2 Archaeological Field Investigations at the Fort Ross Beach Site KENT G. LIGHTFOOT AND ANN M. SCHIFF TJIS CHAPTER DESCRIBES the field program employed in the investigation of the Fort Ross Beach Site (CA- SON-1898/H). The program involved topographic mapping, the profiling of exposed erosional surfaces, and subsurface testing. Below, we outline the field methods utilized, the profie and excavation units investigated, and the stratigraphy, features, and kinds of cultural materials uncovered in archaeological deposits. SrrE DESCRIPTION The first apparent description of the Fort Ross Beach Site (FRBS) occurred in July, 1953, when U.C. Berkeley archaeologists were excavating the foundations of the Stockade walls and undertaking limited reconnaissance in the nearby hinterland. Treganza (1954:18) discusses "Indian Site No 5" located in the rear of the cove directly below the southeast Bastion. Early illustrations suggest that the Russian boat house and tanmery must have been built over the surface of this site. Whether Treganza is referring to FRBS or to archaeologi- cal deposits farther northeast in the cove area is not clear. John McKenzie recorded midden deposits at the base of the terrace in his 1963 map of archaeological sites in the State Park (McKenzie 1963). A detailed study of FRBS did not take place until the summers of 1988 and 1989 when we initiated our field program. The Northwest Information Center (Sonoma State University) assigned FRBS the permanent trinomial number, CA-SON-1898/H, in 1990. The site is located at the base of an uplifted marine terrace on the northwest side of Fort Ross Cove along the channel of the Fort Ross Creek (see figure P.1). The marine terrace is composed of sandstones and siltstones of the Monterey Formation, dating to the Tertiary, that have been tilted and raised 20 to 30 m above the Fort Ross Cove. Erosional processes, bioturbation, and downward transportation of parent materials from the marine terrace have created an extensive colluvial formation along the base of the terrace. The formation processes of the site are described more fully by Price in chapter 4. The colluvial formation of angular siltstone and sandstone fragments, abraded chert pieces, and finer- grained sands, silts, and clays contains archaeological remains. Subsequent erosion of the lower colluvial formation has taken place through the combined forces of high- energy tidal waves crashing across the Fort Ross Cove during winter storms and of high-water discharge from Fort Ross Creek during seasonal floods. The creek, whose channel has meandered across the northern half of the cove over time, has begun to cut extensively into the colluvial formation at the base of FRBS. The colluvial "toe" has been completely removed along this stretch of the creek, destroying a sizable section of the site, and exposing archaeological materials along the creek bed. Repeated winter and spring storms pounded Fort Ross Cove in the late 1980s, resulting in severe damage to FRBS. As a consequence of accelerated erosion, Breck Parkman requested that a full-scale investigation of the site be undertaken on behalf of the Department of Parks and Recreation. Archaeological investigations to assess the significance and extent of the archaeological deposits that remained began in the summer of 1988. On the north side of the creek channel where the colluvial "toe" had been removed, archaeological materials are 24 The Native Alaskan Neighborhood exposed along a 30 m stretch. No colluvial deposits remain on the south side of the creek-only sand and gravel deposits of the Fort Ross Cove beach. The erosional scar on the north side of the creek is a slumped bank that rises 2 to 3.5 m above the creek bottom. The slope of the bank varies in angle, depending largely on the presence of an underlying stratum of consolidated clay sediments that proves to be relatively resistant to wave action, periodic floods, and trowels. In contrast, the upper strata of loosely aggregated sandstone, chert, and siltstone colluvium are very susceptible to erosion. The topography and dimensions of the site are best defined in three profile sections running east to west along the erosional scar (figure 2.1). 1) East Profile. The eastemmost 10 m section of the bank of the erosional scar is lowest in elevation (2 to 2.3 m above the creek bed) and relatively steep in profile. The uppermost colluvium is poorly sorted and unconsoli- dated, while the bottom stratum of consolidated clay sediments creates a slight slump at the very bottom of the erosional cut. A bench exists between the upper bank of the erosional scar and the steep incline of the marine terrace, creating a relatively level area within the site that measures about 10-by-1O m in area. This low bench extends beyond the site to the east where the Ross shipyard, blacksmith shop, and storage sheds were probably located. We arbitrarily define the eastern boundary of the site area by a shallow ravine created by stonn run-off from the Old Russian Road. 2) Middle Profile. The next westward 10 m section of the bank of the erosional scar is higher in elevation (3 to 3.5 m above the creek bed), but less vertical in profile. The bank slumps two or more meters into the creek bed where an extensive stratum of consolidated clay sedi- ments is found. The top of the bank forms the southwest- em edge of the aforementioned bench, beyond which the topography becomes very steep along the cliff face of the marine terrace. 3) West Profile. The remaining 10 m section of the bank of the erosional cut is similar in elevation to the Middle Profile, but much more precipitous in profile. This entire section is characterized by a very steep slope that continues up the cliff face of the marine terrace. No level areas exist in the third section of the site. The dimensions of FRBS, as defined in the field, follow. The East Profile consists of a 10-by-10 m area formed by the erosional scar and bench. The Middle Profile is oblong in shape, 10 m in length along the erosional scar and narrowing in width from 10 to 2 m as the bench becomes more constricted from east to west. The West Profile is defined by a 10 m stretch along the erosional scar and a 2-m-wide strip extending into the steep terrace slope. FIELD PROCEDURES The field program for investigating FRBS involved excavating and profiling the 30-m-long erosional scar, and excavating in two areas of the bench (figure 2.2). We believed this was the most judicious strategy for delineat- ing the overall spatial organization of archaeological materials in the site. 1988 FIELD SEASON We began by establishing a site datum above the erosional cut, slightly east of the site proper (figure 2.1). From this reference point, a topographic map was produced using a transit and stadia rod, with all eleva- tions converted to meters above sea level (asl). Site datum is 4.77 m above sea level. We then laid-out eighteen profile units in the East and Middle profiles, and a 2-by-.5 m unit in the eastern side of the bench. PROFILE UNITS The profile units, each measuring 1 m in length, were staked along the upper edge of the erosional scar. Since the orientation of the erosional scar was between 60 to 70 degrees, the unit corners do not conform to typical north/south oriented 1-by- 1 m grid squares. Each profile unit was laid-out from the two corner stakes in the upper bank to the creek bottom at an angle perpendicular (150 to 160 degrees) to the orientation of the erosional face. The west corner stakes were designated as unit datums, and the coordinates of these corners, measured in meters south and west from the site datum, were used to label profile units in the field (e.g., 4.5S, 4.9W; 6.2S, 8.5W). We collected artifactual materials from the surface of each profile unit prior to excavation. The tenth profile unit (8.3S, 12.9W), counting from the east, which separated the East and Middle profiles of the site, served as a balk and was not excavated. The excavation of the profile units was undertaken to produce a clean, vertical surface that delineated the natural and cultural stratigraphy, as well as associated archaeological materials. Elevations for all profile stakes were shot from site datum and converted into asl read- ings. Separate levels were maintained for each natural or cultural stratum defined in the field. When natural or cultural strata extended more than 10 cm in depth, we divided them into arbitrary 10 cm levels. Trowels were used to excavate upper colluvial deposits, and picks and shovels had to be used to remove the underlying consoli- dated clay sediments. All sediments were screened through 1/4" mesh. Materials were point provenienced in relation to the unit datum (horizontal and vertical readings). Soil and pollen samples were taken from each natural or cultural stratum defined in the field. The above procedures were employed in the excavation of the East Fort Ross Beach Site Investigations 25 I,~~~~~~~~~~~~~~~~~~~~~~7 A.0"X 0VamEt0 wa ~ A a . . . . . . i\ . A . . 0 A, ZGA~~~. .. o 4?~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~1 a)~ ~~~~~~~~~~~~~~C 26 The Native Alaskan Neighborhood Figure 2.2 Photo of the Fort Ross Beach Site from the East, Showing 30 m Long Profile and Bench in 1989 Profile units (4.5S, 4.9W to 8S, 12W). These units required relatively minimal removal of sediments to produce clean, vertical faces. While the widths of the deposits varied, none of the profile units in the East Profile were greater than 1 m in length and .5 m in width, while depth varied with height of the upper bank. The Middle Profile units were more complicated to excavate since they required the removal of a 2-m-wide slump to produce clean, vertical faces. The slump was excavated in the following manner. Profile units were laid-out from the upper comer stakes perpendicular to the erosional scar on either a 150 or 160 degree bearing. Along this bearing, stakes were placed at .5 m intervals from the upper comer stakes. The slump was removed in four separate steps, each measuring 1-by-.5 m. The northwest comer of each 1-by-.5 m step became the step datum. We employed a rather involved provenience system in the field. The first step (0 to .5m) was desig- nated by the western stake of the profile unit (e.g., 8.9S, 14.8W); the second step (.5 to 1 m) by the .5 m stake (e.g., 8.9S, 14.8W-.Sm on a 160 degree bearing); the third step (1 to 1.5 m) by the 1 m stake (e.g., 8.9S, 14.8W-1 m on a 160 degree bearing); and the fourth step (1.5 to 2 m) by the 1.5 m stake (8.9S, 14.8W-1.5 m on a 160 degree bearing). Although the steps were complicated to excavate, they proved effective in the field when we fortuitously uncovered an intact pit feature in profile units 9.7S, 16.8S; 10.3S, 17.6W; and 1O.9S, 18.4W. With the able assistance of John Holson, an archaeologist now with Pacific Legacy Inc., we designated the northwest comer of the second step (.5 m stake) in profile unit 10.3S, 17.6W as the feature subdatum. The 12 steps in the three profile units comprising the pit feature were then exca- vated concurrently, with elevations taken in relation to the feature subdatum. We attempted to point provenience all artifactual materials larger than the excavator's thumbnail within the feature. Archaeological remains recovered in the screen were lot provenienced by level and by step. We maintained separate lot bags for materials recovered within and outside the feature within the same step. Soil and pollen samples were taken throughout the pit feature. EAsTBENcH (ON, 12W) In testing the upper bench area, we placed a 2-by-.5 m unit on the east side of the bench, 10 to 12 m west of the site datum. In designating unit datums, we normally chose the southwest corners unless surface topography was marked. In the latter cases, corners with the highest elevation were selected as unit datums. The East Bench exhibited enough topographic variation to designate the northwest corner (ON, 12W) as the unit datum. We employed the same basic excavation procedures as outlined for the profile units above. 1989 FIELD SEASON We excavated twelve additional profile units in the West Profile, and a 2-by-3 m block in the Southwest Bench (figure 2.1). The same methods outlined above were employed in the excavation with two exceptions. First, we no longer collected pollen samples. Fort Ross Beach Site Investigations 27 Professor Roger Byrne, a palynologist at U.C. Berkeley, assisted us in collecting pollen samples in 1988. An analysis of these samples in his laboratory during the winter and spring months prior to the 1989 field season yielded no discernible pollen. These results indicated that the sedimentary context of FRBS was not conducive to the preservation of pollen. Second, we recognized that a significant shortcom- ing in our 1988 excavation strategy was the use of only 1/4" mesh for sediment screening. Continued reliance on this screen size would bias the diversity and quantity of "micro" materials in our excavation. In the 1989 field season, Thomas Wake constructed a wet screening procedure, with the assistance of State Park maintenance personnel, that pumped water from the nearby Fort Ross Creek at a high velocity into 1/16" mesh. The wet screening of sediments greatly enhanced our ability to recover small faunal remains, beads, and chipped stone debitage. PROFILE UNITS The profile units were laid-out along the upper bank of the erosional scar using the methods described above. Since the vertical slope was quite steep in the west section, separate excavation steps were not necessary to produce a clean, vertical face in any profile unit. We modified the previous field provenience system, and simply designated each profile unit with a letter, begin- ning with A for the easternmost unit (adjacent to the last unit excavated in 1988), and L for the westernmost. Since unit A separated the Middle and West profiles, it served as a balk and was not excavated. Unit H was chosen for water screening. All sediments excavated from this unit were water screened in 1/16" mesh. The sediments from all the other profile units were dry screened through 1/4" mesh. SOuTHwE=STBFNCH (6S, 19W) The reasons for excavating this block were twofold. First, by placing it in the bench's southwest corner, it pro- vided a good comparison to the East Bench (ON, 12W). Second, the excavation block was placed directly behind and upslope of the pit feature detected in 1988. We could then evaluate whether the pit feature was connected to or associated with a larger structure constructed into the terrace slope. The 2-by-3 m block was divided into six 1- by-I m units, each designated by the coordinates of their southwest comer stakes in relation to site datum. The excavation units included 7S, 17W; 7S, 18W; 7S, 19W; 8S, 17W; 8S, 18W; and 8S, 19W. The northwest corner of the block (6S, 19W) the highest point of all the excavation units, served as the block subdatum for taking elevations (figure 2.1). We began wet screening sedi- ments through 1/16" mesh from four units (7S, 17W; 7S, 19W; 8S, 17W; 8S, 19W), and dry screening sediments through 1/4" mesh from two (7S, 18W; 8S, 18W). How- ever, in the lowermost mottled-clay sediments that con- tained primarily lithic artifacts, we wet screened sedi- ments through 1/16' mesh from only 7S, 17W and 8S, 19W. After the 1989 field season, we standardized the labeling of all profile units and steps, and will employ this simplified provenience system for the remainder of the volume. Each profile unit is assigned a 'P' (profile) number, counting consecutively from east to west, beginning with P1, P2, P3, to P30. Excavation steps are assigned letters, 'a' for step 1, 'b' for step 2, 'c' for step 3, and 'd' for step 4. For example, step four (1.5 m along the 150 degree bearing) of unit 9.7S, 16.8W is simply P14 step d. Table 2.1 correlates the proveniences used in the field with those in publications. We present this table primarily for scholars who may use the original field notes, excavation forms, and provenience information for archaeological materials that will be archived in the State Parks archaeological facilities in Sacramento. Table 2.1 Field/Publication Designations for Profile Units Field Publication Field Publication 4.5S, 4.9W P1 10.9S, 18.4W P16 4.9S,5.8W P2 11.4S, 19.3W P17 5.4S, 6.6W P3 12.1S, 20.0W P18 5.8S, 7.6W P4 A P19 6.2S, 8.5W P5 B P20 6.6S, 9.4W P6 C P21 7.1S, 10.3W P7 D P22 7.5S, 11.2W P8 E P23 8.0S, 12.0W P9 F P24 83S, 12.9W PlO G P25 8.6S, 13.9W P11 H P26 8.9S, 14.8W P12 I P27 9.2S, 16.0W P13 J P28 9.7S, 16.8W P14 K P29 10.3S, 17.6W P15 L P30 STRATIGRAPHY AND ASSOCIATED CULTURAL MATERIALS In this section, we describe the stratigraphy, features, and overall distribution of artifacts and faunal remains observed in five areas of FRBS: 1) East Profile, 2) Middle Profile, 3) West Profile, 4) East Bench, and 5) Southwest Bench. EAST PROFILE (Pl-P9) The upper bank of the erosional scar rises slightly in elevation from east (3.77 m asl) to west (4.3 m asl). The entire section was cleaned and excavated to an elevation of 2.9 to 3.0 m asl (figure 2.3). Three deposits are 28 The Native Alaskan Neighborhood Figure 2.3 Photo of East and Middle Profile Units in 1988 defined (figure 2.4). 1) Topsoil. The dark brown (lOYR 2/2) soil is dry, not well compacted, and contains unsorted colluvial materials, including angular sandstone, siltstone, and chert pieces. Most of the colluvial materials are small in size, less than 5 cm in diameter. The soil is thin, no deeper than 10 cm in depth, and in some units (P3, P7) has been completely removed by erosion. Archaeological materials are sparse, including one historic ceramic, a few lithic specimens, and some mollusk remains (aba- lone, snail), and mammal bones. Charcoal is not com- mon, and no features are observed. 2) Midden. The dark brown to black (IOYR 2/1) deposit is relatively loose and non-compact, containing some unsorted colluvial sandstone, siltstone, and chert pieces. The colluvial materials vary greatly in size, from 10 cm in diameter to under 1 cm; most are the latter size. The soil matrix is organically rich, including many fragments of bone, shell, and charcoal. The deposit ranges in thickness from 20 to 50 cm, increasing in depth in the western units (P8, P9). Evidence of rodent dis- turbance is common. A diverse range of artifacts has been recovered, including glass artifacts, ceramic sherds, metal remains, and lithics. An assortment of mollusk, fish, bird, and mammal remains also has been identified. No features are observed. 3) Clay. The underlying stratum is a consolidated brownish-yellow (IOYR 6/6), clay-rich soil that contains angular sandstone and siltstone pieces, and large clumps of clay. The clay-rich sediments are first encountered at elevations of 3.8 to 3.3 m asl, and continue to the bottom of the profile units. The clay matrix contains no glass, metal, or ceramic artifacts and few mollusk and mammal remains. However, a large number of lithic artifacts have been recovered. Charcoal is very sparse. No features are observed. MIDDLE PROFILE (P1 -P18) The upper bank of the erosional scar rises sharply from the eastern corner of unit P 1(4.42 m asl) to the western corner of P18 (5.9 m asl). All units are exca- vated into the underlying clay stratum. The final depths of the units vary from 3.24 to 3.0 m asl in P11, P12, and P13; 3.6 to 3.4 m asl in P14, P15, and P16 (which contain the FRBS Pit Feature); and 3.8 m asl in P17 and P18. Five stratigraphic units are defined (figure 2.5). 1) Topsoil. The dark grayish-brown (IOYR 3/2) soil is composed of angular sandstone, siltstone, and chert inclusions. The stratum is about 10 to 30 cm thick. Recent slope wash has removed the upper stratum from some of P1. P14 contains an unusually large number of angular siltstone fragments. Relatively few artifacts and some mollusk fragments were recovered. 2) Fill. The dark brown (IOYR 3/3) deposit, dry and very loose in texture, is composed of many large angular siltstone rocks, a result of widening the Fort Ross Cove Road in the 1920s. The dirt road, located directly above FBRS, was cut into the marine terrace, producing a considerable amount of rock refuse that overlies some areas of this site. The road-building debris serves as a Fort Ross Beach Site Investigations 29 I ( I*,S CMT 0D t 0 e~~~~c . * :< .i t' CO) C') CO) . * * * . .* s *. . .. . 1 1 C1* xe U)~~~~~~~~~~~n- 10 ~ 0 C" 0 C') C' 0C' 30 The Native Alaskan Neighborhood C' co It C 4 CO C6 'JI I I I I I I i co co _- o c (1 a ) U e ir X m 0 co c Rs V X a o 0 0 Cco 00 0ClO i t i,D Ii2 I I I I a I I I I I I I I - -- qql 0 (0 clJ co C . 9 ui) LiC 4 4 C') C') C') CC) CL6 Ca0 :: Or 0 CL v- C'J C'I) V- a. a. Ln 3c ZZ) Iz, 2. CL. 1%) ?i in elli qu I" bo iz CD a. V- co T-- a. C6 , I *(0 * 0 I I Fort Ross Beach Site Investigations 31 stratigraphic marker in the westernmost units (P15, P16, P17, P18). The fill stratum, as defined in the field, is almost 1.1 m thick in P16, P17, and P18, and only .1 to .6 m thick in the other units. A sparse distribution of lithic tools, mollusk remains, and mammal bones is observed in the fill. 3) Midden. The dark brown (lOYR 2/2) deposit contains a shell and animal bone-rich matrix, not very well compacted, with inclusions of smaller-sized angular sandstone and siltstone rocks. Most colluvial materials are less than 3 cm in diameter. The deposit is only .2 to .3 m thick in P1 1, then increases in depth in P12 to P16, where it is almost 1 m thick in places. In P17 and P18, the midden tapers to only 10 to 20 cm in size before disappearing. A diverse range of glass, ceramic, and lithic artifacts has been recovered as well as an impres- sive assemblage of mollusk shells and animal bones, especially above the FRBS Pit Feature described in more detail below. Charcoal particles are distributed through- out most profile units, and evidence of rodent burrowing is noted in some units. 4) Mottled Brown Clay I. The brown (lOYR 4/3) deposit, fairly compact in hardness, is marked by a clay- rich matrix containing small decomposing red and yellow sandstone pieces, most measuring less than 3 cm in diameter. It underlies the midden stratum west of the pit feature, and ranges from .4 to .6 m in thickness. It is absent east of the pit feature. Charcoal fragments are sparse, and only a few mollusk and lithic specimens were uncovered. 5) Clay. The brownish-yellow (IOYR 6/6) clay-rich sediments compnse the lowermost stratum in all profile units. It contains yellow and red decomposing sandstone rocks, large clay fragments, and some angular siltstone pieces. While few mollusk remains, animal bones, charcoal, or ceramic, metal, and glass artifacts were recovered, lithic artifacts are common. FRBS PIT FEATURE The FRBS Pit Feature in P14, P15, and P16 was originally dug into the underlying clay soil and the brown mottled clay of its west side (figure 2.6). The upper portion of the feature appears to have been dug into the lowermost levels of the midden soil in P14 and P15 (see figure 2.5). The pit, shaped like the hull of a ship, is constructed with its sides sloping concavely toward the bottom. It measures 2 m at its widest point on the 60 degree axis, 1.75 m along its 150 degree axis (extending out from the profile face), and 1 m in depth from the top edge of the pit to the bottom. The wall of the pit is lined with clay, 5 to 8 cm thick. A stone bench is built on the clay surface in the bottom of the pit (figure 2.7). This bench is constructed of large flat rocks, many measuring 25 to 30 cm in length and about 10 cm in width. These are laid down on the clay surface, in three or four courses, at a height that probably reached 45 to 50 cm above the pit floor. The bench was not intact when excavated, and the angle and position of many rocks suggest they had been disturbed from their original position. The bench is built across the width of the pit (along the 60 degree axis). It is not clear how far it extended out from the profile wall, although Figure 2.6 The Fort Ross Beach Site Pit Feature I 32 The Native Alaskan Neighborhood Figure 2.7 Close-Up of Stone Bench in the Fort Ross Beach Site Pit Feature intact foundation stones indicate it protruded at least .9 m. A possible post-mold, measuring 20 by 25 cm and surrounded by four rocks, is located in front of the bench about 1.5 m from the profile wall (figures 2.8, 2.9). The pit feature had been subjected to very hot temperatures. The clay used to construct the pit exhibits signs of thermal alteration, most marked by bright red, orange, and purple colors that radiate out from the interior to exterior surfaces, and fine charcoal powder along the interior surface. In addition, some of the rocks in the bottom of the pit show evidence of fire-cracking. We surmise that a very hot fire was generated in the pit, probably around the stone bench on the floor of the feature. However, while the interior clay surface contains charcoal powder, larger chunks of charcoal were not recovered on the floor or in the pit fill. The paucity of evidence for an intemal source of fire suggests the fea- ture may have been cleaned prior to abandonment or that rocks were heated nearby and then placed inside the pit The floor of the feature appears to have been swept out. Few artifacts were recovered on or near the floor surface. A dense aggregate of mollusks and large mammal bones is observed in the midden deposit directly above the FRBS Pit Feature (figure 2.5). These materials are probably concentrated above the pit, possibly as trash near or on the roof of a structure associated with this feature. When the structure was abandoned, the roof appears to have collapsed inward, and the faunal remains were deposited on top of the already filled-in pit feature. WEST PROFILE (P20-P30) The upper bank of the erosional scar undulates considerably, a product of recent landslides and slope wash (figure 2.10). The surface slopes down from P20, P21, P22, and P23 (5.62 to 5.0 m asl) on the east side, bottoms out in P24, P25, P26, and P27 (5.14 to 4.78 m asl), and rises again in P28, P29, and P30 (5.34 to 5.1 m asl) on the western end. The topsoil is absent in all units. Either it has been removed by recent erosion and/or the upper bank has not been stable enough for its formation. All profile units are excavated to a depth of 3.7 to 3.4 m asl. Four deposits are defined (figure 2.11). 1) Fill. The dark brown (10 YR 3/3) deposit is dry and crumbly, a loose aggregate of colluvial materials, including angular siltstone rocks that are a product of road construction activities above the site. The presence and depth of fill varies significantly across the section. The deposit is over .5 m thick in the eastern units, absent in the middle units, and variable in depth in the western units where slippage scars from previous soil movements are evident. The deposit contains few artifacts and almost no faunal remains. 2) Mottled Brown Clay II. Similar to units P17 and P18, the grayish-brown (IOYR 3/2) deposit is relatively compact, clay-rich, and contains inclusions of decompos- ing red and yellow sandstone. However, it differs in the size range of the colluvial materials by including very large angular siltstone rocks similar to those found in the fill. While most sediments are dry and easy to excavate, Fort Ross Beach Site Investigations 33 moist soil was observed in units P28, P29, and P30 throughout the summer of 1989, suggesting that a spring may be located nearby. The stratum is very thick in most units, in several cases greater than 1 m deep. It contains a diverse range of artifactual and faunal remains. Most of the large angular rocks; ceramic, metal, and glass artifacts; mollusk specimens, and animal bones, however, are found in the upper .3 to .4 m level. Charcoal and lithics are found throughout the deposit. 3) Yellow Clay. Two separate lenses of yellowish- brown (IOYR 5/6) clay extend along the lower levels of units P20 to P26. This clay-rich matrix is marked by many large angular siltstone rocks, no shell or animal remains, and only a few lithic artifacts. 4) Beach Gravel. Beach gravel underlies all the units except P20 on the east, and P29 and P30 on the west. It is likely, however that had these units been excavated to a deeper level, then beach gravel would probably have been found. This stratum is associated directly with the beach deposits in the Fort Ross Cove. As Price notes in chapter 4, this finding indicates that colluvial sediments were deposited directly on top of older beach deposits that once extended along the base of the marine terrace. Figure 2.8 Possible Post-Mold in front of Stone Bench in the Fort Ross Beach Site Pit Feature No shell specimens, animal remains, or European goods were recovered in this stratum, but lithic artifacts were collected. EASTBENCH (ON, 12W) The surface of this unit slopes downward from west to east 42 cm, from 6.98 m asl in the northwestem comer to 6.56 m asl in the northeastem comer. The unit is excavated at the angle of the slope, following the natural stratigraphy, to a maximum depth of about 90 cm below surface. Three deposits are defined (figure 2.12). 1) Topsoil. The dark brown (1OYR 3/3) soil, dry and crumbly, is composed of unsorted colluvial materials, including sandstone, siltstone, and chert pieces, some over 10 cm in length. The deposit covers the entire unit to a depth of 20 to 30 cm below surface. The topsoil was not screened in this uniL 2) Mottled Brown Clay II. The grayish-brown (IOYR 3/2), compact, clay-rich deposit is composed of small decomposing red and yellow sandstone fragments, as well as many large angular siltstone rocks. The deposit is 40 to 45 cm thick in the west end of the unit, and tapers to a wedge only 10 to 15 cm thick in the east end. Archaeological materials include glass, ceramic, and lithic artifacts, and some mollusk shells. 3) Midden. The dark (IOYR 2/1) deposit is charac- terized by a high carbon content, a dense concentration of shell and animal bone fragments, decomposing red and yellow sandstone, and a few large angular rocks. The midden covers the entire unit beginning at a depth of about 6.4 m asl. Time constraints only allowed sampling of this deposit. We sampled a 20-to-40-cm thick stratum of the midden, within which we found a diverse range of ceramic, metal, glass, and lithic artifacts, mollusk shells, and animal bones. Along the northern wall, red clay or 34 The Native Alaskan Neighborhood -o 0 N w 7 U) ~ ~ ~ > 0~~ C.)~ ~ ~ ~ C ELa --a: 05jI CY) r- 2ZI. v- _ ~ _( 0s: Fort Ross Beach Site Investigations 35 Figure 2.10 Photo of West Profile Units . : ,, M a: r , ; . ; ... : :: i: : w ,:i:-T daub was observed and collected, indicating possible in situ burning. Directly west of the discolored soil is a concentration of charcoal, shell, and animal bone. Augering in the bottom of the unit indicates that the midden stratum is at least 1.95 to 2.0 m thick, extending to a depth of 4.33 m asl. SOTrHWESTBENCH (6S, 19W) The surface slopes steeply downward across the six 1-by-I units from west to east and from north to south. From the northwestern comer, 6S, 19W (7.7 m asl), the highest point in the block, the surface drops 1.26 m in the three meter distance to the northeastern corner, 6S, 16W (6.44 m asl) (figure 2.13). From 6S, 19W to the south- westem corner, 8S, 19W (7.08 m asl), the drop is .62 cm in the two-meter distance (figures 2.14 and 2.15). Exca- vation follows the slope of the natural stratigraphy. Most of the units are excavated between .97 and 1.33 m below surface: 7S, 17W (1.22 m); 7S, 18W (1.33 m); 7S, 19W (1.32 m); 8S, 17W (.97 m); and 8S, 18W (1.11 Im). However, we excavated 8S, 19W to the underlying bedrock, about 1.46 m below surface. Five deposits, including the bedrock in 8S, 19W, are described below (figures 2.13 and 2.14). 1) Topsoil. The dark brown (lOYR 3/3) soil is dry, crumbly, and highly organic (grass roots) with small colluvial inclusions. Topsoil covers the entire block, ranging in depth from about 10 to 25 cm. The deposit was not screened. 2) Fill. In the western units of the block, a dense deposit (lOYR 3/3) of angular siltstone rocks exists to a depth of about 40 to 60 cm below surface. The deposi- tion of this rock fill is associated with the construction of the road above the site in the 1920s. The rock fill tapers in depth in the eastern units, disappearing completely in unit 7S, 17W and the eastern edge of 8S, 17W. The deposit was not screened. 3) Mottled Brown Clay II. Underlying the fill is a grayish-brown (1OYR 3/2) deposit of clay-rich sediments that contains red and yellow decomposing sandstone fragments, some large angular rocks, and many fragments of shell and charcoal. The soil varies in compactness across the units, depending largely on rodent activity. The stratum is most extensive in the northwest units (7S, 19W; 8S, 19W) where it is almost .6 to .7 m deep. It is less than .5 m thick in the other units. A varied assem- blage of ceramic, metal, glass, and lithic artifacts was collected, as well as a diverse range of mollusk and animal remains. Charcoal fragments are common throughout the stratum. No features were detected. 4) Highly Mottled Clay. No distinct separation exists between this deposit and the mottled brown clay II. Excavators note that the lowermost stratum is more compact and characterized by a higher clay content, and more decomposing red and yellow sandstone fragments. The thickness of this stratum varies inversely with the depth of the mottled brown clay. A thin layer, .1 to .2 m thick, is exposed in the bottom of 7S, 19W; while 7S, 17W and 8S, 17W exhibit a stratum .3 to .5 m deep. In 8S, 19W, the only unit excavated to bedrock, the highly 36 The Native Alaskan Neighborhood 0* co CY 0 r o cc u * * i * * * * 0 o ' .0 0 CD 0 _ t) cn C C 4 ) J o 0 c .C C) 0 = CO %0x 0~~~ > 0 0O 011011E1 gN1 I I I , I * -- I _ Cs 0 sJ0OCD 0 'i , 0Y CY cL c() EL Z:) U- Co *N u= CL Cu - 0= C\L a) NV --_ 0= a. CY CIY 0."I"+ D1 CL Fort Ross Beach Site Investigations cD ";r c\j 0 (6 C (6 (6 l I I I Il a) 0 0 E 0 o C CU 0D o a c - x 0 _ cc m CO * CU~~~~C o -c co 4~~~~~~c cn Q m >. c: ._ V C *b 0(Ua 0 0 0 0 O c0 -o - 37 0 z 0 v- z 0 --l P- N- -I :s ._ cv z 0 r I I I I I I I I I - - -O 0 (D co CJ 0 vi n-z (6 c 6 c CO 6 38 The Native Alaskan Neighborhood :~(.6 CD C a C ZO co c x co (t *6 C 0 0c (D co uL 11 E (.0 v- c) (0 (I) 11- cD CD co .q - CO) CD (C) CO D ID I en ra 5d cm I II 0 co N-~ C. I I I I I I - I I I I I I I I I Fort Ross Beach Site Investigations 39 mottled clay stratum is .3 to .4 m thick (figure 2.14). The density and diversity of historic ceramic, metal, and glass artifacts decrease dramatically, as well as the quantity of charcoal, shell and animal bone fragments in this highly mottled clay stratum. However, a large number of lithic artifacts was recovered. No features were detected. 5) Bedrock. The bedrock in the northwest corner of 8S, 19W is a brownish-yellow (IOYR 6/6) sandstone (figures 2.14 and 2.15). Many angular, decomposing fragments of sandstone are noted on the contact surface of the bedrock in unit 8S, 19W. CONCLUSION The archaeological investigation of FRBS involved the cleaning and excavation of 28 profile units along the erosional face of the marine terrace, and the excavation of two areas on the bench. Archaeological materials were uncovered in a colluvial formation along the base of the marine terrace. The site is an active landscape. Wave action and floods have removed the colluvial "toe"; slope wash and landslides are continually modifying the steep slope of the West Profile; rodent activity is common; and past road building activities are clearly evident in the stratigraphy. We recognize that some archaeological materials, especially in the West Profile, were probably not recovered in contexts in which they were originally deposited, but have been tansported downhill, probably from NAVS upslope, and mixed with other material culture over time. While dynamic processes have been at work, several lines of evidence suggest an organizatonal structure to FRBS. 1) The Bench. The excavation of the 2-by-3 m block in the Southwest Bench unearthed sandstone bedrock relatively close to the surface. This finding indicates the bench was not created entirely from colluvial deposition, but that its underlying structure is part of the base of the marine terrace. The marine terrace would have been extant when people first began to use the Ross Region, at least 8,000 to 6,000 years ago (Lightfoot et al. 1991:110- 12). The thick midden deposits on the bench could have resulted from long-term in situ use of this topographic place, with some mixing of colluvium materials depos- ited down the slope of the marine terrace, including archaeological materials from NAVS. The Native Alaskan workers may have used the bench as a work area that overlooked the sheds where their baidarkas and fishing/hunting equipment were stored. Eyewitness accounts also suggest that houses may have been constructed down the slope of the marine terace into Fort Ross Cove. Since the sides of the marine terrace have been extensively modified by road construction and erosion, the bench represents one of the few enduring places where intact house structures may still be found. Our excavation of the 2-by-3 m block in the Southwest Bench indicates it will be very difficult to discriminate archaeological materials deposited in situ from those transported downhill from NAVS given the steep slope and mottled nature of the strata. However, there does appear to be some vertical sorting of materials. Historical materials and lithic artifacts are found in the upper levels of the block, while only lithics are in the lower. This vertical distribution may indicate the presence of both prehistoric and historical components on the bench. The most likely place to find in situ deposits is in the East Bench, an area conspicuous by its relatively gentle slope. The topsoil and mottled brown clay soils in ON, 12W appear to have been formed, at least in part, by colluvial action. However, the underlying thick midden stratum may have been created primarily by in situ cultural deposition. This observation is supported by the presence of a possible feature associated with shell and animal bone refuse. It is likely that materials associated with cove activities or local households may still be buried here. However, our testing of the midden is too limited to evaluate this proposition at this time. 2) East and Middle Profiles. The lowest consoli- dated clay stratum in the East and Middle profiles of the site is solid and relatively resilient to erosion. This stratum may have formed on top of the bedrock that extended into Fort Ross Cove. A similar clay stratum is deposited on bedrock in the Southwest Bench and on top of the marine terrace at NAVS. The recovery of prima- rily lithic artifacts and very few ceramic, glass, and metal artifacts suggests the clay was laid down prior to the construction of Fort Ross. The midden deposit overlying the clay in the East and Middle profiles is an extension of the midden materials found on the bench. The common occurrence of historic ceramic, glass, and metal artifacts throughout this stratum suggests it was deposited during and after the Russian settlement of Ross. The FRBS Pit Feature, excavated into the clay stratum and part of the midden deposit, may not be an isolated structure. While no features are found directly behind it, other structures may have been constructed on the nearby bench during the Ross occupation. 3) West Profile. The beach gravel underlying the West Profile indicates that colluvial materials are deposited directly on the beach. The bench does not appear to have extended along this area of the site. There is a good chance that most materials found here were transported down the steep slope of the marine terrace from NAVS. However, despite this mixing, there appears to be some vertical separation of materials. Mollusk shells, animal bones, ceramic, glass, and metal artifacts are found primarily in the upper 30-40 cm levels, while lithics are found throughout, even in the beach gravels. 40 The Native Alaskan Neighborhood Figure 2.14 Profile of West Wall of Southwest Bench 8S 19W 7S 19W I I 6S 19W M.a.s.l. 7.6 Qi 7.4 - 7.2 - 7.0 - 6.8 - 6.6 - 6.4 1 meter [JIJJ Topsoil < Rodent Burrow Fill Stone F7.'. ..Mottled Brown B W[ 1Clay I I Bone n~7~: lHighly Mottled Charcoal U\\d BeClay x Chal Bedrock x Shell M.a.s.l. 7.0 6.8 - 6.6 - 6.4 - 6.2 - 6.0 - 5.8 - 5.6 I I I I I Fort Ross Beach Site Investigations 41 Figure 2.15 West Wall of Southwest Bench. Note Bedrock in Lower Left Corner. Again, it is possible that a prehistoric component underlies this section of the site. REFERENCES McKenzie, John 1963 Historic Resources and Indian Sites at Fort Ross State Historic Park as Identified by John McKenzie, August 20, 1963. Manuscript on File, Cultural Heritage Section, Archaeology Laboratory, Califomia Department of Parks and Recreation, Sacramento. Treganza, Adan E. 1954 Fort Ross a Study in HistoricalArchaeology. Reports of the University of California Archaeological Survey No. 23:1-26.