparse to moderate amount of fire-cracked rock with the highest amount, 31 lbs., occurring at 10-20 an level. Animal bone, which was consistently found in all levels fran 0-200 an, was also present in this component in sparse amounts except at the 10-20 an level where there were 24 pieces. The next tentative component is fran 50-110 an in a stratum of less loosely compacted, very dark gray to very dark gray brown, sandy gravelly silt with many rocks. This contained a possible Excelsior point base from 70-80 an, and either an aberrant form Qf, or a presursor to, a Gunther point at 60- 70 an (Fig. 27b). There were numerous other, less diagnostic tools in this stratum including two reworked bifaces (a chert and an obsidian core from 70- 80 aid 80-90 an respectively. There were also five grinding stone tools which were concentrated around the 70-80 an level. These include two milling stone frag., a hand stone frag., and two bowl mortar frag. Fran this same level, there were 108 lbs. of fire-cracked rock, the highest amount of fire found in 136 this unit and indicative of the concentration of rock at this level. The next component occurs fran 110-170 an in a stratum of moderately canpact, very dark brown to very dark gray brown, sandy gravelly silt. This canponent also contains many tools and is differentiated fran the previous component by the presence of serrated projectile points. Serration is seen by Fredrickson "as an attribute that becanes less frequent" as one moves forward in time (Fredrickson 1973:99). Two Excelsior point bases fran 120-130 an and 160-170 an were found as well as a medial fragment and two point tips, a serrated point base fran 100-110 an, and a serrated tip and a serrated medial fragment fran 130-140 an. Other tools include 8 reworked bifaces, 7 scrapers, and 8 retouched flakes (including 5 fran one level, 120-130). No definite grinding stone implements occurred in this canponent. The debitage count increased in this canponent to an average of over 70 pieces per level with a high of 104 pieces at 110-120 an. The chert to obsidian ratios remain about the same as previous levels. There were moderate anounts (i.e., 20-40 lbs.) of fire-cracked rock per level except for the 140-150 an level which contained 65 lbs. Animal bone was also present in consistant, moderate amounts. Finally, at 140-150 an there was an ananalous sliver of bottle glass found, but this probably fell into the unit fran above. The fourth tentative canponent is in the "transitional zone" fran 170- 200 an, a canpact, dark gray brown to dark brown clayey sandy silt. The only diagnostic artifacts are an Excelsior point base, a serrated Excelsior point, and a milling stone fragment all of which are fran 170-180 an, and a reworked, small sharply shouldered point fran 190-200 cm. All other constituents diminish fran previous levels and with depth. The chert-obsidian ratio increases rather markedly. The last component is poorly defined and extends fran 200 an to the end of the unit at 220 an. As augering showed, this component could continue as much as 50 an deeper since the same soil continued to that depth and artifacts were still occurring at 220 an. This component is in a dark brown, clayey silt. The only distinguishing factor in this stratum is that chert for the first time outnumbered obsidian debitage. As seen fran the preceding description, Son-556 is a very rich and canplicated site and potentially the most important site in the southern area of the project. Its deep, well developed midden will aid in establishing a stratigraphic sequence covering the past three thousand years or more. The housepit on the surface and the possible sandy clay floor may provide valuable information about settlement patterns. The site can also help in answering why a site like Son-551, which, at least in part, is contemporaneous with Son- 556, would consistently have much more chert than obsidian artifacts whereas Son-556 contained much more obsidian than chert. Site Son-557 (Pig Skull Dump) Son-557 is a disturbed, shallow midden site on the south bank of Little Soda Creek (which is a major intermittent tributary to Warm Springs Creek, located a half mile upstream fran Skaggs Springs) 600 meters up fran its confluence with Warm Springs Creek. At the present time a dirt road which proceeds up fran Warm Springs Creek at Son-556 along the south bank of Little 137 Soda Creek stops at this site. A bulldozed garbage pit is located here and is still used by a local inhabitant. This pit, and the subsequent bulldozing made in the area to cover up the garbage, has destroyed most of this site an] disturbed the remainder. The midden was situated on a flat, open terrace 10 meters south of and 1.5 meters above Little Soda Creek. The midden covered an area of 9 by 9 meters when found in the initial survey. At the time of test excavation, some 6 months later, the midden was half that size and the pit had became larger. The midden was bounded to the north by Little Soda Creek, to the south by the hillslope, to the west by an ephemeral drainage ard to the east by the garbage pit and bulldozing. The area of Son-557 is densely forested. The slopes to the south of the site are covered in a dense oak-bay-madrone woodland. Just to the north, along Little Soda Creek, is a dense riparian growth of bay, live oak with some buckeye and wild grape. The site is now open grassland, but scattered stumps indicate that this was once open oak woodland. At the time of excavation, in late June, all the streams in the area had dried up except Little Soda Creek, which maintained a slight flow. This creek also probably dries up by August. The soil on the site was a weakly developed midden of dark gray-brown, loose silty gravel. The surrounding soil was a shade lighter - a gray-brown, loose silty gravel to gravelly silt. Augering was impossible in this soil because the soil was too loosely packed to stay in the auger. A 1 by 1 meter test unit was located 5 meters west of the garbage pit and 2 meters due east of the site datum stake, in an area that was less disturbed than the surrounding area. This unit ended at 40 an in a sterile yellow-brown loose gravel. Son-557 was a shallow midden site with very few artifacts. Unfortunately, most of the midden deposit had been destroyed and the test unit had to be located on the periphery. The result was that only one chert core, 8 chert flakes, 3 obsidian flakes and 3 animal bone fragments were found in 20-25 an of midden. Son-557 was probably a seasonal camp site, located 14 miles up Little Soda Creek fran a major occupation site, Son-556. Approximately a mile up Little Soda Creek fran Son-557 is another seasonal camp site with shallow, weakly developed midden, Son-559. The econamic function of this site is unknown and its present state of disturbance preclude3 further excavation for discovering it. Site Son-559 (Little Soda Creek) Son-559 is a shallow midden site situated on the north east side of Little Soda Creek, a major intermittent tributary to Warm Springs Creek, and is about a mile and a half up this tributary fram Warm Springs Creek. The site is roughly midway between Warm Springs Creek and the top of the ridge whcih divides Warm Springs Creek and Dry Creek. It is on a 50 x 20 m flat shoulder surrounded by steep hillsides. It is between Little Soda Creek to the west and south, an intermittent drainage to the east and south, and is 138 limited to the north by the increasing hillslope. It is about 75 m below ard 250 m southeast of a large flat meadow on which an old house foundation stands. A dirt road proceeding down fran the top of the ridge passes through this meadow. Most of the midden has been eroded away and there appears to be only a small 8 by 10 m remnant left on the north west edge of the flat shoulder and immediately next to Little Soda Creek. The rest of the flat has a sparse to moderate surface flake scatter on a very thin, rocky soil covering serpentine bedrock. The midden here is a very dark gray brown, rocky, sandy silt with angular serpentine rocks which increase with depth. The surrounding soil is a lighter gray brown, sandy silt colluvium with many serpentine rocks. Augering could not be done due to the extreme rockiness of the soil. The vegetation of this site is open grassland with sparse groundcover and scattered shrubs such as toyon, spice bush, scrub oak and manzanita. There is a narrow growth of riparian vegetation along Little Soda Creek with bay, some madrone and coast live oak. In the surrounding area, the vegetation is open oak woodland with Oregon oak and live oak dominant and some buckeye and madrone. Little Soda Creek appears to be a year round stream at this elevation. There was a steady flow of water in it during the time of excavation, early July. This evidently canes fran a spring in the meadow above the site. There appears to be no man-made disturbance at the site. Excavation here showed there was some root disturbance within the midden. Erosion is and has been severe, and it seems most of the midden, which surface scatter indicates once covered this flat, has been washed away. A 1 x 1 m test unit was located on the midden remnant at 15 m due west of the site datum stake aix 3 m northeast of the bank of Little Soda Creek. The unit ended at 40 cns on serpentine bedrock in the western half of the unit and in a decanposing serpentine soil in the eastern half. Son-559, initially thought to be a surface site, had 40 cm of fairly rich midden. A surface collection made during the initial survey of this site included an obsidian tip of a projectile point, a chert concave base projectile point, a chert side-notched triangular projectile point, and a chert Excelsior point. Despite the pranising surface collection, test excavation found no diagnostic tools. The only other tools found were a medial fragment of an obsidian projectile point at 20-30 ac, three utilized chert flakes fran 0-10 an and 2 utilized obsidian flakes fran 20-30 an. Debitage was 21 obsidian and 7 chert at 20-30 an, and 13 obsidian and 3 chert at 30-40 an. There was no obvious fire-cracked rock found in the excavation. 3 shell fragments were found in the first level and there were several bone fragments found in each of the 0-30 an levels. Charcoal flecks were present in 0-30 cm levels and were abundant in the first level. 139 Site Son-555 (oody Woodpecker) Son-555 is a shallow midden site situated on a primary terrace on the south bank of Warm Springs Creek at the confluence of this creek and Seven Oaks Creek. The site is about 20 m south of the Stewarts Point - Skaggs Springs county road on an open, flat area that once was a homestead site. A dirt road goes through the site fron the county road to a man-made, damn1]-up pond 50 m to the southwest. The foundation of a homestead was found 30 m south of the datum stake, and an old redwood fence encloses the 50 x 70 m open terrace on which this site is located. There are also numerous introduced species of trees such as eucalyptus, fig, lilac, anx] apple found in the fenced-in flat. The surface midden of this site was approximately 35 x 30 m in area. It is bordered to the south by a thick growth of fig trees, to the west by a wire fence, to the east by several small fruit trees, and exterded to within 10 m of the county road to the north. The vegetation of the area of Son-555 is mixed riparian anix oak/madrone woodland. The dominant native trees on the site are Oregon and live oaks, bay, madrone, and maple with some manzanita. The slopes in back of, or south of, the site are in a dense oak/bay/madrone woodland. Wyethia, a plant whose seeds were used as a food source, was abundant on the site. Basketry sedges are found 50-100 m down Warm Springs Creek and on the opposite bank from this site. The soil of Son-555 is a dark gray brown, loose, sarxny gravelly silt. The surrounding soil is a much lighter brown sandy silt. Four auger holes were made on the site and indicated the site boundaries already mentioned. In the east part of the midden where it was possible to auger deeply, there was 30-50 an of weakly developed midden. Son-555 is one of many sites that were first occupied by the aboriginies, and then by the homesteaders. Other sites in the South Section that show this double occupation are Son-544, Son-556, aix Son-594. The midden is considerably disturbed both by the homesteading activities aix] by more recent road building. There are recent bulldozer tracks on and around the midden. A 1 x 1 m test unit was located in the least disturbed part in the approximate center of the surface midden at N 100 W, 8 m from the site datum stake. No wet screening was done at this site. The unit erned at 50 an in a sterile sandy cobble layer. Son-555 was a shallow site with weakly developed midden that showed a heavy admixture of homestead artifacts along with aboriginal artifacts in the first two levels. The looseness and occasional ashiness of the soil inxdicate that the first two levels were heavily disturbed. Artifacts from these first two levels include a round bottomed projectile point base (probably an Excelsior point), 3 retouched flakes, 3 utilized chert flakes, 1 obsidian utilized flake, and a chert core as well as square and round nails, a shell button, glass and ceramic sherds, and an iron crank hardle from 10-20 an. The debitage count was 23 chert and 2 obsidian from 0-10, 26 chert anid 6 obsidian from 10-20 an. 140 Site Son-555 (Woody Woodpecker) Level Proj. Retouch Used Core Debitage Fired Animal an. Pts. Flakes Flakes 1bck Bone 0-10 1-c 1-c 1-c 1-c 23-c 6 2 1-o 2-o 10-20 2-c 2-c 26-c 5 1 6-o 20-30 3-c 40-c 4 1 3-o 30-4 0 2-c 1-c 16-c 40-50 2-o 2-c Totals 1-c 5-c 6-c 1-c 107-c 15 4 l-o 13-o Grand 1 5 7 1 120 15 4 Total I 141 The levels fran 20-50 an were much less disturbed and only one historic artifact was found. Very few tools were found (2 retouched and 4 utilized flakes) aid the debitage count, after a high of 40 chert and 3 obsidian in the first level, decreased rapidly. Very little fire-cracked rock or bone were found at any level. Son-555 is one of the few sites that is located on the south, or shady side, of Warm Springs Creek. This site's location may indicate that is was a special-use site, aid not necessarily an occupation site. It could have been psed as a camp site for gathering and processing either basketry sedge or Wyethia, both of which occur locally, or used as a camp site while the chert quarry was being used. Site Son-553 (Double Black Dirt Delta) Son-553 is a deep midden site with one well-developed housepit on the north bank of Warm Springs Creek a mile downstream fran its confluence with Rancheria Creek. The site is 500 m down Warm Springs Creek fran the Baxter Ranch. It is on a primary terrace located in the only substantial widening of Warm Springs Creek Valley in the general area. For almost a mile upstream and for over a mile downstream Warm Springs Creek is in a steep, V-shaped valley with no floodplain. T'o hundred and fifty meters downstream fran Son-553 the creek flows throuah a very constricted canyon formed by a resistant facies of Franciscan graywacke aid by numerous large boulders of this formation. There is now a waterfall in this canyon. It is very likely that a rockslide at one time dammed up the creek, forming a lake in the area of the extensive terraces. These terraces, which are found on both sides of the creek, are 1.5 m above the creek and on the north side, cover an area of 150 m NE-SW by 70 m NW-SE. The midden occurs in two separate ateas; the largest aid darkest midden is 40 x 40 m aid is found on the SE edge of the terrace. It is bordered on the SW by the steep hillslope, on the SE by the cutbank of Warm Springs Creek, on the NE by an intermittent drainage, and imperceptibly grades into non- midden soil to the NW. The secoid midden is smaller and lighter in color. It is 20 m NE of, and across the intermittent stream fran, the darker midden area. This midden covers an area of 32 x 20 m. It is limited on the NE by an ephemeral drainage, on the SE by a barbed wire fence, on the SW by the intermittent stream, and, as in the larger midden, grades into non-midden soil to the NW. On this smaller midden is found a housepit that measures 4.5 m in diameter and 40 cm in depth. The soil in the larger darker midden is a brown to dark brown, loose, gravelly silt. In the smaller midden with housepit, the soil is a gray brown, compact, clayey gravelly silt. Over the rest of this extensive terrace, the soil is a brown, gravelly, clayey silt to silty sand. Three auger holes were made in the area of Son-553. Two of these were made NE of the midden areas and near the cutbank. Neither showed midden present. A third auger hold was made 1 m south of the large housepit. It indicated 25 an of weakly developed midden, then non-midden terrace deposits. Son-553 and the surrounding terraces are in an area of mixed riparian vegetation and open oak woodlaid. Bay laurel, live oak, Oregon oak, cottonwood, aid manzanita are found here. On the north slopes of Warm Springs * 0 0 0 ! * 0 0 * 0 0 0 * 0 0 * 0 0 0 0 * 0 0 0 0 10 20 410 fp i . * Meters * 0 0 * 0 * 0 DMI7~ 0 0 0 l a 0 0 * 0 I.., 0. 0 * 0 * 0 * * 0 0 * 0 * 0 * *9 0 he .I CA-Son- 553 MAP jI 0 Map 24. Son-553 (legend p. 5). 142 fri* f 1. . 10 0 0- 0 * * * j0 * a. . .111 0 0 --.%, oll'o -10- ~20 630 40 100~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Ridden; brown to dark brown dry; fine-grained silt with gravels; friable tQ very friable. Annual grass root zone. 5idden; dark brown; silt with gravels; moist; moderately canpact. Dark yellowish brown; moist; canpact. Ridden; brown to dark brown; silt with sar~i arx] gravels; moist; canpact. Transitional zone; dark yellowish brown; silt with sard aix] gravels; moist; ccmapact. Sub-midden; yellowish brown; clay with silt, sarx] aix] gravels; moist; very canpact. Yellowish brown sard; moist; moderately cCmpact. Irotovina disturbance. Daccinposing sarxdstone. Figure 18. Son-553 Profile. i v V v 0 0 0 u 1) 1 AQ dta ao Ln 00 aN v LA LA CN* HI N (NI U) U) Ui) U) H 0) OH QU MA A L?nr-,~ 4ILI r l U) Uo) to 0 me HI rH )00 A A 0 m C 0 O LUI~ UV o LU C O LA (N N Y LA co 0 ko (N (NI In m to () 2 L 6or 0 lor u I Im L A 0 (I 0 0 0 0 II I, (NI H4r1 Q00 II H.LA H- 0 H- 0 II1 H(, NI 0 Hl 0 A. 0 1. U( U) o HH 0 H 04r-I .-I 0 o II HNI 0 (N 0 HD 0 H1 (N In N 0 00 0 Q r- 0 (A 0 00 144 U) co .I- CZ.. 0D (N (N %O '0 H r-I a) U) a) Y 4) U) H a) co Y~ 8 wrz a) H 4a) r-i 4) U) 0 1N 0 H I 'n .D . LA NI V., 5i (N . (NI Eq. Ln (N LA (NI (00 I CN (NI 0 H 0 0 0 (00 )00 (NI 0Uc) ?;m .rq o U 0 H 0 0 H. 0 0 H 0 2 II 11 A '.i I I 145 Creek, an open oak woodlard of Oregon oak and live oak, with some madrone occurs. On the shadier south slopes, a denser forest of madrone, bay, Oregon and live oak is found. The dammed-up lake which was present at some point during the occupation of this site would have been used as a fishing source, and possibly some of the many shell fragments found during excavation may have been fresh-water mussels gathered here. There exist now several large, deep pools in the general vicinity of Son-553, most notably the plunge pools formed by the waterfall downstream in which are found a large number of trout. The only modern man-made disturbance found on the site is a barbed wire fence passing through the larger midden area. Farm buildings and a corral are found only on the NE erd of the extensive terrace. There was very considerable rodent disturbance found during excavations. Erosion on the site is minimal. Two test units were excavated on Son-553. Unit 1, a 1 x 1 m unit, was placed on what looked to be the center of the large, dark midden area at 2 m due south of the site datum stake. The unit ernded at 100 an in a sterile yellow brown sanx that was honey-combed with krotavina. An auger hole was placed in the middle of the floor of the unit arn went to 230 ans. It contained only alluvial sands and had no cultural material. Test Unit 2 was a 50 x 250 an unit that was located in the large housepit in the smaller, more northerly midden. The unit externded from the center of the housepit eastward to beyond the rim. It was located at 24 m N 60 W of the site datum stake. An attempt was made to excavate this unit stratigraphically, and to search for housepit floors. Unfortunately, this excavation took place in July when the ground had become dry arn compact. There was little differentiation between soils so stratigraphic excavation was impossible and floors could not be differentiated from the normal soil. Therefore, excavation at this unit was stopped after 20 a. Son-553 is a deep, rich midden site that contained two constituents not found in large amounts at other sites. It contained a relatively large number of artifacts made of 19th Century bottle glass, and the largest amount of shell found at any site. Bone was also abundant at this site. Excavations at Unit 1 indicate that there may be two or more separate cultural components found at this site, an upper midden component ani] a lower midden component. The upper midden component, from 0-20 an is characterized by the co-existence of while anid Indian artifacts. There were found in these first two levels, an obsidian corner notched projectile point, as well as two other point fragments, an obsidian burin, an obsidian scraper, a scraper made from bottle glass, 4 retouched flakes, 20 obsidian utilized flakes anx 3 bottle glass utilized flakes, ard 3 square nails. The debitage from these first two levels included 7 glass flakes and had nearly equal amounts of chert and obsidian. This component was further characterized by a very large amount of animal bone, with 220 pieces in 0-10 an, and 98 in 10-20 an. It also contained a large amount of charcoal, arn a moderate amount of fire-cracked 146 rock, and a small amount of shell. Stratigraphically, this camponent occurred in a loose, brown to dark brown silt with some gravel. The lower midden component, from 20-80 an, contained two different strata; one, a dark brown, slightly compact, gravelly silt that extended about 50 an, and the second, a lighter brown, more canpact, sandy gravelly silt. Stratigraphic relationships were obscurred because there was a great deal of rodent disturbance in these strata. The lower midden component contained only one diagnostic artifact, an obsidian corner notched projectile point fran 70-80 an. Other tools fran this component include a chert drill tip fran 50-60 cm, a chert biface fran 20-30 an, 5 chert retouched flakes, and 2 bone awl tips from 70-80 an. Two flakes fran bottle glass were found at 50-70 an, but this low position was almost certainly due to rodent disturbance. The debitage in this canponent showed that chert consistently out-numbered obsidian. In the 20-50 an levels, the ratio of chert to obsidian averaged 3:1; in the 50-80 an levels the ratio was 9:1. There was less charcoal arn less bone in these strata, although the bone consistently numbered 45 to 70 pieces per level fran 20-70 an. This included a dog maxilla, 3 deer teeth, and the distal erd of a right human femur from 50-60 cm. There are also a large amount of shell found in this lower midden component. Most of the shell was concentrated in the 40-60 an levels and consisted mainly of marine shells such as Mytilus, ard Saxidanus. It is not known whether shell was preserved at this site because of the soil chemistry or simply because the site was so late in time that the shell had not yet deteriorated. The transitional stratum between midden and sterile, fran 80-100 an, has also been extensively disturbed by rodents. The only tool found was a small straight based projectile point, found in a krotovina at 100 cm. This layer contained a rapidly diminishing quantity of all constituents until the sterile sand at 100 an was reached. Site Son-554 (Warm Springs Lapidary Center) Son-554 is a chert quarry site with an associated surface scatter found on the south side of Warm Springs Creek along both sides of an intermittent stream tributary to Warm Springs Creek. The site is at approximately 550 feet elevation. The quarry which extends along both banks of the intermittent stream is in a large, open, flat area in an otherwise hilly, wooded terrain. This flat was once the site of a homestead and the foundations of the homestead and several large eucalyptus trees are found 15 m south of the center of this quarry. There are several old redwood stake fences in the area, one of which passes through the major chert source of this site. Son-554 is bordered to the east by another of these fences, and to the west, 250 m up the intermittent stream, by a more modern wire fence. The site extends 20-25 m on both the north and south sides of the intermittent stream. On the SW edge of this area, there was a small, sparse surface scatter of fire-cracked rocks and 3 retouched chert flakes located "'a I I I'' l l Red.. I I ~Qua. 'I ' I - f yij G k, , " . . - ) . 1-3 z Chert Qi CA-Son- 554 Ii MAP lI Extent Of Site Area 0 20 40 Meters uarry Area I z ~ ~ I ~~~~~~~~~~ Z :7 K * : ; _ /) Map 25. Son-554 (legend p. 5). 147 #' I,. kMi. j / / I / I - ..%, A -1 .., - - .... .1, k4 .  -=- . A I.,.-6. .., A.. toild. , -:h- .... .1. 148 about 35 m south of the intermittent stream and along the wire fence. The surface scatter was about 10 x 15 m. The total area covered by this site is about 250 m E-W by 50 m N-S. The quarry occurs in a shale menber of the Franciscan Coastal Belt sequence which includes graywacke, shale arn conglomerate facies. The shale contains veins arn rafted boulders of chert. At this site, the chert is a dense, uniform material that knaps well. Two major sources of this high quality chert, with an accompanying concentration of tools and flakes, were found at this site. The largest occurs 15 m north of the old homestead in an amphitheatre-shaped depression on the north bank of the intermittent stream. It covers an area of approximately 20 x 30 m. The chert here is predominantly red brown with occasional bands of green and light yellow. The second source occurs on the south bank and approximately 50 m upstream fran the larger source. The chert here is a light green to gray green chert that covers an area of 15 x 15 m. On either side of these two sources, there is a sparse to moderate surface scatter of chert waste flakes covering the area of the site. The soil of the site, as well as the surrounding soil, is a dark gray brown, clayey silt to silty clay. There is no midden on this site. Four auger holes were made, but no buried midden was found. An auger hole was put near the chert source on the south bank and although the soil was a canpact, brown clayey silt and was not midden-like, it contained a flake and a core fran 0-45 an. The area of Son-554 is in open grassland with a few scattered oaks. There is a ribbon of riparian vegetation along the intermittent stream. The riparian trees are bay, buckeye, madrone, and live oak. The lower slopes east of the site are covered with dense forests of mairone/oak/bay woodland. The higher slopes on the other three sides of the site are in less dense oak woodland. The stream which flows along the length of the quarry would least 3 months of the year. There are two springs, however, on bank of the stream that probably maintain a year-round flow. be dry at the south Son-554 seems to have suffered some recent distrubance. As mentioned, a homestead was located in the general area. A fence line passed through the larger of the chert concentrations. The open, fairly flat field to the north of the site is known to have been plowed. There may have been a trail that went fran the homestead across the stream to the plowed field, and this trail goes directly over the larger of the chert concentrations, and may have partially caused the amphitheater-shaped depression here. The disturbance on the rest of the site is minimal. Erosion by the intermittent stream is fairly active and the stream seens to be down-cutting rapidly. A 1 x 1 m test concentrations. It was exposed concentration of datum stake. unit was located on the larger of the chert located on a gently sloping area just above the chert tools and flakes at 4 m N 700 E of the site 149 The unit was dry-screened fran 0-40 an. The increasing clay ard moisture content of this soil after 40 an made washing imperative. A pump was brought in ard the soil was washed for the rest of the unit. The qanpactness arid high clay content of the soil made excavation arnd washing slow ard tedious. The unit ended at 90 an in a nearly sterile gravelly clay. An auger hole in the center of the unit went through 60 an of non-artifact bearing gravelly clay to shale bedrock at 151 an. The two chert sources found at Son-554 are accompanied by an extraordinarily large number of chert artifacts. These artifacts include many cores, some retouched and utilized flakes, scrapers, ard a few bifaces. When wet-screening began, a surface collection of the area of Concentration chert debris between Unit 1 and the intermittent strean was made because of the disturbance wet-screening would cause. The area of collection measured 10 x 10 m ard was biased in that only tools were collected. Surface finds included a medial fragment of a chert projectile point, 6 bifaces, 2 scrapers, 7 cores, and 4 retouched flakes, as well as a fragment of a basalt pestle ard a sandstone hammerstone. Excavation of Unit 1 demonstrated that this quarry site had much more depth than was expected. Unfortunately, since this was a quarry and all tools were rudimentary, no diagnostic artifacts were found ard no date can be assigned to this site. Stratigraphically, the unit can be roughly divided into two strata; one, fran 0-40 cm, consisting of dark gray brown compact silt to clayey silt with some gravel. The second stratum, fran 40-90 gn, had a higher clay and gravel content. Excavations uncovered many more cores than were found at any other site, as well as 2 scrapers, 4 bifaces, ard a hammerstone. The debitage count was consistently high fran 0-60 an, after which it rapidly diminished. There were several pieces of redwood fencing found in the 0-10 an level which is not surprising since a redwood fence once stood in this area. Also, an hexagonal shell casing (possibly fran a muzzle loaded gun) was found from 10-20 an. This historic artifact may have been buried by rodent disturbance or could indicate rapid colluvial deposition. The only signs of aboriginal living debris, as opposed to work debris, on or around the site occurs at the surface scatter on the SW edge on Son- 554. Here there were a few fire-cracked rocks and 3 retouched chert flakes, but no midden. This quarry site was fully exploited by the local aboriginies. The amphitheater-like depression was probably caused by digging out the high quality chert which exists in veins fairly close to the surface. The chert was then knapped on the spot. Subsequent erosion fran the stream below and colluvial deposition from the field above reburied some of the artifacts and exposed others. Son-554 will not be flooded but should be excavated because of its 150 great archaeological significance. It is the only quarry site known in the southern area of the area of the project, which ethnographically was the territory of the Dry Creek Pono. The northern area, in different tribelet territory, has a quarry at Sor-582. Although some rare natural resources were shared by different bands, and even different tribes, chert quarries were probably camon enough to be considered the sole property of the local bard. There exists a unique opportunity to canpare the quarrying and flaking techniques of one bank with another. It may also be possible to do source analyses of the chert frai the two quarries and canpare- that with chert artifacts fran various sites. Lower Warm Springs Chronology While no charcoal for radiocarbon determination was recovered in this group a large number of index artifacts is available to construct a rich chronological picture. Son-594 had no midden left to excavate. Barrett's (1908) statement sheds some light on the subject. He says there were a large number of mortars and pestles so this together with its identification as a naned village indicates that the Houx Aspect right up to the historic period. He also says there are other large stone implements which I think can be taken to be, in part at least, milling stones and hand stones. Thus the Borax Lake Aspect was also present there. Son-558 had a projectile point that as noted earlier is a late Houx type. It was such a skimpy deposit that this period may be all that was represented. Site Son-556 has sufficient index artifacts to make its stratigraphic interpretation rather canplex. The persons excavating the site note a midden change at 50 an depth, a change fram dark gray brownto very dark gray brown and fran moderately loose to less loose. However this was not sufficient to indicate in the profile but only a canpact sandy dense at this depth. On the other hand there is a slight but definite decrease in relative anount of chert (relative to obsidian) at the 70 an level and we feel this is a more reasonable place to indicate a break. Three index artifacts cane fran above 70 cm: a small side notched point fran the surface that could easily be late (Fig. 27a); a crude Gunther barbed point fran 60-70 an (Fig. 27b); and a fragment from 40-50 an, that appears to be the barb of a Gunther point (not illustrated). These identify the top canponent as late Houx. The next midden break occurs at 110 an depth and this coincides with a break in the chert-obsidian ratio at 120 an. We assign the levels fran 70 to 110 an to the early Houx Aspect. Within this the levels fran 70 to 90 an have a mortar fragment occurring with two milling stone fragments and a handstone fragment. This would date those levels right around the time of Christ and would imply a considerable time difference between those levels and the ones immediately above, a difference for which there is no physical evidence. We leave this ananily to be dealt with through future excavation. 151 Next below this is a component fran 110 to 170 an where another miAdden break coincided with a break in chippage ratios. There are three quite definite large Excelsior points in this component (Figs. 27c, d, e) so it is assigned to Late Borax Lake Aspect. The renainder we assign to early Borax Lake with the possibility that something yet earlier is at the bottam of the site unassayed. A projectile point found at a depth of 190-200 cm seens to have been shaped with definite intent arnd is almost identical to one recovered fran the Willits site arnd we therefore feel it must represent a definite and presumably early type as yet undefined (compare Fig. 27f with Meighau 1955:Pl. 4H). Site Son-559 has no midden to speak of but three diagnostic projectile points were found on the surface. The smallest Excelsior point shown in Fig. 27g would seen to date from early Houx while the other two (Fig. 27h arn 27i) are from Borax Lake. We assign the site to these two periods but make no differentiation with respect to stratigraphic levels. We are assigning site Son-553 entirely to the Houx Aspect ard the historic period. The top 20 an we consider historic because of the glass chippage. Below that we consider it late Houx down to about 50 an where there is soil and debitage ratio. We consider it early Houx below that. 152 Chronology of Lower Warm Springs Site Son- 594 558 556 559 553 Historic 0-20 Hb X 0-70 20-50 H X Ha 70-110 X 50-100 BLb 110-170 BL X X BLa 170-220 153 Lower Warm Springs Group Chert-Obsidian Ratios by Site and Period 556 Site Son- 559 0.40 (217) 0.12 (210) 0.21 (445) x 553 1.03 (126) 4.17 (150) 6.38 (155) x 0.96 (151) 0.30 (1023) 0.29 (144) 2.85 (431) Late Houx Houx Early Houx Late Borax Lake Borax Lake Early Borax Lake Historic Hb H Ha Bib BL BLa Total Hb: H: Ha: BLB: BL: BLa: - - - Lower Warm Springs Site Group Chert-Obsidian Debitage Ratios by Site and Level Site CA-Son 556 0.50( 18) 0.73( 45) 0.25( 25) 0.60( 16) 0.38( 29) 0.24( 41) 0.30( 43) 0.13( 34) 0.16( 43) 0.07( 59) 0.14( 74) 0.16(104) 0.41( 84) 0.27( 80) 0.00( 40) 0.29( 71) 0.22( 66) 0.78( 48) 0.78( 32) 0.82( 31) 1.80( 28) 1.40( 11) 559 0.71(36) 0.17(49) 0.19(43) 0.23(16) 555 11.50(25) 4.33(32) 13.33(43) 8.00(18) * ( 2) ( ) : debitage frequency * : infinity, or all chert 154 Level CM. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 130-140 140-150 150-160 160-170 170-180 180-190 190-200 200-210 210-220 553 0.96(88) 1.24(38) 3.44(40) 4.50(55) 2.50(35) 10.00(44) 8.00(63) 9.00(30) 2.20(16) 0.00( 2) 155 0 w z Cf) w )~~~~~~~~~C CDI C bib~. 0w Z;~~~~~~~~~ LE O~~~~~~~ v 0 S K[2lE 13 332? Na Map 26. Upper Warm Spring Group. 157 Upper Warm Spring Creek Group The sites of this group are found on Warm Springs Creek at the mouth of Rancheria Creek and above and in the drainage of Rancheria Creek the topography here is dominated by the narrow canyons of the two streams so we would not expect large sites and indeed we do not find sites that are really extensive but several are deep and chronological significant. The land here is dominated by an area between the two streans on which the vegetation reconstructs as mixed evergreen and oak forest which we believe had considerable stands of redwood. This makes it a fringe area fran the standpoint of Mahikaune or Dry Creek Pamo habitat and most probably makes it a specialized area from the standpoint of the people of the tribelet. It also straddles the trail between Dry Creek and the Pacific Coast at Stewart's point, an important trail ethnographically and probably earlier as well in the exchane and transport of goods between the coast and interior. The area is shown on Map 26 but the group does not include sites Son-553 and 554 on the eastern part of the map. Site Son-543 (Days End Blind) Thts is an isolated hunting blind site at the extrene western end of the Upper Warm Springs Group. The blind is 300 m downstrean from the confluence of Warm Springs Creek and Strawberry Creek and approximately eight meters north of the Skaggs Springs-Stewart's Point road. It is situated on the north side of Warm Springs Creek, at the beginning of the hill sloping out of the valley of Warm Springs Creek. The blind is in a ten by five meter, fairly flat, shallow basin along an epheneral drainage. It is bordered to the north, northeast and west by the hillshope of Warm Springs Valley. To the south and southeast the blind has a good view of the lower terraces of Warm Springs Creek. The vegetation of the area consists of open oak woodland with very thick grass groundcover. Oregon oak, coast live oak and black oak dominate, with sane madrone and Douglas fir. There is no appreciable modern man-made disturbance within the immediate area of Site 543. There was, however, considerable rodent disturbance on the bottom of the hunting blind as well as slumping along the sides of it. It is known ethnographically that the Pano used blinds when hunting deer and birds (Barrett 1952:127, 133). Five hunting blinds were found during the field survey. All of these blinds are found in the south, or Dry Creek Pomo part of the Project Area. It is interesting that these seen to be confined to one of the two area tribelets. Since very little is known about hunting blinds, the clearing of two blinds was initiated. Site 543 was cleaned and studied because of its easy accessibility fran the country road. Before cleaning, the depression of the hunting blind measured 139 an by 158 129 an and 36 an deep. There were eight large rocks, two of which formed courses along the northern edge of the perimeter. Several other large rocks were scattered in the immediate area and had apparently fallen fram the northern wall of the blind. These rocks may not have formed a continuous wall but may have been used to anchor brush screening. The blind was first cleared of grass groundcover and then scraped with a trowel to a hard undisturbed surface. All rocks both within and outside the actual pit were left in situ but were exposed and drawn to scale. This method of clearing the bliiO exposed a number of rocks that had scattered downslope fran the original wall perimeter. It, also, expanded andd deepened the blind depression to, measure 3.5 meters by 2.8 meters and to a depth of 47 an. The cleared ground and loose earth were screened ,with a 1/4" mesh but no artifacts were recovered. No discernable animal trails were observed in the immediate area of Site 543, but the ephemeral drainage to the east affords easy access to Warm Springs Creek. Site Son-544 (Serene Flat) This is a rich, deep midden site in the extreme western end of the Upper Warm Springs Group. The site is found between Skaggs Spring-Ste'wart's Point country road and Warm Springs Creek about a half mile west of the confluence of Rancheria and Warm Springs Creeks. The site is situated on a secondary terrace about.15 meters above and 30 meters northwest of Warm Springs Creek. This flat open terrace is bordered on the southeast by the creek, on the northwest by the canyon slope, on the northeast by a year-round spring-fed tributary, and on the southwest by.-an ephemeral drainage. It is in an area of open oak woodland vegetation dominated by Oregon, Coastal and Black Oaks with some firs present on the higher -slopes. Along both drainages bordering the site is a well-developed riparian vegetation with bays, willows, and oaks as well as thick, dense areas of woodwardia. Several species of Brodeia covered the site at the time of excavation. The soil of Son-544 is a very dark gray gravelly silt. This contrasts sharply with the pale brown gravelly silt in the surrounding area which contains many angular, talus-derived pebbles. Three auger holes were placed to determine the extent of midden, all on the southwest side of the site. This is the only open side which is not bordered by tributaries or hillslope. Augering revealed that the midden continued onto the open, flat terrace up to 75 meters southwest of the spring-fed tributary, but did not extend over the entire terrace. Site 544 is known to be an historic homestead site, and was called "Serene Flat" by the original homesteaders. A small wooden house with a brick chimney once stood about 20 meters northwest of the Test Unit 1. Its foundation has been obscured by recent plowing and bricks are scattered in the immediate area. In addition to plowing, the site has been disturbed further by bottle-hunters who have dug several small shallow pits in search of 19th century relics. It is probable that at least the first 20 to 30 an of this _ -ON 0 10 20 Meters Map 27. Son-544 (legend p. 5). 159 160 ~0 -20- -30 - 140 - -50 -' - 60 -0 -100 f cm A. Midden; dark brown to very dark grayish brown; gravelly silt with cobbles; dry; friable. Annual grass root and plow zone. B. Midlden; very dark grayish brown; gravelly silt with cobbles and fire-cracked rock; dry friable. C. Transitional zone; very dark grayish brown to brown; silt with sand and clay; subangular gravels; ccompact; slightly moist. D. Sub-midden; brown to dark yellowish brown; silty sand with gravels; moderately friable to canpact; slightly moist. E. Kortovina disturbance. Figure 19. Son-544 Profile. 161 Site Son-544 (Serene Flat) Scraper Retouch Used Flake Flake 2-c 3-c 1-o 4-c 1-o Cores Debitage 89-c 69-o 1-c 1-c 68-c 69-o 75-c 57-o 73-c 59-0 67-c 61-o 3-c 1-c 99-c lsro 46-o 5-c 1-c 169-c 1-o 49-o 5c 1-o l-o 2-c 1-c 3-c 162-c 47-o 70-c 21-o 73-c 18-o 49-c 11-o 29-c 5-o 39-c 7-o 1-o Totals 4-c 10-c 12-c 21-c 6-c 1062-c 3-o 5-o l-o 7-o 4-o 1-o 520-o 25 1 1582 Level an. Proj. Pts. Reorked Biface 0-10 1-o 2-o l-o 1-o 10-20 20-30 30-40 40-50 50-60 60-70 70-80 1-c 2-o 1-c 1-o 1-c 1-c 4-c 1-c 80-90 1-c 90-100 1-o 1-c 100-110 1-o 1-c 1-c 2-o 4-c 1-o 110-120 1-c 1-o 120-130 130-140 1-c Grarn Total 7 5 11 19 I IlTo 162 Site Son-544 (Serene Flat) Level Pestle Shell Fired Animal Historical an. Bead Fock Bone 0-10 24 100 1 sq. nail 10-20 1 30 46 1 sq. nail 20-30 1 62 124 1 glass frag. 30-40 65 45 40-50 83 19 50-60 76 9 60-70 118 10 70-80 49 8 80-90 18 90-100 24 3 100-110 5 3 130-140 2 Totals 1 1 554 369 163 site has bene seriously disturbed by either plowing, historic foundations or pot-hunter holes. Test Unit 1 was placed at 25 meters north, 450 east of datum in the darkest part of the midden. The unit continued to 140 an with the last 50 an being riddled by krotovina disturbance. Excavations revealed that this was an important, rich midden site. It not only contained a high quantity of lithic debris, but also one of the highest amounts of animal bone found at any site in the Project Area. Moreover, the only shell bead found during excavations came fran this site. Physically site Son-544 can be divided into four strata, the top 20 an being differentiated fran the next 40 an only because it is a plow zone. These two are evidently equivalent and because of various index artifacts, particularly a clam shell disc bead, we attribute the 0-60 an levels to the latest time period. A lower midden component exteids from 60-90 an aid includes the remaining very dark gray gravelly silt and a mottled transitional zone. The amount of animal bone rapidly drops off in this canponent, but conversely, the fire- cracked rock count triples in quantity. This suggests that the very large quantity of faunal remains in the upper midden zone can be attributed to fairly recent deposition rather than to preservation by burning. The proposition of chert to obsidian debitage increases in these lower levels. tbe diagnostic artifacts fran this component are a small side-notched projectile point, a small Excelsior point aid a milling stone. The final, sub-midden canponent (90-140 an) includes a dark yellow brown silty clayey sand with heavy rodent disturbance. It is characterized by low quantities of both animal bone aid fire-cracked rock, and a decrease in lithic remains. The ratio of chert to obsidian, however, steadily increases, reaching 5:1 in this component. Although no diagnostic artifacts were recovered fran this final zone, the chert to obsidian debitage ratios suggests that this sub-midden level is an earlier component than any represented above. Site Son-545 (Horseshoe Flat) This is a weakly developed midden site which is one of five habitation sites located within the Rancheria Creek drainage. The site is 3/4 mile upstream fran the confluence of Rancheria aid Warm Springs Creeks, aid is the furthest upstream of the five Rancheria sites. It is about 40 meters north of aid directly across Rancheria Creek fran the junction of Rancheria Creek Road and Wickersham Road. Site 545 is situated on a first terrace on the north bank of Rancheria Creek, delineated by the confluence of Rancheria Creek aid an unnamed intermittent tributary. The promontory consists of an outcrop of resistant graywacke aid is surrounded by large boulders dam deep pools of water in the creek. The site has open grassland vegetation with a scattering of oak stands. The slopes on the northeast side of Rancheria Creek are now open oak/madrone woodland, and the opposite bank is in dense mixed oak/evergreen 164 forest with redwood, oak and laurel dominant. The soil of the site is a dark brown sandy silt which contrasts with the surrounding lighter soils. Three auger holes were placed both east and west of the seasonal drainage, which is on the upstrean side of the terrace. These revealed no buried midden deposits so the midden at this site is evidently confined to the ten meter square terrace. Test Unit 1 was placed on the first terrace in what appeared to be the darkest part of the midden, south 800 west, 4.3 meters fran datum. It was quickly discovered that the midden was very weakly developed aid continued only to 30 an. The soil became saidier and full of cobbles aid pebbles as the unit continued in depth. Only small pockets of midden remained in the 20-30 an level, and the 30-40 an level was sterile. The single cultural component at Site 545 contained only a few artifacts. Scraper Retouch Used Flakes Flakes Cores Debitage 0-10 1-c 1-c 2-c 1-c 17-c 2-o 3-o 10-20 4-c 1-o 1-o 1-o 20-30 1-c 4-c 30-40 One diagnostic possibly indicating artifact was a chert straight based a Houx canponent. projectile point Site Son-546 (Farmer's Path) This is a surface site situated on a primary terrace on the north bank of Rancheria Creek. It is about a half mile upstream fran the confluence of Rancheria and Warm Springs Creeks. The site is almost certainly an extension of the well-developed midden site just 90 meters downstream, Son-547. The area of the site encompasses a flat, open terrace measuring 40 by 20 meters. An old corral and two abandoned farm buildings are located north aid west of the midden area. The area is on a footpath which was previously used as a road and presently connects Sites 546 and 547. The site is bordered on the south by Rancheria Creek, and on the north, west and east by the hillslope and the farm buildings. Site 546 Oregon Oak is north of the is in open oak woodland with very thick grass groundcover. the dominant tree both on the terrace aid on the slopes to the site. Ma~rone aid coast live oak staids are present on the Level CM. Proj. Pt. 165 terrace. The lower an] opposite slope of Rancheria Creek is now a boggy, open grassland with a dense mixed redwood forest above it. Numerous specimens of several Brodeia species, Mariposa lily, and soaproot were seen on all open areas of both slopes during the spring months. Along Rancheria Creek, the usual riparian vegetation of bay, willow, and cottonwood are present. The soil of the surface site and the surrounding area is a dark yellow brown alluvial sandy silt. Two auger holes were bored in this area to determine if any midden might be buried. In neither case was midden discovered, the alluvial sandy silt continuing the length of the auger (1.5 meters). There was a very sparse scatter of chert and obsidian flakes concentrated in a three by five meter area between the corral and Rancheria Creek. Farming activities have obscured, if not destroyed, traces of aboriginal utilization. No diagnostic artifacts were recovered during surface survey reconnaisance. However, the presence of the collapsed, abanxdoned farmhouse, outbuilding and corral suggest a date in the homesteading period, e.g., 1850- 1900. This site undoubtedly represents a late use of the Rancheria drainage ani] could be instructive in understanding interrelationships between Europeans and Pono Indians during that time period. Site Son-547 (Broken Bridge) This is an important well-developed midden site situated on the north bank of Rancheria Creek. It is about a half mile upstream fram the confluence of Rancheria and Warm Springs Creeks, aix] is the deepest of five sites located within this drainage. The site is on a triangularly shaped primary terrace which is at the confluence of Rancheria Creek anx its first major tributary. The terrace is on the upstream side of the unnamed tributary ax] has its apex pointing southeast. The site is bordered to the east by this year-round tributary, to the south and west by the creek itself, and to the north and west (at the base of the triangle) by the hillside. The triangle measures approximately 30 meters to a side. Fifteen meters to the east of datum and across the tributary is an extension of the primary terrace on which this site is located. It measures 40 by 10 meters and is a flat shaded area. This extension contained a sparse flake scatter, but no midden. The vegetation in and around the site is open oak woodland with oak and bay dominating and allowing moderate grass groundcover. On the slopes to the north, Oregon oak continues to dominate, with some madrone an] coast live oak. On the opposite side of Rancheria, the oak forest slowly gives way to a dense mixed redwood cover. Numerous specimens of brodeia, wild onion and soaproot were found growing on or around the site. A well-developed riparian system is absent along the drainages and we find oak woodland growing up to the edges of the stream banks. The soil of the site is dark gray brown gravelly sandy silt with many 166 CALSon 547 MAP N 0 '/ A AX 0 15 30 Meters Map 28. Son 547 (legend p. 5). . 0 * 167 - 20 -40 50 - 60 ~ 4 ~ ~ 4 a 70 d fb Annua grass 0 V 414 & 7g A ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~~~~~*4?44~~~~~~~~~~~~~~ A Vs rooAt. 1 9 v v P A z4o A VI 4 \\ V~ '9''q 4C P- 110 A ? (4 1E'C & A9~' . A4P. B. MiddAen; very~'A~t% dark:Q grais bro& bitylom A. Midden; vr dark grys brown;sad silty withsub anua rvl;dry; fribe Annual grs D. Troostizone. zn;vrydr ryihbont B. Midn;veydark grayoish brown; srd silty lotgam;ls dry;tl friable Heiavye roto disturbane.ycn C. SuMidden; very bow t dark grayoish brown;sitlom dryelwsbrw;sandy silt with gravels; lgtymit opc. F. Gravels; yellowish brown to dark yellowish brown; sandy clay; moist; mo~derately canpact. G. Krotovina disturbance. Figure 20. Son-547 Profile. 168 Site Son-547 (Broken Bridge) Level Proj. cm. Pts. Biface Reworked Biface Scraper Retouch Used Cores Flake Flakes 0-10 1-c 10-20 1-c 2-c 2-c 1-o 1-c 20-30 2-c 1-o 2-c 8-o 1-o 5-c 1-c 9-o 1-c 2-o 1-o 30-40 1-c 1-o 40-50 1-o 5-c 4-o 7-c 5-o 1-o 50-60 1-c 8-c 9-0 1-o 60-70 2-c 4-o 70-80 2-o 42-c 16-o 85-c 41-o 44-c 21-o 75-c 23-o 1-c 107-c 18-o 91-c 27-o 35-c 12-o 25-c 10-o 80-90 18-c 5-o 90-100 3-c 7-o 100-110 8-c 15-o 1-o 110-120 1-o 120-130 1-o 7-o 2-o Totals 6-c 1-c 4-c 29-c 1-c 533-c 3-o 3-o l-o 3-o 46-o 204-o 9 1 3 5 3 75 1 737 Debitage Grand Totals 169 Site Son-547 (Broken Bridge) Level Hand Pestle Fired Animal CM. Stone Rock Bone 0-10 7 1 10-20 1 1 45 11 20-30 62 21 30-40 52 12 40-50 32 4 50-60 15 7 60-70 19 5 70-80 4 3 80-90 11 6 90-100 6 1 100-110 1 1 110-120 1 120-130 3 Total 1 1 258 72 170 sub-angular cobbles and gravels. The surrounding soil is much lighter, being a yellow brown silt. This entire section of Rancheria Creek is underlain with massive beds of sedimentary rock of the Franciscan sequence. A resistant graywacke outcrop is 20 meters north of the site, up the tributary. This outcrop can be seen in a series of snall, shallow bedrock pools. Four auger holes were placed in the vicinity of the site. The first was placed to the east of the site in the flake scatter, and found no midden. Two holes were bored to the west of Test Unit 1 but were uninformative because they bottomed out on rocks. The fourth auger hole was made 20 meters north of the unit just at the beginning of the hillslope. This hole contained 30 an of silt slopewash, arn then 40 an of midden. This boring showed that the midden extends over the entire terrace on the northwest side of the confluence. An examination of the stream cuts bordering the site showed that the deepest part of the site was in the southwest section. The midden there apeared to be about a meter deep. Unfortunately, Site 547 has had much disturbance. The site is adjacent to an old fallen, wooden bridge. At one time a road passed over this bridge and continued through the site up to the buildings and corral of Son-546, 90 meters upstream. As much as 20% of the site has been disturbed by this road. A salt lick was located on the site stimulating heavy sheep grazing, further disturbing the surface. Being situated at a confluence, this site is subject to heavy stream erosion, especially on the tributary side of the promontory. Finally, excavation has shown there to be heavy root disturbance in the upper 40 an. The nearby bay trees formed a thick matting of roots and rootlets (much more than at any other site) making excavation slow and difficult and stratigraphy uncertain. Test Unit 1 was placed on the south edge of the site, (south 220 east 10 m from datum), just two meters fran the cutbank of Rancheria Creek. The unit was placed so as to avoid the disturbance caused by the road and yet to excavate in the darkest, most central part of the midden. The unit revealed that this site contained a rich midden deposit which continued to 130 cm, ending in sterile stream gravels. An auger hole in the center of the last level continued to 185 an but encountered no further cultural constituents. Four cultural components are present at Son-547 which roughly correlate with the stratigraphic levels shown in the profile. The midden component seems to be separated into two distinct periods. The upper midden levels fall between 0-30 an. Unfortunately, heavy root disturbance has obscured the stratigraphy and made correlation with artifacts difficult. Large amounts of animal bone, fire-cracked rock anx] lithic debitage were recovered fran this zone. The debitage is characterized by a 2:1 chert to obsidian ratio. Diagnostic artifacts consist of two medium size aix one small cornernotched projectile points. One of the larger pieces was found sub- surface in the auger; the other two pieces were recovered in the excavations. The lower midden component basically continues to 60 an and consists of dark gray brown sandy silt. This canponent has fewer tools than the upper midden zone, less animal bone ard smaller amounts of fire-cracked rock. The chert to obsidian debitage ratio has increased to over 4:1. 171 Tools of a diagnostic form include one chert concave based piece, one small obsidian Excelsior point arn one possible Desert Side-notched point and in any case very late. Unfortunately, this zone has ha] very heavy root disturbance. This must account for the side-notched piece which is very late in nature if our interpretation is correct. The next thirty centimeters encompasses a transitional stratum, characterized by dark yellow brown sarndy silt. It lacks any diagnostic artifacts and contains no tools. The chert to obsidian ratio is roughly the same as that of the lower midden component and we include it in the same mixed canponent. Both animal bone and fire-cracked rock continue to decrease with depth. The sub-midden zone (90-130 an) consists of dark yellow brown sandy clay. Few artifacts and nothing diagnostic were recovered fran this stratum. The most interesting fact about this component is that the obsidian debitage consistently outnumbers the chert for these last four levels. This reversal is a unique occurrence among the Warm Springs sites. Site 547 is the richest, deepest site in the Rancheria Creek drainage, and as such was probably the focal site for the other four that surround it. It may have been the primary habitation site in the area with the other used as seasonal camps for fishing, acorns and food collecting. In the late period, this small terrace would not accanodate all of its inhabitants and it is likely that, at least by the protohistoric period, they spread out to include the larger area of Son-546. Site Son-550 (Fallen Oak) Son-550 is a shallow midden site with three possible housepits--it is on the bank of the first major tributary of Rancheria Creek and is 280 m up from its confluence with Rancheria Creek. The site is situated on a flat shoulder of the surrounding hills that shelter it from the north, west, and south. The flat covers an area of 50 by 30 meters and the midden covers almost the entire area, measuring 45 m E-W by 28 m N-S. The midden is bordered by gullys north and south, by a hillside on the west, and on the east by the stream forming the major drainage for this small valley. The site is in open oak woodlaix] with moderate grass groundcover. The dominant trees are Oregon oak anxd coast live oak with some madrone aid black oak. On higher north-facing slopes is found a denser, mixed forest of Douglas fir and oak. At the time of excavation (early June) numerous Brodeia aid soaproot were in bloom. The nearest water is the small stream to the east. This flows fran the beginning of Novemiber to the ern of June. In the summer it contains only isolated pools and a trickle of water. The soil of the site is dark brown gravelly silt with few rocks and pebbles. The surrounding soil was a light brown sandy soil. Two auger holes were placed on the east and west edges of the flat aid in each case 15 to 20 an of midden was recorded, indicating that the midden covers nearly the entire flat. There is no man-made disturbance in or around Son-550. However 172 Site Son-550 (Fallen Oak) Proj. Retouch Pts. Flakes Cores Debitage Shell Animal Bone 0-10 1-c 2-c 1-c 41-c 1 18 1-o 10-o 10-20 1-c 1-c 34-c 2 4 1-o 1-o 5-o 20-bottom 1-c 1-c 2-c 16-c 2 3-o Totals 3-c 4-c 3-c 91-c 5 22 1-o 2-o 18-o Grard Totals 4 6 3 109 5 22 Level an. 173 excavation showed some root and rodent disturbance within the midden. Erosion on the site is minimal. A meter square test unit was located at S 450 E, 6 m from the site datum stake. The unit was placed here because it appeared to be the deepest and most central part of the site. No wet screening was done. The unit was excavated to bedrock which was fran 20 to 40 an (average 28 an) below surface. The unit produced 4 projectile points all indicating that this is a single component site to be attributed to late Houx Aspect. There are three shallow, circular depressions in the site which appear to be housepits. They have diameters of 3, 2.2, and 2.1 meters respectively. Thus we have hamlet of perhaps 20 people occupied only during the winter (because of water supply) anx only during the latest aboriginal period. This is an unusual degree of specificity anx] merits further investigation of its association with tools and other midden components. Site Son-549 (Sunshine Flat) Son-549 is a midden site on the east bank of Rancheria Creek upstream from its confluence with Warm Springs Creek. The site is found on the first inhabitable, flat, open area up Rancheria Creek from Warm Springs creek. The flat is 10 m east aix 2 m above Rancheria Creek anx is about 50 by 60 m in area. It is sheltered on all sides by steep hills. The midden covers an area of 55 m E-W by 40 m N-S. It is limited on the west by the cutbank of Rancheria Creek, on the east anid north by a small drainage that is dry at least half the year, and to the south by a hill. It is in open grasslanxd with a few scattered Oregon oaks. The hills in back or east of the site are in open oak woodland with Oregon oak, coast live oak, aix madrone. Across Rancheria Creek to the west is a dense forest of redwood, bay laurel, aid oak. Water can be found year round in Rancheria Creek, although better drinking water is available at the spring across Rancheria Creek some 70 m west of the site. The soil of the site is a gray brown sandy silt with few rocks aid pebbles. The surrounding soil is lighter in color, being a light brown sandy soil. There is a large, high outcrop of rock 50 m east of the site over which an ephemeral stream occasionally flows. Six augerholes were made on the flat. There showed that a layer of at least 25 an of midden was present over most of the flat, but deepest and darkest on that part of the flat closest to Rancheria Creek. The only disturbance found at Son-549 is the extensive rodent disturbance of the midden. There is minor erosion taking place along the cutbank of Rancheria Creek. Surface finds discovered during the initial survey of this site gave indications that this might be a rich site. Unit 1, located in what surface features indicated was the center of the site, at S 300 E, 3 m from the datum stake, proved to be uninformative. Therefore, after augering had better defined the midden extent, a second excavation unit was located at S 250 E, 10 m from the site datum stake. 174 Site Son-549 (SSmshine Flat) Unit 1 Level Proj. Reworked an. Pts. Biface 0-10 10-20 Scraper Used Retouch Flakes Flakes 2-o 1-o 1-c 1-o 20-30 1-c 2-c 1-o 30-40 1-o 1-c 1-o 40-50 1-c 1-o 50-60 60-70 1-o 70-80 Totals 1-c 4-c 1-c 46-c l-o l-o 3-o 4-o 23-o Grand 1 1 1 7 5 69 Total Debitage lo-c 5-o 5-c 7-o 7-c 3-o 6-c 4-o 13-c 2-o 2-c 1-o 2-c 1-o 1-c Site Son-549 (Sunshine Flat) Unit 2 Proj. Reworked Scraper Used Retou Pts. Biface Flakes Flake Ich s 2-c 1-c 1-c 1-c 1-c 1-c 5-c 175 Debitae Level an. 0-10 10-20 20-30 30-40 40-50 50-60 60-70 70-80 80-90 90-100 Totals 19-c 7-o 21-c 11-o 19-c 13-o 9-c 7-o 30-c 7-o 22-c l0-o 23-c 4-o 24-c 1-o 12-c 1-o 1-c 180-c 61-o Grard 1 5 241 Totals 176 In the first three levels of Unit 1, wet screening was used. However, dry screening proved to be an effective and wet screening was dropped after the 30 an level. No wet screening was used in Unit 2. Unit 1 ended at 80 an in nearly-sterile, compact, yellow brown sand. Unit 2 ended at 110 an in the same sterile sand. Surface finds at Son-549 included two pestle, a mortar fragment, and a leaf shaped chert projectile point. Unfortunately, the point has been lost. Unit 1, as mentioned, turned out to be located a little off the center of the midden. Artifact count was less than 20 pieces per level, virtually no bone was present, and fire-cracked rock occurred only in minor amounts. Only 4 retouched flakes, 1 scraper, 1 reworked biface, 1 biface, and a medial fragment of a projectile point were found, and all in the first three levels. Chert artifacts outnumbered obsidian by at least 2:1 in most levels, except in the 10-40 an levels where obsidian nearly equaled chert. Unit 2, located more within the midden, did have a higher debitage count per level (averaging 30), but contained fewer tools. Only 6 retouched flakes were found. As in Unit 1, chert outnumbered obsidian by at least 2:1 except fran the 20-40 an levels where the ratio was more nearly equal. Fire-cracked rock was present in low quantities except fran 50-70 an, where a rock concentration occurred. This could not be called a feature however, because the rocks were randomly distributed, and neither charcoal, bone, nor artifacts were found in any concentration. This is perhaps the least productive site in the Project in spite of its considerable depth of midden. It would be of some interest to test this site further just because it is so unproductive. There may be unique activity patterns or other conditions which would be important to understand in terms of total lifeways of the Pano or their predecessors. Son-551 (Monday Morning Terrace) Son-551 is a midden site with an associated surface scatter on a secondary terrace on the NW bank of Warm Springs Creek 200 m below its confluence of Rancheria Creek. It is on a flat secondary terrace 60 m north of Warm Springs Creek and 6 m above it. The terrace faces SW and is protected on other sides by the surrounding hills. The midden occurs on the SE boundary of the terrace and on the west side of an ephemeral drainage. The midden covers an area of about 45 m E-W by 30 m N-S. It is bordered on the east by the ephemeral drainage, on the south by the slope of the secondary terrace, and imperceptibly grades into non-midden terrace deposits to the west. The soil of the site is a brown to dark brown alluvial gravelly silt. The midden is not very much darker than the surrounding soil which is a bro rocky gravelly sand. The midden's light color - it is lighter in color than that encountered at any other site - can possibly be explained by its having been leached by groundwater percolating through the porous alluvial deposits of the site. Nine augerholes were made in and around the site. Two augerholes were 177 V CAL Son * 551 MAP 0 0 5 10 Meters Bedrock Outcrop Map 29. Son-551 (legend p. 5). 7%1 L-J, I / bkl ./V 178 - 10- -20- 4 -30- 40 -50 60 01 O0000000000000000 O0000000000000000 000 oo oooooooooooo00ooooooooooooooo ooooooooooooooooooooooooooo )OO 0oo o000 0000000000000000 00000000 0000 000000000o00ooooo0ooo )OOO 000 000000 0000OOOOOOOOOOOOOOOOOOOOCOOC 0OOOOz 00 0 O D Coo0000 00000000(000 00000 0 OOOOOOOOOOOOOOOOOOOOOsDOOO OOOOOOOOC 0 0 0 000000 000000O 0000 OOooooooooooooo000000000oc 000000000000 00000000000000 00000000000000000000000000000 0000000000000o00 0 00 00000000000000o000000 000000000000 00000000o000 0000000000000 00 OO OOOOOOOOOOOOO 00000 00000000000000000000000 OOOoOooooo000 00OOOOOOOOOOOOO 0000000 0000000 OOOOOOOOOOOOOO0000 80 OOoOoooooooooooooooooooooooooo %o~OO, O~O~O~OOo~o~~O!o~o~o~o~o~~Oo~o~~OoCo 000 O00000000000000000000000000000000000(000 OOOOOOOOOOOO000000000000C ._ _ OOOOOtOOOOOOOOOOOOOOOOO OOOOO000000000000O 0OOOOO OOO0tDOOOOOOOOOOOOOOOOOO( 0000000000000000000Q0000000000000oooooooooo 0 0 000000000000000000000000, " 0000000000000o0o00000000000000o0ooo COOOOOOOooooooooooooooooooooooc 00000000000000000 0000000000000000 O 0000000000000000000000000C)QoO OOOOOOO0000000000OOO00000000000000000 OOOO0000000o0ooooo0ooo00o0 0c) -90- ooqoooooooooooooocooooooooooo0o00000000 oo oooooooooooooo0oo0ooooooooo 0000000000000000 00000000000000000OOOC-OOOOOooooooooooooooooooooooooooz( 0000000000 00000000000000000000000000000000000000000000000000000000 OOOOOOOOOugDOOOOOOOOOOOOOOOO 0000000000000000000000000000000tDoooooto 90 000000000000000000000000000000000000000000000000000000000000 000000000000000000000000000000000000 0000000000000000 000000 110 0O09: 0000 J( 00000()0000000 00000000000000000 )000000000(000 000000000000000 0000000000ooooooo 0000000000 00000000000)0000000000000000000000000000 00000000000 0000000000000000000000000*000000.000000000o ooooooooooooooooooooo000000000000000000000000000000000000000000000000000OO 00000000000000000000000000000000000000 00000000000000000000000000000000000 _4 00000000000000000000oooooocooooCoooooo~oooooo0000D 000 00000000000000000000000000000000000000300000 - -*. .----? 00 0 0 00 _ 00 0 0 0 00 0 0 00 0 nroooooooooooot~v>ODOg<*-}vw \ V@OAO 00000000N*???n^ vfvO~vOO~tvOO0000000000000 0 *ur :0 : - 110 00L&()tV{V (VOOVOC)O(O(Ot OOOC)'()O v - A - V .,..^}- --- 120~~~~~~C (I CO j OC 0 Poe~ , Oooooo~cOO00oof<) r i O -120- A. Midden; brown; gravelly silt; dry; canpact. Annual grass root zone. B. Midden; brown to dark brown; gravelly silt; slightly moist; friable to moderately canpact. C. Transitional zone; brown to yellowish brown; silt with sand ard gravels; moist; moderately canpact to canpact. D. Sub-midden; yellowish brown; sandy gravels; moist; canpact. Figure 21. Son-551 Profile. 179 Site Son-551 (Monday Morning terrace) Level Proj. Biface Reworked Scraper Retouch Used Debitage cm. Pts. Biface Flakes Flakes 0-10 2-c 1-c 5-c 5-c 195-c 1-o 1-o 12-o 10-20 1-c 6-c 124-c 2-o 1-o 10-o 20-30 4-c 168-c 11-o 30-40 1-c 1-c 1-c 120-c 1-o 2-o 12-o 40-50 2-c l-c 1-c 1-c 7-c 215-c 16-o 50-60 l-c 1-c 1-c 4-c 190-c 2-o 15-o 60-70 4-c 6-c 231-c 6-o 70-80 3-c 67-c 4-o 80-90 1-c 29-c 2-o 90-100 19-c 1-o 100-110 9-c 110-120 4-c Totals 4-c 5-c 3-c 11-c 37-c 1371-c 3-o 2-o 1-o 1-o 3-o 89-o Grand 7 5 2 4 12 40 1460 Totals I- 180 Site Son-551 (Monday Morning Terrace) Level Cores Pestle Hammer an. Stone 0-10 10-20 20-30 30-40 40-50 50-60 60-70 1-c 1 1 1-c 1-c 2-c Fired Animal Rock Bone 1 63 25 23 61 29 28 1 Total 5 1 1 230 1 181 done in the first terrace below Son-551. Both showed 70 to 110 an of brown alluvial silts and sands but no midden. The augerholes on the second terrace seemed to indicate that the midden covered the SE side of the terrace and extended about 45 m to the west. However, with the soil colors so similar, and with the augering having been done in June when the soil was dried out, it was difficult to be sure of the precise extent of the midden. Son-551 is in open grassland with a moderately thick grass groundcoxver. There is a scatter of small Oregon oak, coast live oak, ma3rone, and manzanita. COn the first terrace below the site are large valley and Oregon oaks. At the time of excavation (early June), numerous mariposa lilies, a food plant gathered by the Pono, were in bloan on and around the site. During most of the year, the nearest water is 60 m to the south at Warm Springs Creek. There are several large pools in the ervirons of the confluence of Rancheria and Warm Springs Creeks that could have been used for fishing. Son-551 has almost no disturbance and erosion here is minimal. A 1 x 1 m test unit was excavated at N 300 E 10 m fran the site datum stake. The unit was located here because surface features indicated it to be the richest part of the site. No wet screening was done. The unit ended at 120 an in a nearly sterile, yellow brown, compact sandy gravel. An augerhole was put in the floor of the finished unit. It continued to 182 an in the same yellow brown sandy gravel before bottoming out on rocks. No cultural material was found in augering. The midden may be divided into two physical layers. Fran 0-70 an it is a dark brown to dark gravelly silt while from 70-110 an it is a compact brown to yellow brown gravelly sandy silt. The cultural divisions do not coincide with this. On the basis of a small corner notched point found on the surface and a pestle fragment fran the 20-30 an level we consider the upper portion to belong to the Houx Aspect, presumably early Houx since there is nothing extremely late from any part of the site. There is a largish Excelsior point from 40-50 an level which may be attributable to late Borax Lake. Chert debitage shows a marked increase at this level as does fire cracked rock. Presumably the late Borax Lake component begins here and continues to the bottom of the site although there may be something earlier which we have not identified simply for lack of evidence. Chronology of Upper Warm Springs It is possible to interpret the chronology at site Son-544 in several ways. The clam disc bead in the 10-20 an levels makes it clear that late Houx is represented. There has also been some thought that the site was occupied during the historic period (some historic Indian site must have existed to account for the name Rancheria Creek). This possibility is rejected on the grounds that such sites always contain quantities of chipped glass and other historic objects which this does not. The first real physical break comes at 60 cm and coincides with a break in the chippage ratios. The question is 182 Upper Warm Springs Creek Group Chert-Obsidian Debitaje Ratios by Site arnd Period Period Hb H Ha BLb BL BLa Post Total Hb H Ha BLb BL BLa Post 544 547 0-20 1.14 (295) Site CA-Son 550 549 551 0-40 5.05 (109) 0-30 2.19 (249) 0-100 2.69 (310) 20-60 1.41 (537) 60-90 3.43 (518) 0-40 13.49 (652) 40-120 17.36 (808) 30-90 3.69 (446) 90-130 4.52 (232) 90-130 0.35 (42) 2.04 (1582) 2.61 (1737) 5.05 (109) 2.64 (310) 15.40 (1460) Late Houx Houx Early Houx Late Borax Lake Borax Lake Early Borax Lake Post Pattern 183 Upper Warm Springs Creek Site Group Chert-Obsidian Debitage Ratios by Site and Level Site CA-Son 547 2.63(58) 2.07(126) 2.10(65) 3. 26(98) 5.94(125) 3.37(118) 2.92(47) 2. 50(35) 3.60(23) 0.43(10) 0. 53(23) 0.00(7) 0.00(2) 549-1 3.00(15) 0.71(12) 2.32(10) 1.50(10) 6.50(15) 2.00(3) 2.00(3) * (1) 549-2 2.71(26) 1.91(32) 1.46(32) 1.28(16) 4.29(37) 2.00(32) 5.75(27) 24.00(25) 12.00(13) * (1) 551 16.25(207) 12.40(134) 15.27(179) 10.00(132) 13.44(231) 12.67(205) 38.50(237) 16.75(71) 14.50(31) 19.00(20) * (9) * (4) ( ) : debitae frequency * : infinity, or all chert Level an. 0-10 10-20 20-30 30-40 -40-50 50-60 60-70 70-80 80-90 90-100 100-110 110-120 120-130 544 1.29(158) 0.98(137) 1.32(132) 1.24(132) 1.10(128) 2.15(145) 3.45(218) 3.45(209) 3.33(91) 4.06(91) 4.45(60) 5.80(34) 5.57(46) 184 whether to assign the entire 0-60 an late Houx or to make an arbitrary break assigning part of it to early Houx; the projectile points shown in Figs. 28b and 28c at 40 to 50 and 50 to 60 an respectively could be in either. We are choosing the second option mainly because the Excelsior point sIhwn in Fig. 28c seems slightly large for a late Houx canponent. This gives us 0-20 an as late Houx aid 20-60 an as early Houx. The lower portions of the site have both physical ard chippage ratio breaks' at 90 an so we are attributing 60-90 an to late Borax Lake aix 90-130 an to early Borax Lake. Site Son-547 must be mixed to some extent because of the heavy root infestation. Because of physical and debitage changes we are assigning the top 30 an to the Houx Aspect and fran 30 to 90 an to the Borax Lake Aspect. This would have the large leaf shaped piece shown in Fig. 28h fran the 20-30 an level too high. The piece shown in Fig. 28k and identified as a possible Desert Side-notched fragment would be much too deep, but then if it is indeed a Desert Side-notched point it would be far out of range geographically as well. For the portion of the site fran 90-130 an we are going out on a limb aid propose it as belonging to the Post Pattern. It is our experience here aid elsewhere that in early portions of sites when a change of chipping material occurs it is always obsidian that disappears (except where obsidian is local as at Borax Lake). Here we find just the reverse and we associate this with a pattern earlier than the others observed here, thus the Post Pattern. We shall discuss the implications of this proposal in the concluding section. Site Son-550 we attribute entirely to late Houx for reasons given in the site description. Site Son-549 can be classified as belonging to the Houx Aspect on the basis of the pestles and mortar fragments found on the surface. The bottom of the site may belong to an earlier period but further excavation will be required to make the attribution definite. As discussed in the site description earlier we consider that for Son-551 the top 40 an is early Houx while the bottom of it is late Borax Lake. 185 SUMLRY AND DISCUSSION Chronology and Debitage Ratios The following table summarizes the archaeological canponents fran the Warm Springs project according to present evidence. This chronology is basically a summary, a sequence of projectile point types and grinding tools, which of itself does not say anything about more important cultural changes that have occurred in the area. W must take up the question of such changes, but before doing so we summarize the chert-obsidian ratios for chipping waste. Among some of the general characteristics of the summary table, it is noted that the overall ratio is 0.94 or a little more than 50% obsidian. This is based on a count of 18,764 flakes, in this count, 5245, or more than a quarter cane fran Son-571, and of these, 93% are obsidian. Fran these numbers we can only conclude site 571 is extremely specialized as an obsidian chipping station. Therefore we exclude it for a moment for both the western group and for the project as a whole, and we obtain the following modified totals: Hb Ha BLb BLa Total Western 2.63 1.13 3.06 61.71 2.36 Group Grand 1.12 1.45 2.15 6.46 1.80 Total This gives us an overall ratio of 1.80 or just under 2 to 1 chert to obsidian. Furthermore the overall changes fram the earliest to the latest period is generally consonant with individual site changes. Accepting these figures as reasonable we can measure them against some particular values given in the table. Looking at totals for groups it is notable that both Yorty Creek and Upper Warm Springs have high chert values, 2.98 and 3.85 respectively. This must be related to their proximity to the two chert quarries. The Yorty quarry (Son-584) is very close to all sites on Yorty Creek, and also to Son- 597, assigned to the Eastern group but at the mouth of Yorty Creek, which also has very high chert values 5.98. The Upper Warm Springs sites are in similar proximity to quarry site Son-554, which while nearby has been formally included in our Lower Warm Springs group. Fram the standpoint of high chert debitage ratios both Lower Warm Springs sites 553 (5.10) and the quarry site itself should be included in the Upper group. These figures thus indicate that people occupying the Yorty Creek and Upper Warm Springs Creek sites used the nearby quarries, which is not surprising. %tat is surprising however, is that the people occupying the other sites relied so heavily upon obsidian, although chert was readily available. In other words, are we to account for such differences simply by a mile or two of distance? It is probable that there was a very casual use of chert by those people near quarries but with more care exercised when the material had to be fetched even a short distance. 186 4i NSIII LA co C- 84 co ( N N (') U ) ec C 0 O OA~ tn Ln in LA LA ~ 44- CN4miJC 0 N%4N O i O LA 8 O co r- O n co r- co kin oS% 4-m4- 1 un u n I n LO unL IL IO LO LA LA L 54 Cl'.~~~~~~~~~~~qwC &4 N sJ4 M '0 q co ON wB m @ I N r- ON co X I u we1oln I oaton 1 IN ? n LA 44 L Ani n LA LA L A LALAn L LA n LA L |3) 0~~~~~~~~~~~~~~~~~~~~~~~~~ '3) 08 > s1 ".1 5i 0O 8! 1 la cn > 0) 0 4 0 r-r--%O Ch (0 0 0 L f% CD CA IV 4C) Sites Chert-Obsidian Ratios by Component Hb Ha BIb BLa Total 567 4.42 2.64 2.84 572 .48 .40 3.06 61.71 1.69 Western 571 .06 .21 .08 568 4.28 12.27 5.00 total 2.63 .16 .59 61.71 .39 593-1 .61 593-2 .14 .49 7.50 7.00 .24 Eastern 582 1.40 4.00 1.66 608 1.40 1.40 597 4.47 19.07 5.98 total .15 q49 1.93 11.83 1o16 576 4.09 1.03 1.65 577 16.00 16.00 Yorty Cr. 584 3.35 3.35 583 9.97 3.81 5.87 total 4.93 1.03 5.85 2.98 Lower 556 .40 .12 .21 .96 .30 Wn. Spr. 553 4.17 6.38 5.10 total .99 .75 .21 .96 .58 544 1.14 1.41 3.43 4.52 2.04 Upper 550 5.05 5.05 dn. Spr. 551 13.49 17.36 15.04 total 1.59 3.44 7.24 4.52 3.85 Grand Total 1.12 .40 1.58 6.46 0.94 Hb: late Houx; Ha: early Houx; Bib: late Borax Lake; BLa: early Borax Lake. 187 188 If we consider site totals, it will be seen that they can be placed in three categories: (1) high obsidian sites (ratio < 1.0); (2) high chert sites (ratio > 5.0); and (3) intermediate, or "normal" sites. According to this breakdown three sites specialize in obsidian technology, sites 571 (preeminent), 593 and 556. Only one component fram these has been attributed to early Borax Lake, the bottom of Son-556. That component is on the borderline of the ratio category, has only a small number of chips, and may be of doubtful attribution anyway. We can therefore conclude that obsidian specialization and its high frequency use began in late Borax Lake times, indicated also by the (unadjusted) grand totals which show a jump in obsidian use at this time with yet further increases in early Houx. The importance of these figures is two-fold. First, the long range trend toward the greater use of obsidian is a measure of the increasing importance of trade in later periods. This will came as no surprise to California archaeologists who have long seen trade as a defining characteristic for the later periods in Central California, (differentiating them fram earlier periods). In this case, though, the measure is rather more precise than we have seen in the past. Thus the grand totals indicate almost no trade in early Borax Lake, a sharp jump in each-of the two succeeding periods, followed by a decrease in late Houx. If we take this last trend at face value, it is an occasion for surprise. The late Houx period corresponds roughly to what has been called Phase II of the Central California Late Horizon. The phase marker for Phase II has always been the clam shell disc bead which was thought to have been a sort of currency, "the life-blood of trade." If this is so, then surely we should see more trade in this period, not less. Thus, the evidence from these Warm Springs Sites will be regarded with some skepticism. Nevertheless it is evidence, and we should consider the possibility that trade peaked at an earlier period and was actually declining just prior to the historic period. The second important aspect of high frequency obsidian counts relates to particular sites. It would appear that, when obsidian cane into heavy use in late Borax Lake times, site Son-571 immediately became specialized in that industry and was the only site to do so in Upper Dry Creek. It continued this dominance in the next period (early Houx), but in late Houx it was replaced by site Son-593, the tribelet center, as the obsidian center as well. This means that in late Borax Lake and early Houx the obsidian specialists were living at a small site detached fram the chief tribelet village. In late Houx the chief village became the center for obsidian manufacture. To see this we campare not just ratios but total numbers. Hb Ha Blb BLa Son-571 - 244-c 136- c 4210-o 655-o Son-593-2 114-c 86-c 45-c 21-c 929-o 171-o 6-o 3-o So far as present evidence is concerned, the switch is conclusive. If further excavation supports this conclusion, we have evidence of a change in social organization. For the two middle periods the situation would seem 189 analogous to certain African tribes in which iron workers were of a different caste than that of the group as a whole and might therefore keep some distance fran them. This distinction would then have been obliterated in late Houx when obsidian specialists were included in the village center. we might wonder whether in the earlier period the obsi pecialists were not also the obsidian traders going from group to gr up carrying their raw material which they chipped locally (to order?) at a place reserved for them. This intriguing possibility seems worth considerable investigation. If we turn now to the other debitage ratio categories, as noted previously, the high chert sites, with one exception, are all located near one or the other of the two chert quarries. The exception is Son-568, not near any known quarry and surrounded by sites with medium or high obsidian use. This is all the more remarkable because its two Houx components date to a period of increased obsidian use. It is almost as if obsidian was shunned. The remaining sites are in no way remarkable with regard to chert-obsidian ratios. In connection with chipping waste ratios we will discuss another problem mentioned only incidentally above. This concerns a category of artifact we have called reworked bifaces in the tables for individual site and which are described below in an appendix by Nancy Whitney. This category of artifacts is made fran a leaf shaped piece, almost always of obsidian, snapped off or truncated and then further modified in various ways. One of the most canmon ways of modification was to use the snapped end as a striking platform for a blow along the long axis of the piece. Thus a long narrow flake would be taken off the edge of the piece which could then be described in European terminology as a burin. Fredrickson (1974:44) is, so far as we know, the first to discuss this in the North Coast Range. He attributes a burin industry to late Borax Lake Aspect and indicates that it continues into early Houx, this on the basis of his excavation at site Lak-261. In the following table we give the provenience of 66 specimens. These constitute all reworked bifaces and debitage identified as having cane fran reworked bifaces and which can be attributed to dated components. The components are as determined fran stratigraphic considerations given for each site group. The summary shows frequencies of 13 for late Houx, 38 for early Houx, 14 for late Borax Lake, and 1 for early Borax Lake. The single specimen fran early Borax Lake is fran a component of doubtful attribution and its presence there makes the attribution even more doubtful. On the whole we feel that our data support Fredrickson's conclusion that the industry begins in Late Borax Lake times and continues into Houx. There are sane other aspects of this chronology that are not quite so clear. If we take the numbers 13, 38 and 14 as representative frequencies of the three phases, we must conclude that the full flowering of the industry is confined to early Houx. The difficulty with this is same one encountered in considering chert-obsidian debitage ratios, namely that overall frequencies are dominated by those of a single site--Son-571. If we exclude that site, the relevant frequencies becane 13, 15, and 11--pretty much the same. The question canes down to this: we have found a major manufacturing center for early Houx at Son-571, have we missed comparable centers for late Houx and 190 Occurrence of Reworked Bifaces by Site arn Component Site Compnent H~b Ha BLb BLa 2 3 3 23 2 3 3 2 2 2 1 4 6 1 3 1 1 1 4 13 38 14 1 66 Group Western Eastern Yorty Lower Wr. Spr. Upper Wr. Spr. Total 567 572 571 568 593 582 608 576 556 553 544 551 Total 2 5 26 2 3 2 2 2 11 1 6 4 191 early Borax Lake? If we did miss them, it would be because of insufficient sampling at the sites in the project or because they were at sites outside the project. Our tentative conclusion is that we did not miss them, that in fact, this specialized industry is to be identified primarily with early Houx. This should be easy to test with further excavation. We have not been able to go into the question of the function of these tools. The answer to this important question may be the key to a lot of the archaeology of the region. It will have to await micro-analysis of these tools and other analyses based on association. If we turn to chronological summary of components, we see total components numbering 14, 11, 9 and 9 for the four periods latest to earliest. This indicates a stable population during this time--we would always expect more components of the latest periods because of accidents of preservation and excavation. Considering only the Western, Eastern ard Yorty Creek groups the comparable figures are 7, 6, 6, and 6 so that for the Upper Dry Creek tribelet we have even more stability. Taken together these figures indicate that the situation regarding tribelet territory and population, outlined in the next section, can be projected back through time as far as the early Borax Lake period. This argues really basic stability of subsistance pattern and social organization for many thousands of years in this part of California. This is in conflict with the conclusions of some others on this subject, notably Fredrickson (1974: Fig. 3), a point we will discuss further below. Whatever general validity the point has it is none the less in accord with the data of this limited area. Territory and Population The situation with respect to boundaries is complicated by reports of several ethnographers with sometimes conflicting statements. To begin with we should discuss the nature of Central California political units. The classic definition of these groups is by Kroeber (1932:258): "Each of these [tribelets] seemed to possess a small territory usually definable in terms of a drainage; a principal town or settlement, often with a chief recognized by the whole group; normally, minor settlements which might or might not be occupied permanently; and sometimes a specific name, but more often none other than the designation of the principal town. Each group acted as a homogeneous unit in matters of lard ownership, trespass, war, major ceremonies, ard the entertainment entailed by them." There has been considerable discussion of the political units of native California in recent years especially by Kunkel (1974) and Bean (1974). Both these authors seem basically to accept Kroeber's definition. Kunkel's discussion is especially relevant to us. In general we are quite sympathetic to his point of view, which is that native California social organization may be more typical of preagricultural peoples situated in favorable enviroments, more typical than "peripheral badlands wanderers" who are 192 supposed by some to be analogous to paleolithic man. However, one aspect of his argument we do not find supported and disagree with it. Kunkel sees the tribelet as being made of more fundamental political divisions which he calls residential kin groups (which may be patrilocal, matrilocal or amnbilocal). He says, furthermore, that "Tribelets were important but somewhat fragile political entities, breaking up fairly often into their component parts--the kin groups--which then combine in new ways." while it may be true that various kin groups changed tribelets fran time to time, our reading of the literature indicates that the tribelet, based on specified territory and with a central town, was the fundamental unit. The degree of attachment of California Indians to their home territory is extraordinary. We have heard of informants who would know the name of virtually every tree and rock within their tribelet territory while nothing less significant than, say, Mt. Shasta outside their territory would warrant their attention. R. F. Heizer (1966:26-30) has fully documented this attachment. Our point of view, then, is that the tribelet, territorily at least, was a relatively fixed entity regardless of whether persons might shift in or out of it. This is the important point archaeologically since it is only the territory we can identify, not the people. When we began archaeological testing at Warm Springs, we more less fell heir to the notion that the project area was divided between the Mahilkaune or Dry Creek Pano, called Mehinkownah by Merriam, and the Mahakamotcemei or Cloverdale Paono, called Makahmochumi by Merriam. This is essentially the division presented by Stewart (1943). As our investigations progressed, it became apparent that Dry Creek Valley above Pritchett Peak contained archaeological resources of what could only be regarded as a tribelet center (especially Son-593 and Son-582), not as an area subsidiary to the Cloverdale center over the hills in another drainage to the east. The ethnogeographic material of C. Hart Merriam indicates support for this judgement. Heizer's (1966) compilation of this material gives as inhabitants of Upper Dry Creek the Sha-kow-we-chum-mi and for Middle Dry Creek the Me-hin-kownah (Stewart's Mahilkaune). There is a problem with the boundary between these two groups. Heizer's map has it running along Warm Springs Creek (the boundary between his 12s aid 12u). This conflicts with Barrett's place names (evidently obtained fram informants of Middle Dry Creek) and with Stewart's boundary as well. We therefore place it further north so that the drainage of Warm Springs Creek is in Mahalkaune territory. In caning to this conclusion then we are putting the West, East, Cherry Creek and Yorty Creek groups in Shakowe territory while Upper Warm Springs and Lower Warm Springs Groups go into Mehinkownah lands. Looking at the archaeology of the boundary zone (Map 1 and Maps 19 anid 20) we see no sites of considerable consequence. Sites 542 and 600 may have been ethnographic sites and are the only sites in the Central Group for which we have indications of substantial midden. These two may have been significant villages, but they are at the eastern end of the Central Group, far below our proposed boundary. They have been destroyed so even that degree of importance is questionable. The other site indicating considerable population is Son-598. This site has 14 well defined house-pits indicating a population of perhaps 70. There is, however, absolutely no midden deposit here, indicating that the village must have existed for the briefest possible 193 time. O0ne wonders if this is not one of Kunkel's residential groups which split off temporarily fran one or the other of the tribelets to settle here in no-mans-land. Between the mouth of Yorty Creek and the cluster of sites at the mouth of Warm Springs Creek we have a housepit (562), a hunting blind (574) a large but ephemeral village site (598), and a minute midden (548). It is true that there may once have been sites here near Dry Creek that have been eroded away (see the gravels on Maps 19 and 20), but we doubt this because there are not even remnants there. Thus, the near blank here indicates a buffer zone between two tribelets and furthermore indicates its existence for a long time, perhaps thousands of years. This suggests that tribelet boundaries, or some of them, may last a very long time indeed. Fran the above we conclude that the Upper Dry Creek or Shakowe tribelet included the drainage of Dry Creek fran Pritchetts Peak most of the way up to Yorkville, the sites recorded herein constituting only the southern half. There appears to be an excellent settlement area just north of the project area but it has not been surveyed. The sites on Warm Springs Creek and just above its mouth on Dry Creek we attribute to Mahalkaune tribelet which had its chief village perhaps at the mouth of Pena Creek to the south. Thus, the sites we have fran this tribelet are peripheral and perhaps subsidiary. We turn now fran the consideration of the tribelet territory to the question of demography. A point of considerable importance, originally raised by Gifford (1918, 1926), is that of the nature and development of the corporate kin group in California and the relationship of these groups to demography. This was brought out by Goldschmidt in 1948 and has been recently discussed by Kunkel (1974) with reference to the Pono in particular. Goldschmidt's point was that lineages and clans arose because of large or at any rate locally dense population. While Kunkel believes the pertinent elements are not precisely lineages or clans, he would probably agree that population was a crucial factor. The burden of the argument concerns the number of persons with whan an individual has regular face-to-face contact, that when this number reaches a certain but unspecified point, it becanes necessary to form social categories for purposes of interaction. Fran our point of view the question is simply how many people are in a tribelet the regular face-to-face group. The tribelet we are most concerned with here is the Shakowe or Upper Dry Creek Pano because we believe we have the princil village of that group whereas the sites of Mahilkaune on Warm Springs Cree are peripheral and will not help us with that group. The problem can be dealt with either by means of individual site counts or by the area density method (Cook, 1976). The site count method involves determination of how many persons lived at each site and which sites were occupied simultaneously. On this first question we can do no better than consult the data of Cook and Heizer (1968) concerning house numbers and village size in California. The data fran their Table 2 contain the 194 following information for the surrounding groups. Aboriginal Group Village Space per House (sq. m) Yuki 73 Pano, coast 79 Pono, interior 158 Wappo, valley 120 Wappo, hill 190 We take the Pano of the interior as our best figure ard then take the total extent of midden as the size of the village and calculate the number of houses represented. If we assume an average of 6 persons per house, we then have a population per village estimate. As regards the point of simultaneous occupation, we want to consider only sites occupied regularly each winter. It is our opinion that only sites with considerable midden deposit should be so considered and surface sites as well such special sites as petroglyphs, hunting blerds, quarries, should be excluded from our calculations. Incidently, it should be pointed out that these opinions are actually hypotheses which should be tested in a variety of ways. Finally, we must deal with the question of chronology. We will present our data assuming that all sites identified as late Houx were occupied at the same time. We are aware that the question of micro-chronology is being ignored here, that if, for example, it was customary for one village to change sites every ten years, a custom gone undetected by us, then our calculation would be in error by a considerable factor. We might point out in this connection that the denigration of chronology by some persons who consider themselves theoreticians is entirely misplaced. If a fact absolutely fundamental to inferences regarding social organization, ard population is such a fact, deperds on micro-chronology then we simply must deal with it. Fran this stardpoint J. A. Bennyhof is more than justified in his concern for typological minutia and perhaps we can dispense with such remarks as "Good grief, yet another paper on California Indian beads!" In any case, we present our data as late Houx considering these the best approximations available. The following sites we consider to be permanent villages of the relevant time period: Late Houx Population Component 567 26 572 12 568 20 593 150-240 (depending on site size) 576 12 584 12 583 20 195 Site Son-593 has a variable estimate because it has been partially destroyed and we are uncertain about its former boundary. Thus the estimate is 102 persons for the small sites making the total of 252 or 342 depending on the Son-593 figure. Our judgement is that in the area above the project boundary there is one more locality about as well suited for habitation as those we have surveyed. This would mean we would expect about as many sites again as are already recorded for a total population about twice the above or between 504 and 684. The population question can also be approached fran the angle of area density. Considering that the northern boundary of the tribelet would be at the bend of Dry Creek downstream fran Yorkville and include the drainages of the tributaries fran there down to Pritchett Peak, this is an area of about 65 square miles. After considerable manipulation, Cook (1956) has decided that 8 persons per square mile is reasonable for this kind of territory. This gives us a total of 520, the same as our smaller figure above. Thus, our figures are quite similar though derived by distinct methods. The question now is whether 500 is enough people to initiate the process outlined by Goldschmidt (1948). In any case, it is a large figure considering that the valley seans to support only about a dozen people today. Site Specialization We have already discussed the subject of site specialization to some extant in relation to chippage ratios and now turn to the general question especially fran the standpoint of land use. In some ways this subject is the most important one in a study of this sort in which many sites are tested but none is excavated extensively. Our breakdown of sites yields the following broad categories: hunting blinds, quarries, petroglyphs, living sites. Living sites include both surface sites and midden sites and can be further broken down as shwn below. Hunting Blinds A single hunting blind (site Son-605) was located outside the project boundary was not included in the site descriptions. Four other blinds were found in the Warm Springs project, one each in the Warm Spring Creek groups and two in the Central Group. These blinds are simply rock rings 2 to 3 meters in diameter located away fran streams in places commanding actual or possible game trails. Unfortunately there seems very little hope of being able to date hunting blinds since they characteristically have no midden or other cultural refuse associated with them. Two of the blinds were cleared and excavated but even then nothing was found. One supposes that they cannot have been of great antiquity since they renained more or less intact but even this is not very convincing. The area surveyed at Warm Springs, centered as it was on the pool of proposed Lake Sonana, contained very little upland away from the streams. Such uplands would be the places one would expect to find hunting sites (and those we did find were so located), hence we believe our present sample leaves 196 this category of site badly under represented, as compared with, for example, midden sites. We feel that if the entire tribelet territory of Upper Dry Creek were surveyed as thoroughly as the project area there might be, ten, twenty, or fifty hunting blinds located. Quarries Just two quarry sites were recorded in the Warm Springs project, site Son-584 on Yorty Creek and Son-554 on a small tributary of Warm Springs Creek. Site 584 is associated with a midden which we have attributed to late Houx. The quarry itself may have been used earlier as well. Site 554 we have not been able to date at all. Nearby sites are heavy chert users in all periods so the quarry may have been used many thousands of years. In the nature of things a reconnaissance of a dam project, while it is likely to discover most valley bottom occupation sites, will find the quarries only if suitable outcroppings also occur in these areas; quarries are as likely or even likelier to be outside the area as within it. Thus our inventory of tribelet areas is as incomplete in quarries as it is in hunting blinds. All we can say is that the quarries we did find were used by nearby sites predominantly (see under chronology). No source analysis for materials was attempted and this should be done in the future. Petroglyphs Petroglyphs are quite an important element of the archaeological resources of the Warm Springs area. All petroglyph sites were in the territory of the tribelet above Pritchett Peak, none in the Central and Warm Springs Creek site groups. All petroglyphs here consist of small pits or cupules with an occasional groove, pecked into the faces of boulders of various sizes. It may be worthwhile to discuss briefly the absence of petroglyphs in the southerly groups. In the section on tribelet boundaries we concluded that sites of these groups were subsidiary to a tribelet center located down Dry Creek, out of the project area. In the course of our field work, we had occasion to conduct a site survey several miles downstream. This took us through what was presumably the center of the tribelet to which our southern sites belonged. However, we found not a single petroglyph site among those sites either. This might indicate that the people living there were not making petroglyphs and never had. Although that may be true, we are inclined to doubt it. All the sites located south of the project area were in fields long cultivated, and hence places in which suitable petroglyph boulders would have been removed. In any case, the downstream survey was only a hundred yards or so on either side of Dry Creek on alluvial fill with little or no natural rock. Thus there could very well be many unrecorded petroglyphs back toward the hills where there are suitable boulders. One petroglyph we did note in the territory of this tribelet was discovered by Dr. D. W. Ritter of Chico, California. It was at the confluence of Little Strawberry Creek and Warm Springs Creek-it consisted of pits and grooves on a small rock outcrop. There may well be other such petroglyphs. The petroglyphs upstream can be summarized briefly. A total of nine of 197 our numbered sites exhibit petroglyphs: site 566, 570, 568 in the Western group; sites 593, 579, 609 and 607 in the Eastern group; and sites 585 and 590 on Yorty Creek. These can be put in four groups for present purposes. (1) Site 566 is outside the project area ard may be associated with other sites up Dry Creek. It is apparently a considerable petroglyph on a flat rock 3 by 4.5 m with scores or hundreds of pits. We have not been able to assign the associated midden to a phase, and it otherwise isolated as well. (2) The two sites on Yorty Creek are also more or less isolated. Site 585 is a large boulder in a gravel bar in the stream ard has a total of 98 pits in it. It is only 300 m downstream from site 576 (Houx Pattern) but there is no clear association. Site 590 consists of two large rocks, also in the streanbed, with about 20 shallow grooves in them. There is no association here either but there are sites nearby of many phases. The other two categories require some additional ccnment. The sites in the Western Site Group category (3), consist of two petroglyph occurrences which are remarkable for size and number of cupules. Site 570 consists of two very large boulders covered with pits. The petroglyphs at 568 are on three similarly large rocks. These boulders are several meters in greatest dimension ard have hundreds or thousands of pits. The inference is unmistak- able that these were very important places to the inhabitants over a long period of time. The petroglyphs at site 568 are not actually with the midden but rather are 90 m upstream from it. Site 570, also without midden, is 280 m farther upstream. These large and important petroglyphs are evidently associated with a group of 3 midden sites which are in any case special sites. The situation is this: the midden sites 572, 571, ard 568 are very close to each other (300 m airline) on the north bank of Dry Creek on shallow terraces. Furthermore, there are other particular features (e.g. site 571 has a tremerdous amount of chipping waste) which we will discuss in following sections. The two groups of petroglyphs are found between sites 571 ard 568, and we feel therefore that they are associated with the site specialties of the little cluster. If we are correct in the association of these sites, then some effort at chronology can be made. The three midden sites contain components of all four of our later phases, thus the petroglyphs could belong to any or all of them. A test excavation was made just beside the petroglyph at site 568. We hoped to recover artifacts in direct association to help date the glyphs itself. However, we found none, not even the hammerstones which must have been used in making the pits. Indeed in the excavation of the three associated midden sites we found only a single specimen that could be described as a hammerstone. The nature of the glyphs is such that a great many hammrstones must have been used. We believe that the petroglyphs may have had considerable ritual or religious importance. If this is so the implements used in making them may also have had sacred significance and consequently were not left around casually but rather cached or otherwise secured in such a way as to make them now difficult to find. The last category of petroglyphs (4) is on the edge of Dry Creek, or even in it, beginning at Son-593 and then continuing sporadically downstream for about a kilometer at sites 579, 609 and 607. At site 593 we found a total of 36 small boulders (.5 to 1 m diameter) with one or more cupules in them. 198 Two of the boulders are on the south midden aid the remainder are on the side of the creek over a distance of about 300 m downstream. The other examples are probably an extension farther downstream of this same scatter. One site, Son-579, is now in midstream but was probably once on a midden since eroded by flooding. It is unusual in having circular grooves, one of which has a pit in its center. Of this group it would appear that at least the two on the midden at Son-593 are attributable to the very latest phase of prehistoric occupation. Just how much earlier they were being made is difficult to say. Perhaps it would be well to discuss now the general problem of petroglyph chronology in the North Coast Range. It is known fran ethnographic accounts that the people of this region made a sort of petroglyph of the type reported here which are called baby rocks. There are several accounts of these among which is one obtained by Barrett (1952:386-87) fran a man who claimed to have personally proved the merits of two cup and groove rocks in Knights Valley, six miles north of Hoplaid. "The procedure was said to be as follows: the sterile pair went to the rock and there first a prayer for fertility was made. Then, by means of a pecking stone, some small fragments were chipped fran one of the grooves or cuppings in the surface. These were then ground to a very fine powder which was wrapped in some green leaves and taken to a secluded spot. Here the powder was made into a paste and with it the woman's abdomen was painted with two lines, one running fran the top of the sternum to the pubes, the other transversely across the middle of the abdomen. Same of this paste was also inserted in the female. Intercourse at this time positively assured fertility, due to the magic properties of this rock." These "baby-rocks" were similar in form to others known further north where they are used for weather control aid are known as "rain-rocks" (Heizer 1953). Heizer aid Clewlow (1973:30) have argued that fram a stylistic standpoint the two types of petroglyphs are the same. These authors are of the opinion that it is a very recent style (30-31). They seem to base their opinion on the fact that since it is known to be late, and since there are very few examples in their sample, then it cannot go back much in time. We agree with them that these are equivalent styles and that they were made recently but not that they are only recent. The scarcity of reported examples we feel is from a lack of field work rather than a lack of petroglyphs. Our experience in the Warm Springs area indicates that camplete on-foot survey of an area does in fact reveal many petroglyphs. That this experience is not unique is shown by that of others (Fredrickson, personal caununication) in the area. The principal reason North Coast petroglyphs are not widely known is that they have not been reported by the general public (this is the main source of information for scholars of this subject). The reason they have not been so reported is two-fold: (1) many or most are obscured -by dense North Coast vegetation and (2) they are so unspectacular that they go unrecognized by the general public (and even some professionals). This latter reason is the most important one - petroglyphs and pictographs elsewhere are easily seen to be the work of man aid are fascinating to amateurs hence became well known, whereas those of the North Coast range are not so recognized or are easily 199 forgotten. Thus in disagreement with Heizer ard Clewlow, we believe there are actually very many of these petroglyphs in the North Coast Range and that they have been made for a long time in the area. The petroglyphs of Warm Springs would seem to us, in fact, to be very good examples of what has been termed the pit-and-groove petroglyph style. This style was first named by one of us (Baumhoff, Heizer and Elsasser 1958) arn] described in detail by Heizer arx] Baumhxff (1962:208, 234-238). It was maintained there that this was an ancient and widespread style and specifically included the examples from Lake and Soncma counties. This style is precisely the equivalent of those at Warm Springs. Heizer and Baumhoff indicated a belief that in the Great Basin the style dated about 5000 to 3000 BC (1962:234). The evidence for that dating was admittedly rather flimsy but it still seems clear that an early beginning date is not out of the question. The fundamental questions on this subject concerns the unity of the style and whether these particular examples are correctly attributed to it. We quote Heizer and Baumhoff (1962:237) on this. "These California petroglyphs differ fran the Nevada specimens in that, for the most part, they are simply pitted boulders; they do not also have grooves. This differentiates them slightly, but the pits themselves are so similar and at the same time so arbitrary that one cannot help thinking they must be part of the same tradition." We agree with this and also accept the earlier finding of 5000 BC as a beginning date for the style hence we conclude that they were being made during all four phases of our sequence. It seems possible that this petroglyph style is a fundamental part of an ancient and widespread culture of speakers of the Hokan languages. So far as we know this notion of such a cultural group was first advanced by Kroeber (1923:130) when he said that in the "first period" (undated) of man in California "people....almost certainly camprise the ancestors of the modern Hokans,..." aiding "perhaps the Penutians." This notion was apparently never abandoned by Kroeber although later discussions are more about the language itself than other aspects of their culture (cf. Kroeber 1955, 1959). Sane time ago one of us took up this notion of Kroeber's and suggested that certain archaeological traits were part of this basal Hokan culture (Baumhoff and Olmsted 1963:280). We would like to propose here the pit-and-groove petroglyphs as an element of that ancient culture. Our feeling is that it was an important part of the ritual of these early Hokan speakers, that the particular ritual declined in the latest period, perhaps replaced by the Kuksu cult of the ethnographic period, but was retained as a survival in its fertility context (weather control farther north). We would say that the large and impressive petroglyphs upstream were made as part of the earlier more fundamental ritual while the smaller ones near Son-593 and downstream may have been later and residual. This hypothesis obviously leaves plenty of roan for testing. We will have more to say about this in discussing overall site configuration. At this point it is sufficient to say that while the evidence for the above case is thin, we nevertheless have a considerable intuitive confidence in it. What we need most here is a method of dating petroglyphs. We know fran experience that this is an extremely difficult business, but it is quite basic to a whole range of prehistoric behavior because petroglyphs 200 are perhaps the most important and abundant ritual relics left to us. Living Sites The living sites will be considered in two categories. Sites above Pritchett Peak will be considered as a unit (Western, Eastern, Yorty Creek, and Cherry Creek site groups), an integral tribelet territory centered on large winter village site(s) associated with smaller sites specialized in a variety of ways. In the southern part of our area, thought to be peripheral to another tribelet, sites will be dealt with idividually on an ad hoc basis. Living Sites: Southern These are the sites of the Central Group and the two groups on Warm Springs Creek. With respect to the Central group, the long stretch of Dry Creek between Pritchett Peak aid the mouth of Warm Springs Creek, we can say that there are 3 small surface sites (562, 548 and 560) and one large surface site (598). In addition, two other sites were destroyed by agriculture. The destroyed sites may have been a 5 house village (542) and a large village (600) but we feel certain from having tested the rennants that they were never very rich and probably not important. In the two site groups on Warm Springs Creek there is a total of 14 living sites. Of these, 8 fall in the categories including small or medium in area, mostly poor aid either very shallow or surface. They must have been temporary camps for small groups. These sites include the following: Son- 558, 557, 559, 555, 545, 546, 550, 549). The remainder include the Skaggs Springs site (594) which is entirely destroyed but which, according to Barrett (1908:220), may once have been an important site and site Son-547, which is dealt with in Appeidix A. The other four sites may be characterised as 5 house hamlets - they are deep, rich, and of moderate size. TNo of these sites have canponents of all four phases (576 and 544), one is early and late Houx (553), and the other (551) canbines late Borax Lake aid early Houx. A striking feature of the Warm Springs Creek groups is the large number. of hamlet size sites (some of the shallower sites are that size too) as opposed to the fact that there are none of the minute but rich stations that are so evident in the Northern Groups (see below). We do not know what to make of this aid simply leave it for further research. Living Sites: Northern For the northern area, it is possible to establish clusters of sites according to trichotomous criteria. The first category is size, dividing sites into small (one to two houses), medium (four or five houses), aid large (20 or more houses). The second division is depth according to which a site is surface, shallow (20-80 an depth), or deep (a meter or more). Finally sites can be categorized as rich (199 flakes per 10 an aid abundant artifacts), poor (10 or 20 flakes per 10 an aid few artifacts), or intermediate. Applying this trichotomy, the data on the following table are obtained. 201 Warm Sprirgs Sites Catagorized by Area, Depth, and Richness of Midden Large rich med poor Medima rich med poor Small rich med poor 593 582 597 566 572 576 567 571 568 584 608 578 583 577 575 603 604 589 601 595 579 586 564 541 581 596 592 588 587 Deep Shallow Surface 202 The sites shown there segregate fairly well into five groups. The first group consists of the large, deep sites 593 and 582. These are undoubtedly the large winter villages or headquarters sites. Site 582 is entirely Borax Lake while 593 is largely Houx. Thus during Borax Lake times the bulk of the population lived at 582 (593 was probably a minor village then) and moved to 593 during the later period. The second group consists of the sites which are deep and of medium size into which we assimilate site 608 which is formally categorized as shallow but at 80 an only just barely so. Of these six sites 597 and 608 are Borax Lake while the other four are Houx. W see these sites as being satelite hamlets attached to the major towns in the first group. This probably argues a larger population in Houx than Borax Lake times although not necessarily so since later components are generally identified more readily than earlier ones. Just what these "satelite hamlets" represented sociologically is far from clear. A casual persual of Barrett (1908) and Kroeber (1932) does not reveal much about such places. Kroeber (1932:257) says there are "minor settlements which might or might not be occupied permanently". These are evidently those occupied permanently. There is not a lot to indicate that they were specialized fran what we have been able to discover archaeologically although more might well be found with additional excavation. One change through time can be noted--the two earlier sites in this group are "rich" while the later ones are only "medium." This means that there was a relatively large amount of chipping debris and other artifact material at the two earlier sites which in turn may very well mean that artisans, particularly stone chipping specialists, were found at the 4 to 5 house hamlets during Borax Lake times but not later. The third group consists of the small, deep sites. Again it seems reasonable to assimilate site 583 to this group since it is within 10 an of being "deep." These four sites do sort neatly chronologically. They are divided as follows: 572 has caupnents of all four phases, 571 the middle 2 phases, 576 the last 2 phases, and 583 the first and last. In analyzing this we begin with the fact that Son-571, particularly its early Houx canponent, is a chipping station pa excellence--nearly a quarter of the obsidian debitage from the entire projecF is fthere. Furthermore about a third of the products fran the specialized reworked biface industry are fran there.. Thus, for the early Houx period, we can say that this is a specialist site. To a lesser extent the same can be said of Son-572 and 583 in regard to their early Houx and Borax Lake components. However, the same is not true of Son- 576, which represents the last phase. This site hai every appearance of a rich site during excavation--dark, rich, deep midden with much perserved bone and charcoal (very unusual). It hai every external appearance of being unusual as well--a tiny, deep site on a small point of land just big enough for a house or two. Yet the midden content is very slight as canpared with other sites. If this site was special, then it was so in non-material or perishable goods. Fran the above evidence we may conclude that during the Borax Lake period certain single house sites as well as 5 house hamlets were occupied by specialists, that during the early Houx period only the one house sites were so occupied, and that during the late Houx period the specialists moved to the main village (Son-593). 203 The two final site categories include those sites which are usually poor, always small, and are either surface sites or have very shallow middens. These sites appear to differ from those above in displaying an occupancy of the most intermittent character; the others, while they may not have been occupied year round, must have been occupied year-after-year over a very long period of time. Unfortunately it is impossible (with few exceptions) to sort these surface sites chronologically. Probably they were inhabited by one or a few families for special purposes, such as acorn gathering or deer hunting, for a few weeks once every several years. The determination of such functions requires reconstruction of micro-envirornents in detail which is impossible at the monent but is not necessarily impossible as a long range goal. In addition to the sites in the five categories above there are two sites which do not group at all. These are a large surface site (579) and a medium sized, shallow midden site (578). They are on opposite sides of Dry Creek fran each other about a quarter mile downstream fran Son-593 which we consider to be the major winter village of the Houx period. As noted in the site description site 579 may have once been more extensive ard been washed out by flooding. One conjecture is that these two sites were adjuncts to the main village at Son-593. The considerations presented above for the living sites of the northern area (above Pritchett Peak) can at this point be canbined with earlier discussions to present an overall scheme for the development of the tribelet there. During early Borax Lake times site 582 at the confluence of Cherry Creek and Dry Creek was the major winter village. At this time site 597, near the confluence of Yorty Creek arn Dry Creek was a 5-house hamlet. One- house sites were at 583 up Yorty Creek ard 572 up Dry Creek. There is slight evidence of occupation at 593 (destined to become the major village later). The only evidence of specialization in this period is the large quantity of chert debitage at site 572 indicating a heavy emphasis on chipping there. We know of no quarry nearby but there could easily be one outside our survey area. Obsidian is very scarce in deposits of this period. Given the imprecise nature of our stratigraphy we suggest the possibility that obsidian may not have been used at all during early Borax Lake times here and that the few pieces we have recorded are due to accidents of deposition, recording, or interpretation. In the late Borax Lake period the same principal village site is continued at site 582 but the satellite configuration changes. Downstream occupation continues at 597 (at the mouth of Yorty Creek) with an increase of debitage and general richness. Between these two a new hamlet (Site 608) is established now and continues only during this period and leaves quite the richest remains of the period; one wonders if it was not a chief's or shaman's residence or something of the sort. Upstream from the principal village occupation continues at 593 (still minimal) while another mile upstream a complex of sites begins which in the following period is to flower spectacularly. At this time the complex consists of sites 571 and 572. Site 572 hai already been occupied in early Borax Lake times and had already shown signs of specialization in chert manufacture. It is time now to discuss these two sites which together form the 204 N~~~~~~~~~~~~ ._.._*> 4 I3 ~o '-4 4 N ' . o + . _ I \ L . _ No p- x Map 30. Settlement Pattern Chrorxzlogy. 0 Q0 N '~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~N N N N~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l U ,~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~U Ma 0 K Seteet atr irnl*v 205 nucleus the canplex just mentioned. In many ways they are twin sites. Only a quarter mile apart, the middens are found on a very narrow terrace just above the north bank of Dry Creek heavily overgrown by oak and madrone. Both sites are on beautiful, deep pools in the creek. The middens are both very small and could not have accacodated more than two houses (one house is more likely). Perhaps most importantly we find that both sites have areas on terraces below the middens imnediately adjacent to the creek which have an abundance of grinding tools. We have designated these as grinding areas but it may be that they were in fact also occupied as living sites and that the evidence of that has been washed away by floods, leaving only the heavier grinding tools remaining. Whatever the explanation of these lower areas maybe, it is evident that these two sites are remarkably similar in external features. The middens, however are not the same. Site 572 has components of all four periods while 571 has only the middle two. Site 572 is categorized as a rich site arK] with justice, especially at the bottom,, but not outstandingly so. Site 571, on the other hard, is by far the richest site we tested on this project, arK is particularly so in its late component (early Houx). Thus we find in late Borax Lake times the beginning of rather extreme specialization at these tiny living sites. It is also at this time that we find a very great increase in the use of obsidian rather than chert (perhaps it is even the beginning of the use of obsidian, as conjectured previously), and here too we find the beginning of the specialized reworked biface industry. Finally, the Excelsior point, which is to typify this and the succeeding phase, begins here. This is an impressive number of changes. One thing striking about these is that they seem to prefigure the florescence which occurs in the next phase; they are, so to say, formative. In the early Houx period there are very substantial changes in settlement. The first of them is that the main village at site 582 was abandoned and the bulk of the population moved to site 593 which had previously been only a hamlet. In the description of that site earlier we mentioned a possible occurrence which may have been related to this move. This suggested that a landslide at one time forced Dry Creek to the north slightly, against what is now the north portion of Son-593 and only after the creek had cut through this was the present large flat available for settlement. If this event occurred shortly before the time of Christ, it may have been what prompted the move to this inviting place. If this is so, however, it is a remarkable coincidence because it happened simultaneously with great changes upstream. Upstream, at the complex around 571 and 572, where we had 2 one-house stations, we now find in addition a hamlet of 5 houses just below at site 568 and another about a mile above at site 567 so that the population of the little enclave changes from about a dozen to 75 or more. At the same time the hamlets south of Cherry Creek (608 on Dry Creek and 597 on Yorty Creek are abandoned. Another remarkable change that occurs at this time is shown in the midden content of site Son-571. This component yielded a total of 4210 obsidian flakes. The next highest count in our digging is 929 and that comes from the late Houx component of Son-593 (also significant); no other is more than in the 400's. Thus there is evidence of overwhelming specialization at this site. Further evidence is found in the distribution of reworked bifaces; 206 23 specimens of a total of 66 for the whole project cane from this component while the next highest figure for the canponents considered here is 3. We can only conclude from the above that tremendous things were happening at the 571- 572 canpiex in the early Houx period. We suggested earlier that we believe the large petroglyphs nearby (one designated Son-570 and the other included with the description of Son-568) were associated with this complex of specialist sites. We also indicated that we were unable to date these glyphs. If the glyphs are part of the site complex, as they seen most clearly to be, then we would argue that some of them were made in Borax Lake times and the bulk of them in early Houx. We have earlier argued that these petroglyphs constitute the principal remaining evidence of what was once a set of fundamental Hokan religious observances. We do not know the nature of this religion, studies of comparative Hokan have not approached anything of this sort yet, but, if it was as important as we think, then it must have been deeply embeded in local social organization. By analogy with North American ethnographic examples we can believe there must have been specialists, priests or shamans, associated with it. Consider this with what we have said previously; we have living at the two stations and associated hamlets (1) the traders, since the obsidian is there, (2) the artisans, because of the obsidian chips and reworked bifaces, and (3) the priests. We have all these things present to sane degree in late Borax Lake, and they came to a peak of intensity in early Houx. In the face of this we can only cane to the conclusion that the rich and powerful lived at the 571-572 canplex while the canmon people lived downstream at Son-593. The evidence here is obviously not conclusive, and the notion needs a lot of further testing, nevertheless it is the most convincing case for such social stratification we have seen in California archaeology. One other thing happened in the early Houx period that requires discussion and that is the introduction of the mortar and pestle as the principal grinding tools. Fredrickson (1974:46) indicates the introduction of this canplex into California earlier than we find it (3000 to 2500 BC), and he presumes that it is associated with acorn processing technology. As a general statement, we have no quarrel with this, but it is evidently not true for our small area; the mortar and pestle complex seems much later. It may be that our evidence is faulty, the sample of grinding stones is regretably small, but what we have is consistent with Fig. 1. On the other hand the number and size of late Borax Lake components is such that there seems no possibility they did not have the remainder of acorn processing technology--it is inconceivable that the number of people represented could have been supported without really large scale recourse to the acorn crop. This aspect of subsistence development seems to us to be a really major question for Warm Springs archaeology for which the answer is not even suggested by the testing thus far carried out. In the final period, the late Houx, we again find major changes in settlement pattern. Site 593 continues as the major village but the midden changes--the debitage changes fran moderate to abundant and, as noted, has the second highest frequency of obsidian in the project. At the same time our upstream canplex sites 571-2, site 571 is abandoned and 572 has only 20 an of deposit. The hamlets continue to be occupied at sites 567 (slightly) and 568 207 (extensively) but now there is no specialization to be associated with them; also there is a hamlet once again on Yorty Creek at site 584. As indicated in the discussion of petroglyphs above, we believe that when the 571-2 complex was for all practical purposes abandoned in late Houx times, then the petroglyphs associated with it were also abandoned. We believe the small petroglyph boulders near site 593 are survivals fran the earlier period now used as in the ethnographic period in connection with fertility rituals. We think that one of the events of this period was the acquisition of the Kuksu cult. The Kuksu or "big heal" religion was centered on a men's secret society which conducted dances or other performances in which the initiates impersonated the god Kuksu. These performances took place in a large earth covered lodge. Kroeber believed that this was a very late development originated by the Patwin and spread rapidly fran there to nearby areas including the one under discussion here. Kroeber's argument was presented in 1925 and was reiterated as late as 1948 (564-568). Kroeber's argument was based essentially on an age-area notion. Such notions are not now fashionable, and we agree that a great deal skepticism is most alvisable in applying them. However, one of us recalls a statement once made by R. H. Lowie the burden of which was that while Kroeber's large scale conjectures were apt to be suspect, small ones that he male about material with which he thoroughly conversant were usually correct. We agree with Lowie's assessment generally, and furthermore, since Kroeber's conjecture about the Kuksu cult being very recent among the Pano seems inherently probable, we believe it marks the beginning of late Houx, and the attendant demise of the earlier religion, the Hokan cult, which was associated with the massive petroglyph boulders. We might mention in passing two ways in which the above conjectures might be tested. For one thing, if we could find a way to date the petroglyphs, we would be on much firmer grounds. This will probably have to be done either through associated artifacts (doubtful but possible) or by means of relative patination (not yet feasible but perhaps so in future). The other test would be by identifying archaeologically the large earth covered lodges which Kroeber said were associated with Kuksu. In this area, the latter will require extremely fine micro-stratigraphy, if our experience is any guide, and even that might not be good enough. We were unable to identify house floors even when housepits were present. In any case, if we are correct that the earlier religion with segregated priests broke down and that the traders and artisans moved back to the main village, then this suggests that the stratification that we saw in the earlier period had now broken down into a more egalitarian situation. Wtether the inception of the Kuksu cult was a causative agent in this or whether it was just one part of a more general change we cannot say. The archaeological inventory does change slightly. A new point type is introduced (Gunther barbed, is this also fran the Patwin?) and the reworked biface industry declines to its late Borax Lake level. Generally speaking late Houx seems poorer than early Houx. The foregoing discussion is further summarized in the table that follows. 208 Site Characteristics by Type and Phase Late Houx Site Depth Cht-Obs No. an Ratio Flakes Rework Proj. Biface Points Village 593 70 0.14 1043 3 6 567 20 4.42 65 - 1 Hamlet 568 80 4.28 586 2 9 584 40 3.35 287 - 2 572 20 0.48 151 3 4 Station 576 50 4.09 280 - 1 583 30 9.79 329 - 2 Early Houx Village 593 40 0.49 257 567 80 2.64 315 2 5 Hamlet 568 50 12.27 146 - 1 571 80 0.06 4454 23 19 Station 572 50 0.40 485 2 2 576 140 1.03 542 2 3 Late Borax Lake Village 582 80 1.40 867 2 6 593 70 7.50 51 - - Hamlet 608 80 1.40 1122 2 2 597 50 4.47 711 - 3 Station 571 50 0.21 791 3 2 572 30 3.06 418 - 2 Early Borax Lake Village 582 50 4.00 60 - - Hamlet 593 30 7.00 24 - - 597 110 19.02 301 - 2 Station 583 60 3.81 289 - - 572 40 61.71 439 - 2 209 It is possible, and for some purposes useful, to view the situation as a developmental sequence in which we would call early Borax Lake incipient, late Borax Lake formative, early Houx, classic and late Houx post-classic. The incipient period would be the first development of tribelet organization, the formative would see beginnings of trade arn specialized industries which culminate in the classic period anx] finally tail off in the post-classic. This would define the cultural growth of the Southern Pono in their distinctness up to eight hundred or a thousanxd years ago when they became more closely affiliated with other Central Californians, especially the Patwin and Miwok. To look at this sequence in another way we can see it as progressive Penutianization of the Hokan peoples. Kroeber says that Kuksu is a basically Penutian cult. This means that the postulated Kuksu spread in our post- classic or late Houx period was in fact the Penutianization of Pamo religion. But this was only the last of a series of changes brought about upon what we regard as the basic California culture of the Hokan people. At some time in the distant past, probably Fredrickson's 2500 to 3000 BC is as good a dividing line as any, California was occupied almost entirely by Hokan speakers (ard perhaps Yukian speakers as well). At this time ancestral Penutians came into the area ultimately occupying most of the favored portions. They may have brought with them a good deal of acorn technology or they may have developed it soon after they came. In any case there is apparently a great increase of population at this time which must have been deperdent on acorn gathering. Soon after this the Penutians began to influence the native Hokans, pushing them out to the margins or isolating them there arx] at the same time exerting cultural influences on the remnant populations. The native Hokan culture apparently differed fran that of the intruders in using harxnstone and milling stone for grinding rather than mortar and pestle, in making twined basketry rather than coiled, and in making pitted boulder petroglyphs while the early Penutians made petroglyphs, if any, more like those of the Great Basin. Fredrickson thinks the mortar anid pestle replaced the mano and metate by 2500 BC while we feel that this change occurred 2000 years later at Warm Springs. Pitted boulder petroglyphs we think remained very important until 1200 AD when they became mere survivals. If bone were better preserved so that bone awls could be recovered in large numbers, we could probably date the introduction of coiled basketry. Unfortunately this is not the case, and we remain in the dark on this issue. These are a few of the changes that came about through contact of the Pono with Penutians. We think a similar series of changes may have taken place among, for example, the Chumash and the Yana. In fact from this point of view one might see the culture history of California in the last 5000 years as the Penutianization of the Hokans, in some places earlier, in others later, and still others never complete at all. At the same time there was a process of adaptation of the Penutians to local conditions. Consideration of both processes will generate a whole series of hypotheses some of which have been outlined above. 211 BIBLIOGRAPHY Barrett, S. 1908 A. Whe Ethno-geography of the Pawo University of California Publ. Ethnology 6: 1-332. ax] Neighboring Indians. in American Archaeology arn 1952 Material Aspects of Prno Culture. Bull. of the Milwaukee Public Museum, 20: 1-508. Baumboff, 1963 Martin A. Ecological Determinants of Aboriginal California Populations. University of California Publ. in American Archaeology and Ethnology 49: 155-236. Baumhoff, 1958 Baumlff, 1963 M. A., R. F. Heizer arn H. B. Elsasser Lagomarsino Petroglyph Site, Storey CouLnty, Nevada. of California Arch. Survey Report, no. 43. M. A. and D. L. Olmsted Palaihnahan: Radiocarbon Support American Anthropologist, 65: 278-284. University for Gloto-chronology. Bean, L. J. 1974 Social Organization in Native California. Antap: California Indian Political and Social Organization. Bal Press. Cook, S. F. 1956 The Aboriginal University of 3. Population of the North Coast of California. California Anthropological Records, vol. 16, no. 1976 The Population of the California Indians 1769-1970. University of California Press. Cook, S. F. 1965 aid Robert F. Heizer The Quantitative Approach to Settlement Size. University no. 64. the Relation between Population and of California Arch. Survey Reports, 1968 Relationships among Houses, Settlement Areas, Aboriginal California. IN: K. C. Charn, Archaeology. Fredrickson, 1973 and Population in (ed.), Settlement David A. Early Cultures of the North Coast Rarnes, California. dissertation, University of California, Davis. Ph.D. 1974 Cultural Diversity in Early Central California: A View frao the 212 North Coast Ranges. The Journal of California Anthropology 1: 41-53. Gifford, E. F. 1918 Clans ard Moieties in Southern California. University of California Publ. in American Archaeology, 14: 155-219. 1926 Miwok Lineages and the Political Unit in California. American Anthropologist, 28: pp. 389-401. Goldschmidt, W. R. 1948 Social Organization in Native California ard the Origin of Clans. American Anthropologist, 50: 444-456. Harrington, Mark R. 1948 An Ancient Site at Borax Lake, California. Southwest Museum Papers, 16. Heizer, R. F. 1953 Sacred Rain-Rocks of Northern California. University of California Arch. Survey Reports 20: 33-38. 1966 Languages, Territories, and Names of California Indian Tribes. University of California Press. Heizer, R. F. and M. A. Baumhoff 1962 Prehistoric Rock Art of Nevada and Eastern California. University of California Press. Heizer, R. F. and C. W. Clewlow, Jr. 1973 Prehistoric Rock Art of California. Ballena Press. Kroeber, A. L. 1923 The History of Native Culture in California. University of California Publ. in American Archaeology and Ethnology, 20: 125- 142. 1932 The Patwin ard their Neighbors. University of California Publ. in American Archaeology arn Ethnology, 29:253-423. 1955 Linguistic Time Depths So Far and their Meaning. Int. Journal Amer. Ling., vol. 21: 91-104. 1959 Recent Ethnic Spreads. University of California Publ. in American Archaeology ard Ethnology, 47: 259-281. Kunkel, P. H. 1974 The Pano Kin-Group arn] the Political Unit in Aboriginal California. Journal of California Anthropology, vol. 12, no. 1. Meighan, Clement W. 1955 Archaeology of the North Coast Ranges, California. University of California Arch. Survey Reports 30: 1-39. 213 Meighan, Clement W. ard C. Vance Haynes 1968 New Studies on the Age of the Borax Lake Site. Masterkey 42: 4- 9. 1970 the Borax Lake Site Revisited. MacDonald, 1975 Science 167: 1213-1221. IRderick ard Elizabeth Honeysett An Ecological Survey of the Wann Springs Archaeological Project Area Sonana County, California. Report prepared for the U.S. Corps of Engineers. McCullough, Dle R. 1971 The Tule Elk, its History, Behavior, ard Ecology. California Press. Ritter, Eric W., Brian W. Hatoff, ard Louis A. Payen 1976 Chronology of the Farmirngton Complex. American 334-341. University of Antiquity 41: Stewart, 0. C. 1943 Notes on Pano Geography. University of California Publ. in American Archaeology anx] Ethnology, 40: 2942. Treganza, A. E. 1958 Salvage Archaeology in the Trinity Reservoir Area, Northern California. University of California Arch. Survey Reports, 43: I. 215 APPENDIX A POSSIBLE PRE-BORAX LAKE COMPONENTS The four-phase sequence fran Early Borax Lake to Late Houx (Fig. 1) we have argued represents a continuous occupation by the Pono or their ancestors dating' fran 5,000 BC up to the historic period. In doing our field work we were alive to the possibility that earlier material might also exist. Most recent work in the North Coast Range suggests that some categories of earlier remains may exist but were not found by us. The recent work referred to is at Lake Berryessa, the reservoir behind Monticello Dam on Putah Creek some 50 miles southeast of Warm Springs. The drought of 1976-77 was such that the lake level there had dropped about 50 feet by May, 1977. The sides of the lake thus exposed were largely devoid of vegetation and also been modified by the action of the water, in some places covered with silts apparently the result of lateral slope movement but in others eroded by beach action or by the lateral movement itself. One of the results of these processes was to expose archaeolgoical remains of a very different nature than those found prior to the construction of Monticello Dam. At this point we are conjecturing that these remains were covered by a soil horizon making them impossible to discover and the lake action dissolved or otherwise moved this covering soil. It is entirely possible that similar remains are present in the Warm Springs area but went undiscovered because they are similarly covered by recent soils. Really extensive augering is the only way that these could be discovered. Aside fran this possibility we know of only two possible candidates for pre-Borax lake components in the Warm Springs area: The Yorty Creek gravels and the bottom layer of Site Son-597. Yorty Creek Gravels We noted earlier in the summary given for the Yorty Creek Group the existence of a possible early component on this small creek. Three sites there have a sub-midden gravel component, Son-593, 584, and 577. These sites are all on Yorty Creek, in the northeastern part of the project area. At 583 and 584 the gravels were encountered at 90 acis and 60 acs below the surface, respectively, and extend to 120 ans and 135 axs where they overlie bedrock. At 577 the gravels were encountered at 100 axs and excavated to 180 ams where water table percolation brought the excavation to a halt. In addition, we have found artifacts in the gravels in the stream-cut banks of Yorty Creek, extending fram the eastern edge of the project area at Son-583 to just upstream of Sot-576, the site at the Sheep Shearing Station. These gravels generally are one to two meters below the surface overburden. Chert artifacts only have been found in the gravels exposed by the stream cuts. At 583 and 577 the artifacts in the gravels are entirely of chert; 216 obsidian drops off canpletely. At 584 there are roughly equal amounts of chert and obsidian. At all locales the chert industry can be characterized as consisting of large percussion flaked informal tools made on locally available red, brown, and green fine-grained siliceous rocks such as chert and jasper. No grinding stones or diagnostic artifacts such as projectile points have been found in association with this industry. Approximately 140 artifacts (tools and flakes) have been found in the gravel component of the tree sites; of these about 60, or 43% show some evidence of having been water worn; the remainder are fresh appearing. In the opinion of Mr. Dick Thompson, geologist, Corps of Engineers, San Francisco District, who examined the sites and stream cuts, the artifacts in question have not been moved any great distance and then redeposited but have been water rolled locally with respect to their point of origin or original site context. Six of the most convincing artifacts are shown in Fig. 30. We noted earlier that these must be earlier than Early Borax Lake. Just where it is to be placed in this open ended time period is not now possible to say. Neither is it possible. to define an "industry" or "culture" more precisely so as to relate to other early California manifestations. It is our estimate that a large amount of excavation and additional observation of exposed gravels together with assistance from geological specialists will be required before any archaeological answer are obtained. Site Son-547 Son-547 is stratigraphically one of the most interesting in the project. The top layer of the site is terribly enbeded with roots. Orlins finds that this is the heaviest root growth ever encountered by him in an archaeological site. One expects that some mixture of artifacts would be caused by this churning and since the roots were most abundant in the top 30 an one would expect the most confusion at this level. On the contrary, it is found that this level is consistent stratigraphically and also interms of chert-obsidian ratios. The next zone (30-90 an), which has a smaller growth of roots, fluctuates strikingly in chert-obsidian ratios and has artifacts from three different time periods. One can only hope that this represents nothing more serious than a sampling problem which could be obliterate by additional excavation. Finally, in the deepest levels (90-140 an), we encounter the anomalous situation of debitage consisting almost totally of obsidian in contrast to the other sites in the Project which have chert as the principal or exclusive component of deep level debitage. We grant that a total of 55 chips is a small sample but the consistency of the low chert-obsidian ratios in the four levels adds sufficient weight so that its significance can hardly be thought to be spurious (see Table p. 164). What then is the explanation? At this depth it can scarcely be due to root mixture and, in any case, root mix would have been more likely to bring down chert than obsidian since chert is the dominant material in the site as a whole. We must look elsewhere for an explanation. We note that the nearest obsidian source, Mt. Konoci, is only about 25 airline miles distant so it is not difficult to imagine people obtaining material from there either by trade or transport (although we do not know that this was the source). The striking feature is the contrast with other early deposits both here and elsewhere in California in which only strictly local material is found whether it be chert, 217 obsidian, or sone other substance. In the usual case of early deposits we thus have evidence of extrene localism or, in other words, of tribelet organization. The difference between this and later situations is that in addition to tribelet organization we also have extensive trade networks. The situation at Son-547 indicates either a very early system of trade relationships or, alternatively, a pretribelet system of social organization very much more free ranging than has been present in California in more recent times. To assume the first alternative is to assume a once wide trade network subsequently cut off entirely only to begin again at yet a later date. %tile this is possible it seems to us that the second or per-tribelet, free ranging hypothesis is preferable and we therefore tentatively accept it. Admittedly the present evidence is a rather slerner thread to support any considerable hypothesis but then it is, after all, only a hypothesis. 219 APPENDIX B AUXILIARY TECHNIQUES Because of the exigencies of court action against the Corps of Engineers a variety of archaeological aids which could have assisted in analysis of the Warm Springs materials were not performed. Among the most obvious of these were faunal analysis arn obsidian analysis (both source and hydration). In the North Coast Range the abundance of rainfall combines with acidity of soil to make preservation of bone almost impossible. Perusal of the tables included with each site description reveals that this statement almost but not quite true. Several of the sites, notably Son-593 arn 576, produced fairish quantities of bone. Unfortunately nearly all the bone consists of tiny unidentifiable fragments so its usefulness will ultimately be derived through techniques different than those usually employed. However, there were a small number of identifiable pieces arn a large excavation sample may yield sufficient, quantities to be useful. Obsidian source arn hydration analysis is another matter. This could very well be quite revealing with regard to trade patterns and chronology. This is being undertaken at present and we hope to make it available in the near future. We did perform both a palynological assay anx a water separation or floatation project. The results are presented herewith. Palynology Several samples fran one site (Son-582) were submitted for purposes of pollen identification to Mr. James West of University of California, Davis. West centrifuged the samples and prepared slides (which are still available). He reports that no pollen remained in the samples. It may be that other sites have superior pollen preservation qualities; we would think that Son-593 and 576 would have the best chance in this regard. Water Separation A floatation analysis of five of our sites was carried out by Terry Zontek ard Jane Melville. They report the following procedures and results. Their detailed site by site results are also available. Procedure At each site tested a ten centimeter wide column sample was removed fran one wall of the test unit excavated. The sample was taken out in ten centimeter depth increments, producing a ten centimeter cube, representing a one percent sample of each ten centimeter unit level. The vertical arnd horizontal boundaries of each sample were scratched on the wall with an icepick using a meter stick as a straight edge. In most cases the samples were removed beginning with the 0-10 an level, surface representing 0 an. A dustpan was placed along the bottom border of the sample as the sample was cut out of the profile with a square-nosed trowel. The samples were then bagged, 220 sealed, arni labeled with provenience data. The tools for the water separation method were an ordinary fine mesh tea strainer ard a large galvinized washtub with the bottom replaced by a 1/16" mesh screen strengthened by braced metal cross bars. For each sample processed by this procedure a place was chosen which had a pool of water at least 18" in depth through which ran a moderately swift current. The depth was important for subnersing the washtub and the current necessary to carry the soil away form the work area. The washtub was placed on the bottom of the pool and the sample was poured in with one hard while the other produced a mild turbulence. The light fraction was then scooped off the water surface with the tea strainer, the tub raised ard lowered to further release floating material, and that material collected. Both the light fraction ard the heavy fraction on the bottom of the screen was bagged, sealed, and labeled with provenience data. The first step of laboratory sorting was done with tweezers and a hard lens. The sorting categories for the light fraction were organic material (charcoal, seeds, seed husks), bone and other organic matter (roots, rootlets, etc.); for the heavy fraction, bone, lithic debris and rootlets. The organic matter was then examined under a Bausch and Lomb variable power dissecting microscope. A clear plastic ruler with 1/4" markings was used to measure specimens to insure that items that would have been recovered in a 1/4" screen were eliminated. After sorting, tabulations were made of the items in each category. The separate categories were then placed in film cannisters, sealed, ard labeled. Results Column samples for water separation-microanalysis were examined fram five sites within the Warm Springs Dam Project Area. This material has not yet been analyzed; however, the following quantities of material were collected fram the samples: Site (CA-Son-) Lithics Bone Seeds excavated levels W-s unit w-s unit Ws only 582 (11 levels) 51 746 20 35 24 % 6% 36% 100% 571 (13 levels) 680 5225 490 433 10 % 12% 53% 100% 576 (20 levels) 82 843 302 672 4 % 9% 31% 100% 553 (10 levels) 75 408 250 663 -- 16% 27% 221 556 (22 levels) 181 1034 389 376 25 % 15% 51% 100% Totals 1069 8256 1451 2179 63 % 11% 40% 100% w-s (water separation) % (w-s/unit + ws totals) The quantity of seeds recovered through this program was disappointingly small; however, it should be emphasized that not a single seed was found in test unit excavations using 1/4" mesh screen. The low seed count may be partly attributed to destruction of unburnt seeds by reaction to local soil chemistry. Water separation procedures will be potentially valuable for collecting any remains that have been preserved in feature fills such as hearths. On the other hand, over a thousand pieces of lithic debris were recovered by water separation. This amounts to an average of 11% of the total lithic count from these five sites. In a future study these additional pieces will be extremely helpful in understanding the nature of the complex reworked tool industry at Warm Springs. Perhaps the most imnportant category of items recovered by the microanalysis program was the faunal material. Over 1,400 small pieces of animal bone were collected, both burnt and unburnt, resulting in 40% of the total amount of bone found at these five sites! The fact that these pieces are small does not diminish their importance. Normal unit recovery with 1/4" mesh screening is biased towards collecting large ma-unalian remains. Through a water separation program it is possible to also recover reptile, rodent, bird and fish remains which may well have constituted a major part of the animal food resources. A program of this type coupled with regular excavation procedures is necessary for any large-scale excavation of an archaeological site. Too much is lost through 1/4" mesh screening providing biased ard partial recovery of archaeologically significant remains. For this reason we suggest that a water separation-microanalysis program accompany any future excavation plans in the Warm Springs Dam Project Area. 223 APPENDIX C A TYPOLOGICAL STUDY OF THE REW)RKED BIFACES OF THE WAM SPRINS ARCHAEOLOGY PROJECT BY NANY WHITNEY Complete Bifaces For the purpose of this study of the reworked bifaces of the lithic industry at Warm Springs, all the projectile points, spear points, knives, blaes aid other symmetrical bifacially worked tools are collectively referred to as "bifaces," regardless of function. It is clear that all categories mentioned above were occasionally reworked; therefore, all bifacially worked tools are considered potential core pieces. A tool was counted original dimensions were considered to be whole missing. and measured as a complete preserved on the piece. even if a small section of biface if the maximum Thus, a tool may be the original piece is Thirty-four canplete bifaces were found on the test excavation units during Phases I Archaeology Project. These canplete bifaces may six distinct shapes: the surface or collected fran and II of the Warm Springs be divided typologically into foliates straight based concave based side-notched corner-notched barbed 4-o* 1-o 9-c 2-c 6-o 5-c 1-o 3-c 3-c 12-o 22-c *o-obsidian c-chert Note that 12 pieces were made of obsidian and 22 fran of the total number were chert. It is interesting five of the 89 reworked pieces were manufactured fran were clearly preferred for purposes of reworking. chert, or well over half in this regard that only chert. Obsidian bifaces Inccnplete Bifaces This category includes all fragments of projectile points, spear points, blades, knives arnd other bifaces which cannot be shown to have been reworked. The terms "base" proximal erd missing. arK] "tip" were used for pieces with the distal or Therefore, a base fragment may include either a small, 13 total 1 total 2 total 11 total 4 total 3 total 34 total l 224 canplete base or the majority of a piece which is missing only its tip. A midsection is missing both erds. There were 146 biface fragments recovered fran surface survey reconnaissance or test excavations. These incanplete bifaces fall into three categories: tips 30-o 18-c 48 total bases 43-o 39-c 82 total midsections 10-o 6-c 16 total 83-o 63-c 146 total Eighty-three (57%) of the total number of fragments were manufactured of obsidian while 63 (43%) chert pieces were recovered. It is important to recall that only 35% of the cauplete bifaces were obsidian and to note that 94% of the reworked bifaces were made frcm obsidian. It appears that most obsidian bifaces were fragmented ard appear either as unworked fragments or as reworked pieces. Fifty-six percent of the fragments occur in the form of obsidian ard chert bases. Many of these pieces had been broken at the thickest part of the body. This suggested that the breakage may not have been accidental. All incaoplete bifaces were then examined to see if there was clear evidence of intentional breakage. It was postulated that if the fragment had a small projection or hinge on the break, then the piece was probably held against a hard edge arn snapped off. In fact, 120 of the 146 biface fragments, or 82%, contained this hinge. In the other 26 cases the hinge was either not present or, in the case of some of the coarser cherts, could not be distinguished. In addition, it was found that 76 fragments, or over half (52%), of the examples had a punch mark directly opposite the hinge on the other surface. It seens clear that not only were most of the bifaces intentionally snapped, but also they were truncated by holding a punch in the center of one surface of the piece. When sub-divided into the fragment categories, we see that 58% of the total number of biface fragments that can be shown to have been intentionally truncated are bases. For obvious reasons, one would expect to find may tip fragments but relatively few truncated bases, assuming that the breakage was accidental. Hinge plus nch mark Hinge only Total tips 20-o 2-c 8-o 7-c 37 bases 25-o 17-c 9-o 19-c 70 midsections 10-o 2-c 1-c 13 55-o 21-c 17-o 27-c 120 All of the figures suggest that the inhabitants of the Warm Springs area were intentionally truncating the bifaces by removing the bases in order to create a striking platform for reworking their tools. 225 Six distinct shapes of incanplete bifaces can be recognized fram the fragments recovered: foliates 37-o 19-c 56 total straight based 2-o 14-c 16 total concave based 1-c 1 total side-notched 5-o 5-c 10 total corner-notched 4-o 5-c 9 total barbed 2-o 1-c 3 total 50-o 45-c 95 total Note that over half (59%) of the types are leaf shaped arn] that most of these (67%) were of obsidian. Examination of foliates in terms of fragmented sections revealed that nearly 80% of the foliate fragments are bases and that 78% of these bases were made of obsidian: tips 4-o 3-c 7 total bases 30-o 14-c 44 total midsections, 3-o 2-c 5 total 37-o 19-c 56 total As more camponents of the Warm Springs lithic industry were examined, it became increasingly clear that the inhabitants of this area were intentionally truncating bifaces for the purpose of reworking them. In normal archaeological contexts biface fragments are indicative of projectile points, spear points, knives anx blades which had been broken through use or during manufacture and discarded. However, many of these fragments may be regarded as the debitage of this industry. The base sections were snapped off of the original piece to provide the widest striking platform possible for reworking the biface. Further, it seems that a particular kind of biface was preferred in this industry as well as a particular material; namely, obsidian foliates. Reworked Bifaces: Technique of Manufacture The eighty-nine reworked biface fragments recovered fram the surface and the test excavation units at Warm Springs were examined with regard to a particular hypothetical method of manufacture. It was found that by using this approach, all but eleven pieces could be categorized. Thus, 88% of the pieces can be interpreted within the framework of the system described below. Figure 1 is a schematic representation of a foliate. Each step of manufacture involved in producing the final reworked form is labelled next to or on the piece. 226 FIGURE 1 The first step in the manufacturing process involved removing either a base or tip fram the biface in order to produce a platform fram which the rest of the piece could be shaped. With the unworked bifaces short base fragments are numerous: bifaces were truncated at the basal end in order to obtain the widest section of the biface for striking platform. Next, on the thicker bifaces, a flake was removed fran the center of one or both surfaces of the biface in order to thin the piece. This flake was usually struck fran the platform towards the tip. Two of the actual "thinning" flakes were measured and examined. Many of these flakes remain in the debitage level bags, uncounted, since they were originally unrecognized. The third step was to remove an edge fran the side of the biface, which will be referred to as "truncating the edge longitudinally." As with the "thinning" flakes, this edge was removed by striking a blow fran the platform which had been formed by the initial transation. These longitudinally truncated edges will be referred to as "spalls." Fifteen spalls of this type have been recorded in the inventory records, most of these being fragments of the original longitudinal truncation. A final step, in the case of the largest bifaces, involved removing the remaining tip fram the piece, giving the tool a rectangular rather than a triangular shape. With the smaller bifaces, this step was apparently not necessary. It may be that the tips of the larger bifaces had to be removed in order to decrease the length of the finished tool. Completion of these steps yielded what must have been the desired product. A complete tool, then, can be defined as a fragmented foliate resulting fran one or more transverse truncations, having one original edge with the opposite edge truncated longitudinally. Fifteen canplete tools were recovered as well as fifteen tool fragments. It is significant that fifteen spalls (or spall fragments) were also found. The average dimensions of the fifteen complete tools were calculated; the average length is 3.0 an, the average width is 1.49 an, and the average thickness is .92 am. Thirteen of the fifteen tools have a "hinge" on the transverse truncation, and one-third of these thirteen have a pu.nch mark 227 opposite the hinge, indicating that these tools were intentionally truncated. The tools invariable exhibit secondary retouch. Both step and marginal retouch are consistently found on the remaining original biface edge and on the negative flake scar left by the longitudinal truncation. Furthermore, one-third of the pieces have step retouch on one or both sides of the transverse truncation. If sharp edges remain on the piece, either fran the removal of the "thinning" flake or fran the truncations, these were removed. All fifteen of the tools bore at least some evidence of this process. Also, the spurs left on the various truncation scars were removed in many instances. Naturally, the production of these tools created debitage. This debitage was often retouched and utilized, aix can be organized into nine separate types. Figure 2 is a schematic representation of a leaf-shaped biface with each type labelled next to or on the piece. 7 3 FIGURE 2 Type 1 comprises camplete bases that have been removed fram the biface and subsequently retouched. One of these has been found in the debitage. The piece has been heavily retouched on the transverse truncation. In several cases it seems that these bases were further truncated in order to form a number of snall triangles, which were subsequently retouched. Six of these were found in the reworked biface industry. These are labelled "step 2" on the base of the foliate in Figure 2. The third type of retouched debitage involves reworking the flakes removed fram the truncated biface as "thinning" flakes. Two of these flakes have been retouched subsequent to their removal. Type 4 includes midsections which have been thinned. Seven examples of this type of retouched debitage have been inventoried. Occasionally this piece is further truncated as shown on the foliate in Figure 2. This truncated piece, type 5, is then retouched and used as a tool in its own right. Six of these are known. All fifteen of the spalls resulting from the longitudinal truncation of the biface show some retouch. They are generally very small pieces. Despite 228 their diminutive size these spells seem to have been utilized as tools. They are labelled "type 6" on the foliate in Figure 2. Type 7 includes tips that have been longitudinally truncated or "spelled" before being removed fran the core piece. Subsequent to their removal, they are retouched on their truncated edges and re-used. Six retouched tips were included in the reworked biface industry. Of this total, five have been "spelled." The eighth type of retouched fragment comprises midsections that are either only retouched, or "spelled" and retouched. Of the five midsections in this type, two have been truncated longitudinally and three have been only retouched. All five pieces have been retouched subsequent to their remval fran the original biface. The final type of retouched debitage comprises bases that have been modified into end scrapers. There are two examples in this category of which one was formed on a base that had a "thinning" flake removed and the other on a base that was both "thinned" and "spelled." Both were formed on rounded bases that had been heavily retouched. Of the fifteen complete tools only one was manufactured fran chert. Four other chert reworked biface pieces were recovered; two are spalls and two belong to the category of "miscellaneous" unidentified pieces. Three of the five chert pieces have been burnt including the complete tool aid two spells. The small number of chert reworked bifaces indicates that obsidian was preferred as the raw material in this industry. This preference may be due to the fact that obsidian is substantially easier to work than chert. Reworked Bifaces: Selection of Complete Bifaces to be Used as Cores The complete and incomplete unworked biface fragments suggested that the reworked biface industry utilized mostly obsidian foliates as core pieces. The measurements of the tools fram the reworked biface process were compared to the measurements of the various camplete biface shapes to determine which type was most caommnly reworked. The following is a list of the average lengths, widths and thicknesses for the six biface types as well as for complete reworked biface tools. length width thickness foliates straight based concave based corner-notched side-notched barbed reworked tools Only obsidian reworked biface measurements are in 3.7 an 5.5 2.60 1.90 (3.0) an 1.89 an .73 an 41 1.93 .60 3 _ - 0 1.65 .40 4 1.35 .41 11 1.55 .35 2 (1.49) an (.92) an 15 pieces pieces pieces pieces pieces pieces pieces pieces were considered in these averages since 94% of the industry consists of obsidian fragments. Incamplete parentheses. 229 Transverse, longitudinal and facial truncations made it impossible to record original measurements of the reworked pices. Of the 14 obsidian reworked biface tools, six were not thinned and retain conplete thicknesses. The average thickness of the unthinned pices is .96 an. As this is an incanplete thickness, it becanes significant that this measurenent is thicker than that of the camplete pieces. Comparing the reworked tools with the original biface shapes, it is clear that the average thickness of the reworked biface tools is much larger than any of the original bifaces, caning close only to foliates. The width measurenents are incanplete on the reworked biface tools because of the longitudinal truncation. Note, however, that even with this incomplete width these tools are wider than side-notched pieces and came close to the width of the barbed type. If one adds the average width of the fifteen spalls recovered to the average width of the fifteen reworked biface tools, it seems likely that an approximate width for the original piece can be attained. The average width of the tools is 1.49 an plus 1.00 an for the spells, or 2.49 an for the average width of the original bifaces before reworking. This width is, again, larger than any of the complete or incomplete unworked biface types and only approximates straight based and foliate pieces. The length is incomplete because of the one or more transverse truncations of the original piece. Even with this truncated length, the reworked tools are longer than either the corner-notched or side-notched complete bifaces. The length closely approximates that of the canplete foliate average length but is much lower than the straight based length. Only three straight based pieces were made of obsidian and, therefore, considered in this canparison. Thus our sample for this category is snall. All of these figures indicate that the reworked pieces are closest in shape (length, width, and thickness) to obsidian foliates or, possibly, straight based pieces. Assuming that the thicker the piece is, the larger the piece, then, the largest obsidian bifaces were being reworked. In fact, it would seem that the size that was preferred is so large that it is not represented among our complete biface pieces, foliate or straight based. The largest bifaces were reworked so caomonly that few fragments of the originals are left. The chart on the next page (Figure 4) plots average thickness to average lengths of the complete and incanplete unworked obsidian obsidian foliates. In this way the shape of these pieces is reflected in a proportion of thickness to length. kten the average thickness and length of the reworked tools is placed on this graph (shaded area), it becomes clear that only a very few of the canplete or in complete unworked pieces cane close to the original measurements of our bifaces. Obviously, the largest obsidian bifaces were reworked in this industry. These pieces measured over 3.0 an in length and were more than .90 an thick. This reworked biface industry is known fran two other locales in the North Coast Ranges: fran the Clear Lake region as found at Lak-261 230 (Fredrickson Lake County. relationship they have not 1973), and fran Little Indian Valley (Orlins 1971) is eastern However, at this time it is not possible to understand the between the Warm Springs industry arn the other industries since yet been adequately described. 231 EXPLANATION CF ILUBSTRATIONS ? = reconstructed dimension () = actual but partial dimension Specimen Number Material 1. w. th. wt. Fig. 22a 70-304 b 70-313 c 70-320 d 70-365 e 70-562 f 70-578 9 70-615 h 70-613 i 70-626 j 70-643 k 70-647 Fig. 23a 70-539 b 70-551 c 70-661 d 70-662 e 70-675 f 70-676 g 70-723 h 70-736 i 70-745 obsidian obsidian red chert red & black obsidian obsidian brown chert obsidian red chert obsidian lt. grey chert obsidian obsidian green chert green chert green chert obsidian grey chert brown chert obsidian green chert - 1.7 1.9? 0.9 2.0 0.8 - 1.6 0.2 0.5 0.3 0.8 2.6 1.5 0.6 - 1.6 0.9 - - 0.7 4.5 2.3 0.8 - 2.0 0.6 3.8 - 0.8 - 2.3 1.6 1.1 2.6 1.9 3.2 1.7 2.1 1.5 2.3 1.3 - 1.2 - 1.9 5.7 2.5 6.0 1.9 0.8 0.3 0.3 0.6 0.5 0.3 0.4 0.7 0.8 0.7 (1.4) 3.0 1.5 (2.1) 2.6 (7.0) Excelsior Gunther Excelsior Excelsior Cascade? 9.0 early side-notched (2.0) Excelsior - Excelsior 0.5 late side-notched 3.1 1.1 late side-notched 0.6 late side-notched - late side-notched (5.3) early side-notched 12.1 5.1 early side-notched Type -- - & AD - ( Fig. 24 Specimen Number la 70-55 b 70-68 c 70-259 d 70-462 e 70-391 f 70-403 g 70-1547 h 70-163 i 70-233 j 70-1349 Fig. 25a 70-83 b 70-85 c 70-6 d 70-120 e 70-125 f 70-1527 g 70-171 h 70-212 i 70-218 j 70-252 Fig. 26a 70-5 b 70-1535 c 70-1051 d 70-1235 e 70-1236 Material obsidian obsidian obsidian obsidian obsidian obsidian grey-green chert brown chert quartzite grey chert tan chert obsidian tan chert green chert obsidian obsidian green chert obsidian obsidian obsidian obsidian reddish chert obsidian tan chert obsidian 1. w. 4.1 1.7 3.5? 2.3 - 1.4 2.6? 1.2 3.6? 2.0 1.5? 1.4 2.6? 2.0? 2.3? 1.2 - 2.2 - 2.0 - 2.4 - 1.5? 3.7 1.7 2.3 1.6 1.6 1.4 2.0 1.6 2.3 1.3 - 1.7 4.3 2.3 - 2.3 2.7 1.5 - 1.9 2.7 2.0 - 1.5 2.0 1.0 th. 0.9 0.8 0.3 0.8 0.9 0.4 0.3 0.5 0.8 0.4 0.7 0.3 0.6 0.3 0.3 0.3 0.5 0.8 0.8 0.9 0.4 0.7 0.6 0.5 0.3 wt. 7.8 (4.6) (0.2) (4.6) (4.8) (0.7) (0.6) (2.5) (3.0) (1.4) (5.0) (0.3) 3.4 0.8 0.3 0.4 1.1 (4.4) 6.8 (8.6) 0.7 (3.1) 2.9 (0.8) (0.3) Excelsior Excelsior Excelsior Excelsior Gunther Gunther Excelsior Gunther (in gravels) Gunther Gunther late side-notched Excelsior Excelsior Gunther (?) early side-notched early side-notched late side-notched 232 . l Sp-ecimen Number Material 1. w. th. wt. 70-1538 70-10 70-1053 70-1109 70-1145 70-12 70-1449 70-1491 70-1508 70-1515 70-1521 70-2* 70-1* obsidian obsidian obsidian obsidian gr., striped chert obsidian red chert reddish chert obsidian greenish brn. chert obsidian red chert red chert i 70-3* red & green chert Ia 70-1212 obsidian b 70-1167 brn. chert c 70-1174 red & green chert d 70-1182 reddish chert e 70-8 white chalcedony f 70-950 obsidian g 70-962 reddish brn. chert 2.5? 2.5? 2.9 2.1? 2.8? 2.2? 1.8 2.4? 3.3? 4.5 2.5 2.5 4.1 3.7 1.6 2.0 3.2 1.6? 1.8 1.6 1.3 2.0 1.6 1.2 1.9 0.4 0.4 0.3 0.4 0.5 0.4 0.5 0.8 (1.0) (1.0) 1.3 (0.7) (2.0) (0.8) 1.8 (3.6) 1.9 0.6 (2.5) 2.3 0.8 7.4 1.8 0.7 late side-notched late side-notched late side-notched late side-notched late side-notched Excelsior Excelsior Excelsior 3.0 1.7 0.5 2.6 Excelsior 1.7 0.9 5.6 early concave base 2.1 0.7 5.0 early side-notched 1.4 1.3 1.9 0.3 0.4 0.7 0.6 0.7 3.5 late side-notched late side-notched Excelsior 4.4? 1.8 1.0 (8.3) Cascade? 2.4? 1.7 0.3 (0.6) 2.8 1.7 1.5 1.4 0.3 0.3 1.3 0.4 233 Tvpe Fig. 26f g h i Fig. 27a b c d e f g h Fig. 2E 234 Specimen Number Material h 70-963 gr. & tan chert i 70-968 reddish brn. chert Fig. 28j 70-973 obsidian k 70-977 basalt or Fig. 29a 70-919 gr. chert 1. w. th. wt. Type - 2.3 0.9 (7.2) early leaf-shaped - 2.1 0.7 (5.0) early concave base 2.6? 1.8 0.9 (4.5) Excelsior - - - (0.2) Desert Side-notched? - 2.0 0.4 (1.2) b 70-896 green trans- 2.3? lucent chert 1.7 0.3 (0.6) Gunther related c 70-904 obsidian d 70-905 gr. chert 2.0 1.4 0.3 (0.4) same 2.5? 2.0 0.4 (1.3) e 70-912 green trans- 3.2 lucent chert 1.7 0.4 0.9 Gunther related f 70-830 obsidian 2.8? g 70-855 tan chert 3.9? Fig. 30A 70-277 Near site Son-584. 1.9 0.4 (1.3) late corner-notched 2.2 0.8 (5.5) Excelsior In gravel creek bank north of the confluence of Yorty Creek ard a small tributary. This is a large keeled piece of brown chert. It has secordary flaking on its edges. Wt. 51 gr. 30B 70-274 In gravels below and Large, thick scraper Wt, 18 gr. 5 m SE of Unit 1 of site Son-576. of pink chert with yellow band. 30C 70-181 30D 70-931 30E 70-179 Site Son-577, 90-110 an depth. Green chert. The nibbling siown at the top of the piece appears to be battering to make a platform rather than retouching. Water rolled. Wt. 21 gr. Site Son-585, in gravels. Reddish chert. Looks like handaxe. Wt. 79 gr. Site Son-577, 80-90 an depth. Dark grey chert. Chipping at top and top left is recent and probably not man made. Chipping at bottom aid on right edge is man made and has cortex formed over it--same cortex over entire piece except areas at top and top left. Wt. 18 gr. 30F 70-280 Near Son-584. Same location as Fig. 30A. material but with white barding. Another 80 gr. 30G 70-188 Also same handaxe. Wt. Son-577, 110-120 an depth. Green chert. A flake stuck fran top ard then trimmed with most of outside cortex left. Wt. 38 gr. *$pecimens shown in Fig. 27g, h, and i are clearly fran material quarried at nearby Son-554. 235 236 WARM SPRINGS PROJECTILE POINTS NORTH SECTION - WESTERN GROUP CA-Son- 568 20 -30 cm 40- 50 cm 60-70cm 90-100cm CA-Son-571 10-20cm 30- 40 cm 60-70cm CA-Son- 571 60-70cm 70-80cm 90-100cm 100-110cm Figure 22. (Actual Size). 237 WARM SPRINGS PROJECTILE POINTS NORTH SECTION - WESTERN GROUP a 0-10cm d b CA- Son- 575 30 - 40 cm i I e CA-Son- 572 0-10cm 0 f 0-10cm h CA-Son-572 110 - 120 cm 120 - 130 cm Figure 23. (Actual Size). g 80-90cm 238 WARM SPRINGS PROJECTILE POINTS NORTH SECTION-EASTERN AND CHERRY CREEK GROUPS CA- Son- 582 20- 30 cm 50-60cm CA- Son - 593 30-40 cm 50- 60 cm CA- Son- 592 Surface CA- Son- 597 30- 40 cm CA- Son - 597 40-50 cm HX ji CA-Son- 601 20-30 cm Figure 24. (Actual Size). c 0- 10 cm Surface f It 239 WARM SPRINGS PROJECTILE POINTS NORTH SECTION-YORTY CREEK GROUP a CA- Son-583 0-10 cm 10 20cm d 20- 30 cm h 10-20cm 60-70 cm CA- Son 584 30- 40 cm i s~~~~~~~~~~~~~~~~~ CA- Son- 576 I 70-80 cm 0-30 cm 100 - 110 cm Figure 25. (Actual Size). 240 WARM SPRINGS PROJECTILE POINTS SOUTH SECTION-CENTRAL AND LOWER WARM SPRINGS GROUPS ha kab CA-Son-542 Surface 0-10 cm Tc Surface I .I Id CA-Son- 600 0- 10 cm 20- 30 cm . f CA- Son - 553 0- 10 cm CA-Son-553 0-10 cm 70-80 cm Figure 26. (Actual Size). Surface 100 cm 241 WARM SPRINGS PROJECTILE POINTS SOUTH SECTION- LOWER WARM SPRINGS GROUP Surface 160 170 cm IW b CA- Son - 556 60-70 cm e CA- Son - 556 170 180 cm 120 -130 cm 190-200 cm CA- Son- 559 Surface Figure 27. (Actual Size). Surface Surface i 242 WARM SPRINGS PROJECTILE POINTS SOUTH SECTION - UPPER WARM SPRINGS GROUP CA-Son- 544 40-50cm 50-60cm Af CA-Son-547 10- 20 cm 60-70cm 20-30cm CA-Son-547 30- 40 cm 40-50cm Figure 28. (Actual Size). Surface 0-10cm 20-30 cm k 50- 60cm 243 WARM SPRINGS PROJECTILE POINTS SOUTH SECTION - UPPER WARM SPRINGS GROUP M~~~a CA- Son- 545 0-10 cm CA-Son-550 10-20cm 10-20cm 20 cm- Bedrock CA-Son- 551 40-50 cm Figure 29. (Actual Size). 0-10cm Surface A F XB 0B ACT~~ tE E cm Figure 30. Artifacts fran Yorty Creek gravels. 244 CONTRIBUTIONS OF THE UNIVERSITY OF CALIFORNIA ARCHAEOLOGICAL RESEARCH FACILITY Number 40 October 1979 AN ARCHAEOLOGICAL ASSAY ON DRY CREEK, SONOMA COUNTY, CALIFORNIA UNIVERSITY OF CALIFORNIA Department of Anthropology Berkeley * CONTRIBUTIONS OF THE UNIVERSITY OF CALIFORNIA ARCHAEOLOGICAL RESEARCH FACILITY Number 40 October 1979 AN ARCHAEOLOGICAL ASSAY ON DRY CREEK, SONOMA COUNTY, CALIFORNIA Martin A. Baumhoff and Robert I. Orlins UNIVERSITY OF CALIFORNIA Department of Anthropology Berkeley TABLE OF ODNTENTS Preface * .................... ............................. viii Introduction ................ * * * * * * * * ......... 1 Geography ................. ....... ......... . ..........**** 1 Archaeology ......***************************************** 2 Site Description ....**@*************9********eee................ 9 Western Group .......... .9 Western Group Chronology ......... *. ...*. . . ............... .. 38 rEaK-stern Group . . . . .............................. 43 Eastern Group Chronology .... .............. 43 Yorty Creek Group .. *..... ... .......................... 79 Yorty Creek Group Chronology .......101 Cherry Creek Group ......................................... 105 Cherry Creek Group Chronology ............................... 112 Central Grout)................................ 115 Central Group Chronology .. *0 * * *** ..................... ... 122 Lower Warm Springs Group . . . . .. . . . .. . . . . .. . . . . . 125 Lower Warm Springs Chronology ............................... 150 Upper Warm Springs Group ... ..................... .. ... ....... 157 Upper Warm Springs Chronology .... ............ ............... 181 Summary and Discussion Chronology and Debitage . .......... ................. . .. 185 Territory and Population ..... .......................... 191 Site Specialization sees............................. 195 Bibliography ................................. 211 I Appendix A. Apperndix B. Appendix C. Explanation Possible PreBorax Lake Ccmponents Auxiliary Techniques A Typological Study of the Reworked Bifaces of the Warm Springs Archaeological Project. by Nancy Whitney ....................... of Illustrations ..................................... It 215 219 223 231 LIST OF ILUUSTRATIONS Figures 1. North Coast Rarne Chronology ................................ 2. Legered for Site Maps 3. Son-567 Profile 4. Son-571 Profile 5. Son-568 Profile 6. Son-593, Unit 1 7. Son-593, Unit 2 8. Son-582 Profile 9. 10. 11. 12. 13. 14. 15. 16. 17. 180 19. 20. Son-608 Son-597 Son-576 Son-577 Son-584 Son-583 Son-603 Son-601 Son-556 Son-553 Son-544 Son-547 Profile Profile Profile Profile Prof ile Profile Profile Profile Profile Prof ile Profile Profile 0000*@.. 0@000 0000.6 ooeooee e~g.... 0 00 @ 060 00 0a Profile Pr.o.0 le06. . . a .0..0 0 0 a0........ 000 00 0 0.a0.......0.0. * .......... ....a.. 0 00 00 0 060 00 0 0 @6 *0 0 a0 00 0 0 00 6w * a .... ... a 6 0* .....................a......a a0 00 00 0a O 6 @6@ @0 00 060 0 ~00 0 0 0 a0 0 0 0 0.0 21. Son-551 Profile ..........................................060 22. Projectile Points, Son-568, 571 23. Projectile Points, Son-572, 575............................ iii 3 5 14 28 34 46 47 57 65 69 82 88 93 98 109 111 131 143 160 167 178 236 237 24. Projectile Points, Son-582, 592, 593, 597, 601 .............. 238 25. Projectile Points, Son-576, 583, 584 239 26. Projectile Points, Son-542, 553, 600 * 240 27. Projectile Points, Son-556, 559 *****,*************..........* 241 28. Projectile Points, Sonr-544, 547 ....... 242 29. Projectile Points, Sonr-545, 550, 551 .243 30. Artifacts from Yorty Creek Gravels 244 iv Maps 1. Archaeological Site Groups .................................. 1 2. Western Site Group 10 3. Son-567 . . .. . .a a@***@ 0 60a000a0@. *0@*@6 *** ......... 13 4. Son-572 ***,**,****a..... * ***.*.....a. .a....................... 20 5. Son-571, Second Terrace 25 6. Son-571, First Terrace 26 7. Son-568 . 32 8. Eastern Site Group . *0***e**** .. .................... ......... 42 91. Son-593 ...........*.......................S....... 0 0 .0@ @06 44 10. Son-582 ..... .................. 56 11.4 Son-6 08 e e e. e. . ein. e a. .e . a. e. .e . .ee e. e e . .e . e 63 12. o-9 .................................. 68 13. Cherry and Yorty Creek Site Groups .......................... 77 15. Son-577 . 87 16. Son-584 ........................... ..................... 92 17. Son-583 97 18. Son-601 . ...... 0.................................... 110 19. Central Site Group West ....................***............ 113 20. Central Site Group East .............. . ........ 114 21. Sonm-600 .... ........ 120 22. Lower Warm Springs Site Group 123 23. So~n-556 130 24. Sor- 553 142 25. Son-554 147 v 26. Upper Warm Springs Site Group ............................. 155 27. Son-544 . ............ 159 28. Sont547 166 29. Son--551 177 30. Settlement Pattern Chronology .,, 204 vi PREFACE In Spring, 1974 the Corps of Engineers approached us about doing a very small archaeological reconnaisance in the Warm Springs Dam project. The situation had arisen in the following way. The Warm Springs Dam had been planned for many years by the Corps as a flood control project related to the annual or least biennial floods on the Russian River (the dam in on a tributary). In 1964 in accordance with then current practice the Corps of Engineers, through the National Park Service, contracted with A. E. Treganza of San Francisco State College for an archaeological survey. Not much time or money was allocated to this survey so that it was quite minimal - only a few days were devoted to it and only in areas adjacent to the road. A total of eleven sites was located arn none was recammerKded for excavation. In the years that followed a considerable national outcry on the subject of environmental degradation developed which resulted in several laws and executive orders relating, among other things, to preservation of archaeological resources, a subject long ignored by all levels of government ard also by private enterprise. Locally the interest in environmental preservation manifested itself in concern for the damage that would result fran building Warm Springs Dam. At that moment, in Spring, 1974, the particular concern was for archaeological site Sor-600 which was thought by some to be the ethnographic village of Tako-ton, ard therefore of interest to Pamo Indians now living on Dry Creek. Col. James Lammie of the Corps of Engineers, desiring an evaluation of this situation, asked us to make a determination. A minor amount of excavation there suggested that whatever the ethnographic identity of the site might be, farming operations had so badly disturbed it, that its archaeological potential had virtually disappeared. This was reported to the Corps of Engineers. In the meantime it had been determined that the original survey by Treganza was inadequate according to the starndards mandated by recent changes in Federal rules. The Corps of Engineers therefore provided the money for a complete survey, and a contract by the National Park Service with the University of California was entered into ard work was begun October 24, 1974 with a crew of 6 under the field direction of Orlins. The survey continued for two months, arn luckily there was fine weather during most of that period. A total of 55 archaeological sites was recorded including 10 which had been found by Treganza in the original survey (one of the original 11 turned out not to be a site). The survey covered 85% of the project area (the other 15% was done later), and we now consider that although a few small sites may have been missed that virtually every surface site is now known. There may of course be subsurface archaeology that has not come to light particularly along Yorty Creek as discussed in the description of that group. Having completed the survey, negotiations were begun in February, 1975 concerning the next phase of the project-testing and evaluation. Field work was begun in late May, 1975 with Orlins again field director and a crew of 10. Test excavations were performed into mid-July when the field crew dispersed, viii and three persons returned to Davis to assist in completing the report. After excavations had beeen under way a short time, it became apparent that insufficient time had been allocated to preparation of the final report. This was therefore delayed, and as a consequence the Corps of Engineers became increasingly uneasy. They were required to appear in court, and while it would have been possible to obtain a postponement, the project had already been delayed more than a year and inflationary factors, particularly at that time, were increasing to the cost of the dam considerably. It is possible to argue that they should have begun thorough archaeological evaluation much earlier, and in fact it has been so argued, but on the whole we fiind it difficult to place blame in view of the fluctuating nature of the legal and regulatory requirements. In general we found the San Francisco Office of the Corps of Engineers very reasonable to deal with and not at all obstructive. In any case because of the delay and because of internal confusion within the National Park Park Service, the contract between the University and the Service was terminated in February, 1976. At that time our draft report was in their hands, and they performed an act which they call "assembling" which so far as we can see consists of retyping, reproducing, and omitting certain portions. The Park Service gives the following statement "...any errors of omission or content are to be attributed to the contractor." These authors refuse to accept such responsibility since they did not even see the final copy until after it was issued. It is instructive at this point to outline some of the difficulties faced in dealing with interlocking bureaucracies. About a half dozen different bureaus, depending on how you count them, were involved in this effort. The U.S. Army Corps of Engineers primary charge, as we understand it, is with flood control and inland waterways. In addition they are required to observe environmental regulations imposed by Congress and/or executive order. In our experience there are three ways they can do this in the case of archaeology: by hiring their own archaeologists; by contracting other individuals or organizations; or by working through Inter-Agency Archaeological Services, an agency of the National Park Service. Inter-Agency Archaeological Services have had a longish history but a rather uneven one because the level of professionalism has been uneven. As is the case, we suppose, with most governmental organizations the persons responsible have been better bureau- crats than professionals (archaeologists). Inter-Agency Archeological Services do not, or at least did not in this case, undertake the archaeological investigations themselves but contracted them to other organizations retaining for themselves more-or-less the position of overseer. In the present case the contractor was the University of California and this is another level of bureaucracy we will deal with below. But first let us deal with a heirarchy tangential to this Federal maze. There exists in Washington something called the National Register of Historic Places. Nominally the function of this is to identify places important or unique historically or archaeologically which can then be preserved or otherwise dealt with appropriately. Perhaps more important is the provision that on Federal land such places must be dealt with appropriately. The office that deals with this seems also to be related to the Inter-Agency Archeological Services but only at a higher level than we dealt with on the project. Also ix tconneted with this tangent is the Historic Preservation Office of the State of California. As we understand it the naninations for the National Register of Historic Places are (when Californian) first assessed by this office. Another function of the State organization is to pass judgement on all enviromental impact reports even if they refer to Federal lard. The legality -of this seems questionable to us, but in any case it provides another level of interference. Finally the contractor was the University of California, and it is in itself a formidable bureaucracy. One has innumerable difficulties here carrying out archaeological obligations, difficulties in the form of approval of contracts, accounting procedures, purchasing procedures, use of University vehicles aid the like. These, however, are ones we deal with every day as membrs of the University. In the case of this project we came up against a particularly formidable obstacle in University hiring policy. There are two aspects of this, one having to do with racial aid other kinds of illegal discrimination and the other relating to prior rights of University employees or students to University jobs (however funded). The first part hinges on what is called an affirmative action program undertaken in response to action by the Department of Health, Education aid Welfare under anti-discriminatory legislation. The effect of the program is that in addition to eliminating discrimination it eliminates speed. This was most unfortunate in the present case since speed was all-important to the Corps of Engineers. It seemed most unfair to us that, frcm what we could observe, both the Corps of Engineers aid the Inter-Agency Archeological Services could hire virtually anyone they wanted within days while the University, because of the action of yet another Federal agency, was hamstrung in the same activity. The University's own hiring policies also slowed us up at a time when speed was of the utmost importance. The result of this concatination of agencies, Corps, Offices, Registers, Departments, and Campuses as well as laws, orders, policies, rules and programs, was to make it seem at times impossible to do archaeology at all. One spends full time dealing with these problems and no time doing archaeology or any other kind of scholarship. The difficulties of the situation are twofold. In the first place it is in the nature of bureaucracies to deal in abstract requirements and policies usually not related to the particular problem at hand--thus the well-known inflexibility of these institutions. But even if this were not the case, the objectives of these organizations is at best different fran and at worst contradictory to the objectives of scholarship or science, namely to discover the nature of the world. Thus the Corps of Engineers is interested in flood control, not in how Paom Culture came to be the way it was. The National Register is interested in registering things. Even the Inter-Agency Archeological Services, which one might imagine had some basic interest in the problem, seems more interested in purely formal requirements, whether or not they have anything to do with the particular issue. We do not know how to deal with the difficulties mentioned here or even whether it is particularly useful to mention them. They are outlined here as a matter of personal and ethnographic interest rather than as a proposal for organizational reform or anything of the sort. x What follows is the result of the archaeological reconnaissance in the Warm Springs area. It is basically a descriptive report, necessary because so little descriptive archaeology for the North Coast Range has appeared in pr int. Acknowledgements A great many individuals contributed time and efforts to this project. The official field crew was the following: Peter Banks (Assistant Field Supervisor), David Devey, Joan Hellen (also photographer and artist), Elizabeth Honeysett, (also ecologist), Willian Lindenau, Donald McGeachy, Jane Melville, Scott Patterson, Pamela Roberts, Miriam Stafford, Sonia Tamex, Nancy Whitney (laboratory supervisor), and Terry Zontek. Terry Zontek and Jane Melville also did floatation studies. Nancy Whitney and Joan Hellen were most helpful in canpiling material for the report. Roderick MacDonald and Elizabeth Honeysett prepared an ecological survey for the project. Melva Orlins served as project secretary. A good deal of time was contributed gratis by the following: Sari Fredrickson, Jonathan Borah, Thomas Clancy, Roxie Towe, Linda Bell, Randy Mason, and Andrew Devey. We think it fair to single out Sari Fredrickson and Linda Bell as having been especially generous with their time and effort. Sonana State College generally, David Fredrickson and David Peri in particular, made generous contributions to the project. Our relations with the U.S. Goverrment were not always friendly during this project so it is worthwhile noting that we felt and still feel that friendly and willing cooperation was given us by Mr. Garland Gordon of the Park Service and Dr. Richard Lerner of the Corps of Engineers. We are grateful to them both. The Department of Anthropology of the University of California, Davis was most supportive throughout. We wish particularly to thank James West, D. L. True, Jeanie Anderson, Barbara Beckman, and Gayle Bacon. xi INTRODUCTION Geography The area under consideration is in California's North Coast Range in Northern Sonama County about 20 miles southwest of Lakeport on Clear Lake. It is in gently rolling to low mountainous country with elevation fram about 400 to over 1000 ft. elevation. Mean annual rainfall here is over 50 inches and is quite variable, ranging fran below 30 to over 100 inches per year. The bulk of the rainfall falls between the first of November and the last of March. The temperature is indicated by data fran nearby Healdsburg wher8 the mean annual temperature is 590 F, ranging fran 400 in January to 70 in July and August. Overall the topographic and climatic conditions can be described as salubrious and pleasant. It has always been our contention that the most inmortant aspect of the environment for a hunting-gathering people like the Pomo is the native flora. This is true not only because they relied directly upon plants for their resources but also because the nor-aquatic fauna upon which they also relied is itself intimately related to floral patterning. Fortunately this matter has been taken up in some detail in an ecological survey of the project area (MacDonald and Honeysett 1975). This report was prepared for the present project and is available for consultation. We rely upon that report in the following summary. The vegetation is intimately related to soils and the above report relies upon that to divide the areas into four soil-vegetation types as follows: 1. Mixed Evergreen and oak forest soils. "These important timber soils supported a mixture of Doug-fir, Redwood, and hardwoods." This constitutes about 19,000 acres or 36% of the project area. The soils are in the Hugo, Josephine, Hugo-Josephine and Sites series. On soils of this category there seem to have been a considerable amount of redwood and Douglas fir growth and may therefore have supported some elk population. This is on the edge of the Roosevelt elk distribution (McCullough 1971:10) and this would have been the kind of country they would have preferred. The hardwoods included black oak, tan oak, and coast live oak, in that order. The first two figure importantly in the production of acorns for human use. Overall one would say that while this land was very important to the aboriginal inhabitants it was slightly less so than the next category. 2. Woodland - Grassland soils. Three soils also comprise about 36% of the total acerage. This actually seems to us to be two separate groups. The Laughlin and Suther series both have large areas of open grass (more than 50%) but also dense woodlands significant components of which are Oregon oak and black oak, both important acorn producers. The second sub-group consists of the Yorkville, Montara, and Sobrante series on which there are widely scattered small patches or individual trees. It would appear that no economically important tree crop occurs in the latter sub-groups unless it is the digger pine. Digger pines produce good crops of pine nuts but we have never heard that these were in any way important in the North Coast Range. Possibly there occur here also examples of lone standing oaks of relevant species (particularly valley oak). These might be important because such 2 trees far outproduce, tree for tree, individual trees found in groves. We feel that the Group 2 soils have the most important nut production to be found in the area. In addition it is the most important feeding ground for deer. This therefore may be the most important land from the standpoint of aboriginal econanics in the Warm Springs Project area. 3. Woodland-Chapparal Soil. This is the Los Gatos soils series canprising about 12,000 acres or 22% of the project area. We quote fran MacDonald and Honeysett (1975:24). "80% Brush and hardwoods with sparse grass. Shallower soils support a diverse 'high' chaparral mixture. Small redwood clumps occur in ravines. A grass-woodland develops on the deepest phases." The oaks that occur here are mostly coast aid interior live oak, both inferior in acorn production. The primary importance of this land must have been in producing deer feed. It was probably less important than either of the two preceeding. 4. Chaparral Soils. There are a little over 3,000 acres of these or 6% of the total and the comprise the Stonyford, Maymen, and Hennebe series. These dense chaparral areas were only important on their margins where important deer feed was found. They were probably the least important in the area. The distribution of these soil-vegetation types is shown on maps with archaeological sites plotted on them accanpanying each site groups below. The significance, if any, of the relationships shown there will be considered as the sites are described and discussed. Besides vegetation it is important to consider location and abundance of game animals and fish. Deer and elk have both been discussed in connection with vegetation zones above. The other important question relates to the availability in aboriginal times of game fish in Dry Creek and Warm Springs Creek. There seems no doubt that there was a salmon run in Dry Creek in aboriginal times, at least we have been told this by Fish and Game people, but it seems very unlikely that it was a large one. Apparently silver salmon and steelhead trout both run up the Russian River (Baumhoff 1963:174) and a small part of these went up this tributary. In addition there must have been trout and other native freshwater fish. The fishery then may have been a significant element of native subsistance but probably not a major or definitive element. These geographic factors in aboriginal life will be considered in the discussion of archaeological sites. Archaeology It is important that in addition to providing descriptive material we also outline our ideas of the broad outline of North Coast Range archaeology. For present purposes we present a cultural sequence which relies basically on that of Fredrickson (1973, 1974) but also incorporates additional typological features fran the Borax Lake site (Harrington 1948; Meighan and Haynes 1968, 1970) and fran the Willits site (Meighan 1955). Fredrickson's sequence gives a total of five periods for the chronology of the region summarized as follows: Borax Lake Pattern Borax Lake Aspect Early L CM 0 0 0 Late arm-I- Il 3 cD 0 0 (I 0 0 Berkeley Pattern Houx Aspect Early Late 0 2 ax 0 North Coast Range Chronology. 3 m a. Post Pattern a) 0 0 0o 4- 5P 0 0 01 0 0 0 Figure 1. i i i i d I *-wQ l OD l 14 I -E D 1 -_*4 I I I I I 4 -& - I 4 Post Pattern. This is tiought to date fran 10,000 to 6,000 BC. This complex is positively identified by Fredrickson only at the Borax Lake site. It has fluted points and crescent stones illustrated by Harrington (1948) ard by Meighan and Haynes (1968). Only one canponent fran Warm Springs i proposed as being fran this period but there may be others undiscovered simply because we have not excavated sufficiently the deeper portions of some sites. Early Borax Lake Pattern. This follows a possible haitus lasting fran 6,000 to 5,000 BC. and persists until 3,000 BC. It is characterized according to Fredrickson, by the large, square stemped Borax Lake point and by the use of milling stones. We have found none of the square stemmed points but we believe the large side-notch points and large concave based points also date from this period. At the Borax Lake site the concave base points occur with Borax Lake points (Meighan and Haynes 1970: Figure 4). Neither Fredrickson (1974:45) or Meighan and Haynes (1970: note 3) place the concave based points in early Borax Lake but into the next one. However Fredrickson notes the early obsidian hydration readings from Borax Lake and says "Lak-261 may well represent sampling error [i.e., concave base points may be early]' (Fredrickson 1973:198). In any case we tentatively include this concave based type in early Borax Lake Pattern though it may begin earlier in that period than square stemmed Borax Lake points. We place the large side notched points in the same because they are stratigraphically, equivalent at the Willits site (Meighan 1955: types 11 and 7). Late Borax Lake. In this phase two new projectile point types are introduced. The mortar and pestle is also introduced although the hardstone - milling stone canplex persists. The time span of the phase is fran 3000 to 500 BC. The two new projectile points are, we believe, variants of one another and are called here the Excelsior point and the leaf shaped point. Fredrickson (1973:199) says "The defining characteristics of the Excelsior point are a triangular, straight-edge body and a convex base which is frequently ogival in outline; that is, it resembles a pointed arch. A frequent but not necessary attribute of the Excelsior series point is the presence of a definite shoulder at the junction of the body and the base." These points probably would not have been recognized except for Fredrickson's excavation of the Houx site (Lak-261) where there were a great many of them (see Fredrickson 1973:201). At the Houx site these are all large points but in Warm Springs and also in Indian Valley all of them are smaller and becane even more reduced later in time. That these are variants of a leaf shaped with curvate rather than straight sides seems obvious fran the Willits site (canpare Ileigham 1955: P1. 4B with P1. 4C and D). These also are reduced in size later in time. Early Houx Aspect.. The time period here is roughly fran 500 BC to A.D. 1200. In this phase grinding tools are mortar and pestle only. Smaller versions of the Excelsior and leaf shaped point continue. A sall side notched point begins here; this may be simply a smaller version of earlier large side notched point but aside fran the noth there is no great resemblance. Late Houx Aspect. This runs fran A.D. 1200 to historic times. Typological content of the early Houx Aspect catinoms but with the addition of Gunther barbed points. These were defined by Treganza (1958:13-16) and 5 Midden i;i?iii Surface Scatter @"~-4$ Oak Woodland BrushBoulders mh~Jz%& Rock . . . . Sand, Gravel 'fiit~ - Swamp a Test Unit * Datum oAuger Hole * Cupule Rock 0 Fire Hearth -. Housepit .-...*.11....-Suggested Extent of Midden J ASlope or Embankment Creek ....-.-.Spring -- Tributary, Seasonal Stream Roodway- Paved Roadway-Dirt, Unimproved Trees FO Oak El Bay ? Maple ? Cottonwood E Buckeye ? Box Elder * Madrone 0 Alder LA Manzanita Figure 2. Legend for Site Maps. 6 seem always to be late in Northern California. This outlines the basic sequence in the area and gives the primary types we will be using to determine the chronological placenent components (Fig. 1). In determining what is or is not a component (a single phase within a site) we have had recourse to yet another device: the ratio of chert to obsidian chips within a 10 an level of a one meter pit. Since the criterion is likely to be controversial we will attempt some justification here. The testing of the sites in the Warm Springs project was done by digging one or, infrequently, two one meter pits (nine units were dug in Son-600 but that was a special case irrelevant to the present discussion). That meant that the time diagnostic artifacts occurred at the very low frequency for any one site - the greatest number of diagnostic projectile points is 6 fran Son- 547 and 2 or 3 is the usual number. If we had put, say, 10 units in each site we could make component divisions based upon diagnostic artifacts but this was impossible in the circumstances. Waste chips, however, cane not singly but in 10's or 100's per level ard we concluded that our best irdicator of change was a change in the ratio of chert to obsidian chips - we discuss the inferences, other than that of change, that can be made fran this in our concluding section. One of the dangers of using these ratios stems fran the possibility of a single flint knapper working one or a few pieces of raw material in one place at one time thus greatly changing a culturally determined ratio locally. For that reason we accept as an indicator of cultural change a ratio which changes not just once but over several levels in the same direction. We recognize that this method is imperfect but regard it as the best that can be done in the circumstances.