|
 |
Group Environment
July 30, 2001 Prepared
by: |
|
Cultural Resources Context CULTURAL RESOURCE CONTEXT FOR
THE
Alexander, Burke, Cabarrus, Caldwell,
Catawba, Cleveland, Gaston, Iredell, Lincoln, McDowell Mecklenberg,
Rowan, Rutherford, and Watauga Counties, and Chester, Fairfield, Kershaw, Lancaster,
and York Counties,
Prepared for: Duke Power Lake Management
Prepared by: THE LOUIS BERGER GROUP, Inc.
May 2000
Table of Contents
List of Tables
The Catawba-Wateree River System (river system) includes portions of 14 counties in North Carolina and 5 counties in South Carolina. This context is a synthesis of the various sources on the cultural history of the Catawba-Wateree River System, and provides a framework to assess the relative importance of cultural resources located within the project area. The project area consists of all land within 0.25 miles of the each of the 11 lakes in the river system (Table 1); however, the following contexts focus on the entire Catawba-Wateree region. The context is divided into three sections: environmental setting, prehistoric period, and historic period. Table
1
To understand the nature of human-land relationships since the end of the last Ice Age approximately 12,000 years ago, the first section of the context will describe the character of the natural environment in the project area and present a model of Holocene environments in the river system. The Catawba-Wateree river system project area begins in the foothills of the Blue Ridge Mountains, continues across the Piedmont of North and South Carolina, enters the Fall Zone of South Carolina, and ends in the upper Coastal Plain of South Carolina. The Blue Ridge province extends from Georgia to Pennsylvania (Spencer et al.1989:T157:1). It is bounded on the east by the Piedmont province and to the west by the Appalachian Mountains. The geology of the Southern Blue Ridge is much more complex than that of the Northern Blue Ridge, it is a northeast-plunging anticlinorum that contains a core of Grenville age high-grade gneisses that is flanked and overlain by Paleozoic metasediments and metavolcanics (Clark et al. 1989:T150:4). The relief consists predominantly of strongly sloping to very steep uplands, stream terraces, and narrow, nearly level flood plains (Mathis 1995:2). Elevations in the westernmost portion of the project area, in McDowell County, North Carolina, range from 980 to 5,665 feet above mean sea level (amsl). The foothills represent the transition from the Piedmont province to the Blue Ridge province. The terrain in the foothills, as the landscape is characterized by discontinuous mountains and hills that intermingle with the hilly relief of the Piedmont (Tuttle 1997:2). In Wilkes County, North Carolina, which is immediately north of the project area, elevations range from 880 to 4,100 feet amsl. Geologically, the area falls within the Inner Piedmont Belt, a zone of stratified rocks of thinly layered mica schist and gneiss, interlayered with amphibolite, calcsilicate rock, hornblende gneiss, quartzite, and some marble (NCGS 1988). The Piedmont comprises a dissected plateau whose surface features principally reflect erosion. The landscape in the Piedmont is characterized by hilly relief, with moderately sloped valley walls and a dendritic drainage pattern (House and Ballenger 1976). The uplands overlooking river and stream valleys consist of steep-sided ridges and ridge spurs. The slopes of which form a deep, almost sheer-sided valley. Along the river valley bottom, the relief is virtually level. Wide interriverine zones separating the major river valleys in the Piedmont show little variation in local relief. Rock types in the Piedmont are primarily metamorphic consisting of schists, gneisses, and slates. Granitic igneous rocks are also present where intrusive activities have occurred (Kovacik and Winberry 1987:16). These activities have resulted in the formation of dikes of basalt, diabase, andesite, gabbro, and other igneous rocks throughout the Piedmont. Residual quartz chunks are also abundant in localized areas (Overstreet and Bell 1965). The project area is underlain primarily by units of intrusive granite, metamorphic granite, mica gneiss, and amphibolite. The mica gneiss is of particular interest with regard to prehistoric land use because this rock type contains large amounts of quartz (Campbell 1982), which would have been utilized in prehistoric chipped-stone tool industries. Prehistoric populations might have also made use of mica sources contained within this bedrock unit. The Fall Line runs across the Midlands of South Carolina and divides the Piedmont and Coastal Plain geophysical provinces. At this location, the resistant crystalline rocks of the Piedmont straddle the unconsolidated Coastal Plain sedimentary rocks (Kovacik and Winberry 1987:18). In such an environmental setting, a frequent occurrence are rock outcrops and series of rapids that can stretch over a mile in length along river courses. The region referred to as the Fall Line is not easily discernible. Kovacik and Winberry (1987:18) have noted that the reason for this problem is that some rivers have cut through the sedimentary into the underlying crystalline rock, and rapids can shift locations during periods of high and low water. The geological formation of the Fall Line affords the archaeologist research challenges and opportunities not found in the neighboring Piedmont or Coastal Plain provinces. Narrow floodplains which contain bench-like terraces and levees were created mainly through vertical accretion of the crystalline hard-rock valley walls. An advantage in this Fall Line is that, with the right conditions, vertically broad and stable alluvial landforms are formed which contain overbank, levee-like terrace deposits. These landforms are well known for their depth and for their intact stratigraphic sequences, frequently spanning the entire Holocene period. The succession of pedostratigraphic units and their preservation is due to infrequent overbank depositional events, which also serve to seal discrete occupational components (Sassaman et al. 1990:34). Conversely, the same geomorphic traits can promote scouring and erosion during flood episodes. The result is a common phenomenon of near-channel lateral scouring and the flood-stage stripping or truncation of the upper portions of earlier floodplain deposits. The archaeological connotation is that such unstable surfaces have no integrity and are not likely to have been intensively occupied. Erosion artificially accelerated by historic land clearing and tillage has probably had a more profound impact on many of the region's soils and landscapes than the combined effects of the preceding millennia of natural processes. Tillage-induced erosion of sloping terrain results in the loss of soil material from the upper horizons of higher landscape positions with subsequent deposition of the eroded soil along lower toeslope positions. Much of the eroded soil may also find its way to stream systems, where, as transported sediment, it may ultimately be deposited in the alluvial settings of valley bottoms. These processes of erosion, transport, and deposition have not only been severely exacerbated by the influences of man, but they have also been so widespread that the floodplains and low-lying terraces of Piedmont stream systems are nearly everywhere mantled by appreciable deposits of historic-age alluvium. Rock types in the Fall Line are primarily metamorphosed sedimentary varieties, some interspersed with volcanic flows, consisting of argillite, rhyolites, and tuffs. In general, the Fall Zone resembles the Carolina Slate Belt formation, noted for its composition of amphibolite and argillite along with muscovite and chlorite schist. Minor portions of the Fall Zone consist of thick volcanic rock beds (Sassaman and Anderson 1994:14-15). The project area is underlain primarily by units of intrusive granite, metamorphic granite, mica gneiss, and amphibolite. The mica gneiss is of particular interest with respect to prehistoric land use because this rock type contains large amounts of quartz (Campbell 1982), which would have been utilized in prehistoric chipped-stone tool industries. Prehistoric populations might have also made use of mica sources contained within this bedrock unit. Coastal Plain physiographic province, which is the largest landform region in South Carolina, comprises approximately two-thirds of the state’s total area (Kovacik and Winberry 1987:18). Its topography varies from nearly flat and featureless to a rolling surface similar to the lower Piedmont (Kovacik and Winberry 1987:20). Elevations range from sea level near the coast to about 91 meters (300 feet) at the edge of the Sandhills (Kovacik and Winberry 1987:20). The sedimentary rocks that underlie the Coastal Plain are made up of muds, silts, sands, and other substances of marine origin. After deposition, these materials were consolidated to form shales, sandstones, conglomerates, and coquinas in horizontal layers. The oldest subsurface rocks in the Coastal Plain occur nearest the Piedmont margin; the youngest rocks occur adjacent to the coast (Kovacik and Winberry 1987:20). A distinctive feature of the South Carolina Coastal Plain are the Carolina bays, so-called because of the bay trees that characterize the vegetation found on their edges (Kovacik and Winberry 1987:21). The Carolina bays are generally oval or elliptically shaped, and resemble swamps with standing water and buttressed trees. They range in size from 1.6 to 2 hectares (4 to 5 acres), to thousands of acres (Kovacik and Winberry 1987:21). The bays’ axes regularly parallel each other and are oriented in a northwest-southeast direction. A sandy ridge may encircle a bay but often forms only on the southeastern rim. The bays are thought to have formed from shallow lakes during the middle-to-late Wisconsinan glaciation (Kreisa et al. 1997:5). Carolina bays are mostly likely to occur in the southeastern portion of the project area. A portion of the Carolina Sandhills, which are part of the Coastal Plain province, are located in the southeastern portion of Fairfield County. The Sandhills are the result of seacoast development as well as fluvial and aeolian deposition and consist of distinctive parabolic dunes and linear dune forms (Kreisa 1997:5). Dune formation is believed to have been initiated during the Pleistocene, but periods of aeolian activity in the Hypsithermal may also have contributed to dune development (Miller 1979; Thom 1970). The source for the dune sand appears to have been flood plain sediments or exposed sandy sediments on south-facing slopes (Thom 1970). The prehistory of the project area can be divided into several periods and subperiods (Table 2). These periods represent changes in culture, technology, and subsistence among human groups that inhabited southeastern North America since the end of the last glaciation at ca. 14,000 years ago. Table 2 - Archaeological Periods
3.1.1. Paleoindian (12,000-10,000 B.P.) Paleoindians represent the first known human populations to occupy the region. These groups encountered an evolving post-glacial landscape in the western portion of the project area, but may have encountered an essentially Holocene environment. Archaeological evidence indicates that humans arrived in the Northeast and the Southeast between 15,000 and 11,000 B.P.(Anderson et al.1996; Dent 1995). In general, the accepted date for evidence of Paleoindian occupation in the east is ca. 12,000 B.P., when the presence of Clovis fluted points occur in the archaeological record. Excavations at the Big Pine Tree Site at Allendale (South Carolina), Meadowcroft Rockshelter - 13,950 B.P. (Pennsylvania), Big Eddy - 12,950 B.P. (Missouri), and Cactus Hill (Virginia) - 16,000-15,000 B.P., however, suggest that human groups using a pre-Clovis technology may have been in the midwestern and eastern United States before the arrival of groups using Clovis technology (Carr et al. 1996; Fiedel 1999; Goodyear 1999; Ray et al. 1999). Rising temperatures at the end of the Pleistocene resulted in accelerated glacial retreat in the northern latitudes and a diachronic replacement of, and population shifts in, the flora and fauna of the region (Delcourt and Delcourt 1986). A rapid change in climate at this time resulted in a shift in frontal patterns, a retreat of the former boreal forest that had been established as early as 17,000 B.P., and the development of the mesophytic deciduous forest as a panregional phenomenon (Cox 1968; Craig 1969; Delcourt and Delcourt 1987; Jacobsen and Grimm 1986; Kneller and Peteet 1993; Peteet et al. 1993; Shane and Anderson 1993; Wilkins et al. 1991). A slight temperature reversal occurred between 10,800-10,000 B.P., resulting in an increase in boreal taxa (Gunn 1997; Peteet et al. 1993). The transition represented in the shifts in the distribution of plant taxa at the Pleistocene/Holocene boundary also is represented in the faunal record of the period. The results of archaeological excavations at sites in western and Midwestern North America demonstrate that in the late Pleistocene Paleoindian groups in those regions subsisted on megafauna such as woolly mammoth, mastodon, and bison (Graham et al 1981; Kelly and Todd 1988; Loy and Dixon 1998; Meltzer 1988; Purdue and Styles 1980; Tankersly 1990). While megafauna were dominant during the early Paleoindian period, most were extinct, extirpated, or began a northward re-adjustment in range by 10,000 B.P., and were replaced by contemporary faunal populations that inhabited the temperate deciduous forests of the southeastern United States (Mead and Meltzer 1984; Purdue and Styles 1980). Thirty-two genera of megafauna became extinct including herbivores, carnivores, and giant rodents (Graham 1979:62). The late arrival of human populations in the eastern United States near the end of the Pleistocene, and after the extinction or extirpation of Pleistocene fauna, meant that Paleoindian groups in the northeastern United States focused on migratory fauna (e.g., caribou), while groups in the Middle Atlantic and the Southeast focused on the exploitation of large, modern mammals (e.g., elk, deer, and moose) supplemented by foraged or collected plant and animal taxa (Carr et al. 1996; Custer 1990; Dent 1995; Gardner 1989; Kuehn 1998; Purrington 1983; Tankersly 1998). Some researchers believe that Paleoindian groups in the Southeast hunted Pleistocene megafauna before 10,900 B.P. (Anderson 1996:151). The Paleoindian period can be characterized as a time of extensive mobility, as small groups ranged across the landscape in search of food and high-quality cryptocrystalline rock or metavolcanic rock (Custer 1990; Daniel 1998; Gardner 1989; Goodyear 1979; Kelly and Todd 1988; Ward and Davis 1999). The preference for high-quality lithic raw materials have led some to suggest that Paleoindian groups were "tethered nomads" (Turner 1989), or that they employed cyclical settlement systems that were focused on lithic outcrops (Custer 1990:23). In low-biomass environments where preferred lithic resources were rare and where other resources may have been scattered rather homogeneously across the landscape, occupations located at or near quarries may have functioned as major base camps from which groups exploited other resources (Stevenson 1985). Evidence in support of this model of Paleoindian subsistence economy has been recovered at a series of Paleoindian sites in the Shenandoah River drainage (Gardner 1989). This analysis, and a re-evaluation of earlier work, indicates that the subsistence and settlement patterns of Paleoindian groups were characterized by small groups or bands that occupied a series of transient camps along the smaller, upland streams rather than in the broad bottom lands. These camps were relatively small and probably were occupied seasonally to exploit locally abundant and accessible resources. Larger sites, possibly base camps, existed in locations at which high-quality lithic resources also could have been exploited. These quarry-related base camps often were located near major streams or in large, open, river valleys. Raw-material analyses of fluted points and associated tools indicate that Paleoindian knappers had a proclivity for the use of high-quality, cryptocrystalline, lithic materials such as jasper, chert, and chalcedony. In the North Carolina Piedmont, however, metavolcanic rocks present in the Carolina Slate Belt were the most suitable raw material. Most of the Paleoindian tools in this region are made of rhyolite. Rhyolite outcrops are concentrated in the Uwharrie Mountains of Montgomery and Stanly counties, which are to the east of the project area (Ward and Davis 1999:38). The artifact type that characterize the Paleoindian period is the well-made fluted point that includes the fluted Clovis form (Early Paleoindian, 12,000-11,000 B.P.); Cumberland, Debert, Quad, Regan, and Suwannee (Middle Paleoindian, 11,000-10,500 B.P.); Plano-derived forms, Dalton, Meserve, and Holcombe Beach (Late Paleoindian, 10,500-10,000 B.P.). Fluted points were first discovered in the 1920s in the southwestern United States in direct association with the bones of extinct Pleistocene mammals such as mammoth and bison. They have not been recovered from similar contexts in the eastern United States (Dent 1995; Ward and Davis 1999). Goodyear et al. (1979:91) note that formal variability among Paleoindian point types may reflect chronological or spatial differences. The late Paleoindian period (10,500-10,000 B.P.), is represented by the Hardaway/Dalton phase, including a lithic assemblage of hafted bifaces, expedient bifaces, diversity of flake scrapers, flake chisel, flake blanks (Cable 1996:110). Terminal Dalton occupations in Tennessee date to 9790+/-160 B.P. (Broster and Norton 1996:294), and terminal Hardaway occupations to 9990+/-140 B.P. in Alabama (Driskell 1996:328). Side-notched forms such as Charleston and Big Sandy have been dated to 10,490+/-360 B.P., 10,330+/-120 B.P., and 10,345+/-80 B.P. in recent excavations in Tennessee (Driskell 1996:325-326). Similar dates were obtained for Charleston types at St. Albans in West Virginia (Broyles 1971; Niquette et al. 1992). Some researchers have placed Dalton/Hardaway in the late Paleoindian (Gardner 1989; Tankersly 1990), while other have established a separate cultural period (Purrington 1983). The Paleoindian tool kit also includes scrapers (especially distal- and lateral-edge unifaces), spokeshaves (concave unifaces), hammerstones, abraders, gravers, wedges [pieces esquillees], as well as multi-use flakes and bifaces (Gardner 1989). It is apparent from the artifacts recovered at the Flint Run Paleoindian Complex that two types of flake/core reduction technology were employed during the Paleoindian period: free-hand core and bipolar core (Gardner 1989:19). Blade core technology appears to be prevalent at Paleoindian and Early Archaic occupations in the eastern United States. Blades from conical blade cores often are found at Clovis occupations (Carr et al. 1996; Freeman et al. 1996; Morrow 1997:53), but blades usually "form only a small percentage of Clovis lithic assemblages" (Parry 1994:91). However, the paucity of blades at Paleoindian occupations may be a consequence of the production of end scrapers from snapped and broken blades (Freeman et al. 1996:389). Prismatic blades are a hallmark of the earliest Paleoindian occupations at Meadowcroft Rockshelter in Pennsylvania (Carr et al. 1996) and at "Pre-Clovis" sites, but Gardner (1989:19) noted that blade cores are completely lacking at Paleoindian occupations in the Shenandoah Valley of Virginia. In the mountains and foothills of western North Carolina, late Paloeoindian (Quad-like and Hardaway forms) artifacts have been recovered in the uplands, but the majority have been recovered in flood plain contexts in portions of McDowell and Rutherford counties, but outside the project area (Jurney and Downing 1974). One of the most important of Paleoindian sites is located in the Piedmont province of North Carolina. Archaeologists disagree about how old the site is, but most agree that it dates to at least the late portion of the Paleoindian period and that it also contains a significant Early Archaic component (Ward and Davis 1999:34-37). Joffre Coe’s excavation of the site began in 1937, but large-scale excavation did not begin until the 1950s (Coe 1964:57-60). An unprecedented volume of artifacts has been recovered by Coe and other archaeologists from the site (Daniel 1998:129). One other Piedmont project has resulted in the recovery of Paleoindian material. In the 1970s, large-scale block excavations were conducted at two sites outside the project area. Excavations resulted in the recovery of Hardaway-Dalton projectile points from stratified deposits (Cable 1996). Paleoindian site distribution studies in adjacent states (Anderson 1992a:75; Charles 1986:16) illustrate that sites cluster near the fall zone in South Carolina. Unfortunately, no early Paleoindian sites with intact stratigraphy and extensive assemblages have been excavated in South Carolina. Paleoindian artifact concentrations have been noted along major rivers bisecting the fall zone and in the Coastal Plain (Anderson 1992b:18, 34; Gunn and Wilson 1993:14), as well as in upland areas (Elliot and O’Steen 1987:144). A Late Paleoindian assemblage was recovered at the Taylor Site in Lexington County, located southeast of the project area (Michie 1996). While Clovis, Suwannee, and Dalton forms were recovered on the surface of the site, Late Paleoindian artifacts recovered during excavations at the site including Dalton forms, a Dalton adze, pieces esquillees, and blades (Michie 1996:254-263). Within the project area, Paleoindian fluted points have been recovered at eight archaeological sites in Burke, Caldwell, Iredell, and Mecklenburg counties in North Carolina. Clovis and other Paleoindian forms also have been recovered on the periphery of the project area in Ashe, Wilkes, Watauga, Madison, and Cherokee counties (Keel 1976; Purrington 1975, 1983:108; Reid and Lautzenheiser 1998). Late Paleoindian forms (Quad-like) have been recovered in a variety of environmental contexts in McDowell and Rutherford counties of the Blue Ridge foothills (Jurney and Downing 1974). In the Piedmont, fluted points and other Paleoindian forms have been recovered in several other counties that are near or within the project area including Burke, Cabarrus, Catawba, and Lincoln counties (Hargrove 1996:3; May 1989; Novick 1995:3.2). In South Carolina, Paleoindian points have been recovered in all five counties within the project area (Anderson and Sassaman 1992; Charles and Michie 1992; Elliot and O’Steen 1987). 3.1.2 Archaic (10,000-3000 B.P.) The Archaic period in the greater Southeast is typically divided into three subperiods: Early (10,000-8000 BP), Middle (8000-5000 BP) and Late (5000-3000 BP). A focus on riverine resources, and the presence of semi-sedentary or sedentary human groups, appears to have its genesis in the Hypsithermal. The focus of Middle Archaic prehistoric economies on the exploitation of riverine locales continued into the Late Archaic, although Late Archaic sites are frequently encountered in upland contexts. Early Archaic Period (10,000-8000 B.P.) In the Early Holocene, reduction in the moisture regimen appears in the palynological data as early as 10,000 B.P., when the cooling that occurred from 10,800-10,000 B.P. is reversed (Kneller and Peteet 1993; Shane and Anderson 1993; Wilkins et al. 1991). By 9000 B.P., westerlies increased in strength, frequency, and duration and temperatures were 2 to 30 degrees C warmer than present (Webb and Bryson 1972; Wendland 1980; Wright 1968). There was a collapse of the mesophytic forest at this time in parts of the southeastern United States, and they were replaced by oak-dominated forests in the Piedmont and Appalachians and by pine forests in the Coastal Plain (Delcourt and Delcourt 1987). The cultural sequence for the Archaic, including the Early Archaic, was established by Joffre Coe (1964) while working on several stratified archaeological sites in the Piedmont of North Carolina. Occupations that date to this period are marked by the presence of a variety of small, corner-notched formed hafted bifaces, such as the Palmer and Kirk corner-notched types, and steeply retouched unifaces (Coe 1964:120-122). Diagnostic artifact types of the Early Archaic also include Big Sandy side-notched and bifurcate forms (Anderson and Joseph 1988; Coe 1964:67-70). Evidence of the Palmer/Kirk phase (10,000-9000 B.P.) has been identified at archaeological sites throughout the eastern United States. This period is characterized by a marked increase in the use of non-local lithic materials, a decrease in number of sites, more sites in flood plain settings, and increased mobility (Carr 1998:49-50). Formed hafted biface types include Small Kirk Corner Notched, Large Kirk Corner Notched, and Kirk Stemmed. Kimball (1996:157-159), however, believes that there was no Palmer-Kirk-Pine Tree temporal order, but that formed hafted bifaces that are side-notched and have large hafts with ground excurvate bases (e.g., Charleston and Big Sandy) could be defined in Kirk phase levels, and that in upper levels small Kirk Corner Notched and large Kirk Corner Notched co-occurred with side-notched forms such as Palmer and Pine Tree. Dates for Kirk occupations range from 9928 B.P. at St. Albans in West Virginia (Broyles 1971; Niquette et al. 1992) to 9175+/-240 B.P. at Icehouse Bottom in Tennessee (Chapman 1977). A similar range of dates have been documented in Pennsylvania (Carr 1998), Virginia (Gardner 1989), and Kentucky (Creasman 1995). By 9000-8500 B.P., the Bifurcate phase (i.e., Kanawha, MacCorkle, St. Albans, and LeCroy types) was established and is characterized by sites in flood plains but more upland settings, intensified use of local lithic raw materials, and fewer formal flake tools (Carr 1998:50). Some researchers believe that bifurcates are concentrated in the Piedmont of North Carolina and South Carolina (Anderson 1995:46). But recent research indicates they occur throughout the Appalachian Summit, the Allegheny Plateau, and the Ridge and Valley provinces of the southeastern and Mid-Atlantic states. Kimball (1996:159) suggests that the types follow the following chronological order: St. Albans to LeCroy to Kanawha. Gardner (1989) and Stewart and Cavallo (1991) view bifurcates as dating to 8500 B.P., i.e., the Middle Archaic period. However, recent radiometric dates place the bifurcate tradition firmly in the period from 9500-8500 B.P. Bifurcates have been dated to: 8545+/- 80 B.P. in Kentucky for LeCroy types (Prentice 1992:49); 9219 B.P. at St. Albans in West Virginia for LeCroy types (Broyles 1971; Niquette et al. 1992); 9141-8998 B.P. at St. Albans in West Virginia for Kanawha types (Broyles 1971; Niquette et al. 1992); and 9420+/-90 and 9360+/-130 at Sands Eddy in Pennsylvania (Bergman et al. 1996). In North Carolina, a radiometric date of 7960+/- 90 B.P. was obtained for a mixed Bifurcate Phase/Kirk Phase stratum (May 1989:14). There is a debate underway relative to the settlement strategies used by Early Archaic groups in the Southeast. This debate has its genesis in the tremendous quantity of data generated by recent archaeological investigations in North Carolina, South Carolina, Tennessee, and Virginia. The models offer competing views of human behavior and decision-making among Early Archaic groups. Data derived from archaeological studies conducted on the South Atlantic Slope, Anderson and Hanson (1988) suggest that settlement strategies during this period varied on a seasonal basis. During the winter, Early Archaic groups followed a logistical strategy whereby populations were concentrated in residential camps and critical resources were obtained during forays to points where such resources were present. During the remainder of the year, the settlement/subsistence strategy was characterized by foraging, and residential sites were moved to resource locations as specific resources became available. The annual round would have included regular congregations of these smaller groups, probably in the fall, for information exchange and mate acquisition. Another characteristic of these annual rounds was their linear range between geophysical zones; groups roamed between the Atlantic coast in the winter through spring and the uplands of the Piedmont in the summer through fall. However, these movements were concentrated within major drainages that extended from the Piedmont to the Coastal Plain (Anderson 1996; Anderson and Hanson 1988). Finally, in the "terminal" Early Archaic, population growth lead to the circumscription of resource areas (Anderson 1996). Some archaeologists have questioned the validity of such a seasonally mobile model, given the rich diversity of natural resources in the Piedmont (Ward and Davis 1999:58). An alternative hypothesis has been that people where "tethered" to the Uwharrie Mountain rhyolite sources and moved between drainages in the southern Piedmont, i.e., the Uwharrie-Allendale Model (Daniel 1998:194-202). This model has been developed to explain site locations in North Carolina (Uwharrie Rhyolite) and South Carolina (Allendale Cherta). Similar models have been developed by Gardner (1989) for Virginia and by Carr (1998) for Pennsylvania. Other researchers are exploring the nature and function of Early Archaic sites, as diachronic change in settlement patterns and site types has been suggested for the Early Archaic. In Tennessee, Kimball (1996:184) notes that "Kirk and Bifurcate site distributions should not be lumped together as though they represent some homogeneous Early Archaic settlement system." In South Carolina, Sassaman (1996:73-75) has shown that Kirk phase sites in bottomland contexts include tool concentrations that represent specific activity areas and features such as fire-cracked rock (FCR) concentrations, debitage clusters, and deer bone concentrations. In these settings, "tool diversity is high," tools are made from locally available materials, and formal and informal tools are abundant (O’Steen 1996). In upland contexts, the artifact assemblages consist of exhausted hafted bifaces, debitage, and exhausted formal unifaces representative of "short-term hunting forays" (Sassaman 1996:75). Sites with a relatively large number of curated tools discarded and a low number of expedient tools are indicative of the bulk processing of resources, as the low number of expedient tools is the result of the short duration of the site use (Cable 1996:31). By the beginning of the Bifurcate Phase, "the proportional representation of diversity of expedient tool components increases dramatically" representing "a major shift in the functional use of the site" due to longer single occupations or more recurring occupations (Cable 1996:136). In McDowell County, North Carolina, examples of Palmer and Kirk types have been recovered in flood plain contexts, but outside the project area (Jurney and Downing 1974). In the Piedmont of North Carolina, excavations on stratified Early Archaic sites have been undertaken since the late 1940s. The most famous sites, Gaston, Doerschuk, and Hardaway, were excavated by Joffre Coe (1964). These sites produced the cultural sequence that "is the backbone of the Archaic period in North Carolina and throughout much of the eastern United States" (Ward and Davis 1999:51). These sites established the Palmer-Kirk-Stanly-Morrow Mountain-Guilford-Savannah River sequence. Excavations in the Upper Coastal Plain/Lower Piedmont along the Savannah River drainage have identified a few intensively occupied Early Archaic sites, including Lewis East and Pen Point in South Carolina (Benson 1994:21; Ledbetter et al. 1994:250-251). Most of these sites represent locations with intact stratigraphy and have yielded impressive artifact assemblages, with dense quantities of lithic debitage. One such Early Archaic site is Windy Ridge in Fairfield County (House and Wogaman - Windy Ridge). Archaeological surveys in York County resulted in the recovery of Early Archaic artifacts at several archaeological sites in highly eroded contexts (Brockington 1980), and in Kershaw and Lancaster counties, bifurcates (LeCroy Type) have been recovered at several sites in the land between Lake Wateree and the Lynches River. Twenty-six Early Archaic sites in the project area in North Carolina have been identified in Burke, Caldwell, Iredell, and Mecklenburg counties. In the portion of the project area in South Carolina, 27 Early Archaic occupations have been identified in Chester, Fairfield, Kershaw, Lancaster, and York counties. Middle Archaic Period (8000-5000 B.P.) By the beginning of the Middle Holocene (8000-4500 B.P.), a marked change had occurred in the moisture regimen throughout the United States, i.e., the Hypsithermal, also known as the Holocene Climatic Optimum and the Atlantic climatic episode. In the Northern Hemisphere, the summer days were longer and brighter and the winter days were shorter and dimmer, resulting in long, hot, and dry summers and long, frigid, snowy winters (Gunn 1997:143). The Hypsithermal maximum probably occurred at 7000-6400 B.P. with mean July temperatures 0.5 to 2.0 degrees Centigrade warmer, increased evaporation, a longer growing season, 10 to 25 percent less precipitation, and an increase in the occurrence of fires (Bartlein et al. 1984; Fredlund 1989; Wendland 1980). Zonal atmospheric patterns were dominated by the Pacific air mass (Delcourt and Delcourt 1987). As a result of moisture stress, the water table in many areas probably decreased significantly (cf. King and Allen 1977; Schwert et al. 1985), as did lake levels (Brugman 1980). In many areas, marshes and swamps began to dry up (King 1980; Kneller and Peteet 1993) and streams became choked with sediment (Gunn 1997; Voigt et al. 1998). The panregional nature of this event is well documented by pollen evidence that shows an expansion of xeric biotic communities and the shift from mesic to xeric taxa in the biotic communities of the Middle Atlantic region including Virginia (Fletcher et al. 1993; Gaudeau and Webb 1985; Gunn 1997; Kneller and Peteet 1993; Peteet et al. 1993), the Southeast (Watts 1979; Whitehead and Sheehan 1985; Wilkins et al. 1991), the Midwest (Butzer 1977; King 1980; Shane and Anderson 1993), and the northern United States and Canada (Anderson 1985; Brugman 1980; Bernabo and Webb 1977; Schwert et al. 1985; Wright 1992). In the central and southern Appalachian Mountains the incidence of pine increased, possibly as a function of the increased frequency of fires, forest-gap dynamics due to strengthening storms, and the presence of rocky soils where pine could out compete deciduous taxa (Delcourt and Delcourt 1985:20, 1986; Watts 1979:464). In these areas, the hemlock-oak forest was replaced by a mixed deciduous forest of oak, birch, hickory, beech, and American chestnut (Delcourt and Delcourt 1987:29-30). In the Piedmont, climatic conditions may have resulted in a Plains-like environment of scrub oak savannahs and pine forests of the Coastal Plain (Gunn 1997:145). With the amelioration of moisture stress that had been the hallmark of the Hypsithermal, a time transgressive replacement of xeric taxa by more mesic taxa began around the end of the Middle Archaic period (5000-4400 B.P.), i.e., the onset of the Late Holocene (Delcourt and Delcourt 1986). Hafted bifaces of this period are large, notched and stemmed forms such as the Stanly, Morrow Mountain (I and II), Guilford, and Big Sandy II (Coe 1964; Purrington 1983; Smith 1986; Steponaitis 1986). Artifacts representative of these types have been found in archaeological contexts throughout the southeastern United States. Sassaman and Anderson (1994:23) state that "Morrow Mountain Stemmed points are the most common diagnostic Middle Archaic artifacts in South Carolina." There appears to be an overlap between Morrow Mountain and Guilford in South Carolina (Kreisa et al. 1997). The common occurrence of ground stone mortars, pestles, manos, metates, nutting stones, grooved axes, and celts in Middle Archaic occupations suggests a pronounced involvement in plant harvesting (Ford 1977). This ground-stone technology serves as a distinctive hallmark between Early Archaic and Middle Archaic stone-tool assemblages (Chapman 1985). Innovations in lithic forms continued, and is reflected in the Middle Archaic material remains, e.g., the appearance of atlatl weights, netsinkers, and grooved axes. Middle Archaic groups appear to have depended on local lithic raw materials in the production of chipped stone tools (Purrington 1983:122). Relative to the preceding Early Archaic period, when settlement patterns encompassed broad linear territories that crosscut geophysical zones to exploit specific seasonal resources, Middle Archaic settlement patterns reflect limited movement between regions. Middle Archaic populations instead expanded their settlement ranges within geophysical regions and exploited more diverse resources. Settlement and artifact data from this period suggest "a strategy of small co-resident group size, frequent residential movement, generalized subsistence, low-investment technology, and social flexibility" (Sassaman et al. 1990:10). Sassaman (1988), as well as others (Blanton and Sassaman 1989; Sassaman et al. 1990), asserts that Middle Archaic populations were mobile and changed residential locations frequently to take advantage of specific resources as they became available. He suggests that tools used in resource procurement and processing were of an expedient type and were manufactured from local materials. Blanton and Sassaman (1989:61-62) have noted that there are more sites in interriverine zones, but the size and artifact density of sites are greater in floodplain settings. In addition, they have identified that in South Carolina more Middle Archaic sites are found in the Piedmont than in the Coastal Plain. Similar results have been noted for North Carolina (Ward and Davis 1999:63). In contrast, Goodyear et al. (1979) suggest that Middle Archaic populations grew increasingly less mobile in their settlement systems. According to these researchers, Middle Archaic groups established base camps along river floodplains. Available resources located in upland environments were exploited during forays from the residential camps in the river valleys. Since residential sites dating to the Middle Archaic period have not been documented in the region, the former model of settlement during this period is more strongly supported. Goodyear et al. (1979:111), however, point to the presence of storage pits and burials at sites in the region. Furthermore, a Middle Archaic component at Mims Point, along the Savannah River, yielded a relatively large assemblage of Morrow Mountain points, features, and a human burial (Sassaman 1993). The intensity of occupation reflected by this component may indicate a residential camp. Goodyear et al. (1979) also argue that the dependence on locally available raw materials is an indication of increased sedentism. Early Middle Archaic occupations are rare in the Appalachian mountains and foothills, but Morrow Mountain and Guilford forms have been recovered throughout the mountainous and foothills of western North Carolina (Purrington 1983:121-122). In an archaeological survey of portions of McDowell and Rutherford counties, Morrow Mountain was the most frequently encountered artifact type in upland and flood plain contexts, while Guilford types were recovered on lower slopes and on flood plains (Jurney and Downing 1974:110). Middle Archaic sites, represented by Morrow Mountain and Guilford types, have been identified in upland contexts (e.g., 31CW132 and 31CW140) in Caldwell County (Ayers and Kooiman 1996). Important Middle Archaic sites in the North Carolina Piedmont include the Lowder’s Ferry Site in Stanly County, and the Doerschuk Site, in Montgomery County, where Coe identified the hafted biface sequence for the Middle Archaic period (Coe 1964). Forty-five Middle Archaic occupations have been identified in the project area in the following counties in North Carolina: Burke, Caldwell, Gaston, Iredell, McDowell, and Mecklenburg, including a quarry site. In South Carolina, 43 Middle Archaic occupations within the project area have been identified in Chester, Fairfield, Kershaw, Lancaster, and York counties. Late Archaic Period (5000-3000 B.P.) The onset of the Late Holocene interval (4500 B.P.) resulted in "relatively unstable, century scale oscillations" in climate (Gunn 1997:146). With the beginning of the Sub-Boreal climatic episode (5000-2760 B.P.), xeric taxa were replaced by mesic taxa as temperatures remained cooler than modern temperatures and effective moisture increased (Fletcher et al. 1993; Hall and Lintz 1984; McMillan and Klippel 1981). The effects of the mid-Holocene climatic optimum were still felt in portions of eastern North America (Bhiry and Filion 1996:319); however, a mixed deciduous forest of American chestnut, beech, birch, hickory, and oak in the Allegheny Plateau and the Appalachian Mountains was established by ca. 4500 B.P. (Delcourt and Delcourt 1987:29-30). By 4000 B.P., the eastern United States could be characterized as the "deciduous forest region" with spring and summer moisture from the maritime Tropical air mass and autumn and winter weather dominated by the pacific air mass and the southern anticyclone (Delcourt and Delcourt 1987:100). This period is characterized by two phenomena: the re-expansion of spruce and fir in upper elevations of the Appalachians and the dominance of the oak-chestnut forest in the central and southern Appalachians (Barnosky et al. 1988:180-181; Delcourt and Delcourt 1985:20-21). With the increase in precipitation, flood frequencies in the eastern United States increased from 5000-3000 B.P. Evidence of shifting alluviation rates during this time period are seen in several regional examples from West Virginia, Kentucky, and Tennessee that date to between 4000-3000 B.P. (Bettis 1994; Brakenridge 1984:19; Broyles 1976; Voigt et al. 1998). Traditional descriptions of the Late Archaic focus on: (a) regionalization of hafted biface forms; (b) an increased diversity in tool forms; (c) the presence of items indicative of exchange networks; and (d) the diversity in Late Archaic site types. Ritchie (1932), Ford and Wiley (1941), and Caldwell (1958) hypothesized Late Archaic social complexity as restricted to small familial bands practicing seasonal transhumanence, i.e., a seasonal round, with economies that focused on efficient exploitation of forest resources, and whose material culture was unsophisticated and lacking ceramics. For example, highland areas would be used in the fall for nuts and game animals (Wilkins 1977), while valley bottoms would be used for larger aggregations at base camps and in spring and summer for fish, mussels, and plants (Cleland 1976, Gardner 1987). Subsequently, this notion has been modified as a consequence of archaeological excavations that have documented: regional trade networks (Winters 1969); fiber-tempered pottery (Reid 1984; Sassaman 1993); increasing mortuary complexity (Charles et al. 1986; Farnsworth and Asch 1986; Marquardt and Watson 1983; Warren and O'Brien 1982); and the presence of semi-sedentary and sedentary villages (Brown and Vierra 1983; Ritchie and Funk 1973). Taken together, the results of recent research indicate that Late Archaic groups were involved in a series of dynamic relationships with their cultural and natural settings. Stemmed forms of the Iddins/Ledbetter/Otarre cluster are found in the mountains and foothills of western North Carolina, often in association with the Savannah River Type (Keel 1976; Purrington 1983). Late Archaic sites in the Appalachian Summit date to ca. 4500 (Creasman 1995:Table 6-1; Pullins 1999:Appendix E; Voigt 2000). Savannah River is the dominate hafted biface type recovered in Late Archaic contexts in the Piedmont and Coastal Plain of North and South Carolina. Terms such as "Transitional Archaic" (Witthoft 1953) and "Terminal Archaic" (Faulkner and Graham 1966; Mouer et al. 1981) have been applied to differentiate Late Archaic groups at the onset of the Early Woodland. Prufer and Long (1986:35) state that transitional is "an unfortunate term" and Murphy (1975:117) notes that, "the Transitional Period is as poorly named as it is poorly understood". The terminal Late Archaic often is marked by the introduction of a ceramic technology and regional varieties of broad-bladed (i.e., "broadspear"), parallel-stemmed hafted bifaces, e.g., the "Savannah River Complex"(Coe 1964:123-124; Kinsey 1972; Ritchie and Funk 1973; Turnbaugh 1975; Witthoft 1953). Originally described by some as an adaptation to Coastal Plain habitats (Kinsey 1972; Mouer et al. 1981) and to have occurred solely in the Middle Atlantic and Northeast (Ritchie and Funk 1973:71), recent archaeological investigations have documented Late Archaic occupations with broadspear forms in forested and mountainous areas of Kentucky, North Carolina (Purrington 1983), Ohio, Ontario (Spence and Fox 1986), Virginia (Blanton et al. 1993), as well as West Virginia (Moxley 1983; Wilkins 1978, 1985). Use-wear studies of the broadspear form, initially proposed to have been a specialized tool type for use in riverine contexts, often served as a multifunction tool, much in the manner of other hafted bifaces in the Late Archaic. In the "Transitional Archaic," steatite bowls have been proposed as an integral part of the artifact assemblage in the eastern United States (Coe 1964; Kinsey 1972; Mouer 1991; Mouer et al. 1981; Ritchie and Funk 1973). However, recent AMS dates obtained on organic-residue samples from steatite bowls in collections indicate that many post-date the appearance of pottery (Sassaman 1993; personal communication 1996). In North Carolina, the introduction of ceramic technology follows Savannah River. At Doerschuk and Gaston, the Late Archaic Savannah River occupation was separated from the ceramic-bearing Early Woodland Badin occupation by sterile alluvial sediments (Coe 1964; Ward 1983). A different history in the development of ceramic technology occurs in South Carolina. Fiber-tempered pottery has been identified in coastal South Carolina as early as ca. 4200 B.P. (Sassaman 1993). Sand-tempered Stallings Island pottery is the earliest ceramic type in South Carolina, but is restricted to the southern portion of the state. This ware is tempered with fiber and occasionally exhibits surface treatment, such as punctuation and incising. Sand-tempered Thoms Creek pottery represents a later ware and displays the same kinds of surface treatments, and is found in the Coastal Plain and Piedmont of South Carolina (Price 1992; Sassaman 1993). As Custer (1989) notes, "one of the striking characteristics of Late Archaic regional settlement patterns is the appearance of large base camp sites in the major riverine" settings, a view supported by the presence of relatively large sites, "base camps," that usually are found in bottomland contexts. These types of sites also are found in bottomland contexts along the tributary streams of rivers, as has been demonstrated at recent excavations in Alabama, Georgia, Kentucky, South Carolina, and Tennessee. These sites often are characterized by extensive midden deposits, numerous cultural features, and a wide range of material remains (Adovasio 1982; Jeffries 1990; Winters 1969) that may have been occupied "year round" (Ritchie and Funk 1973:41-44). Braun (1987) believes that riverine contexts with their concentrations of aquatic and wetland resources offered a resource base that, when complemented with the use of cultivated and domesticated plant taxa, "made it possible for opportunistic groups to begin a more intensive, sedentary use" of these habitats. Caldwell (1958) defined the Late Archaic as a period in which prehistoric groups practiced "primary forest efficiency," i.e., they practiced "broad-spectrum" (Flannery 1968) or diffuse (Cleland 1976) subsistence strategies. While the subsistence economies of Late Archaic groups in the eastern United States apparently exhibited a stronger orientation to riverine resources (Mouer 1991; Phillips and Brown 1983; Ritchie and Funk 1973; Winters 1969), recent research suggests that between 4000-3000 B.P. these groups relied on the cultivation of domesticated native plant taxa (Fritz and Smith 1988; Voigt and Pearsall 1989). Smith (1992:49) demonstrates that the period from 4000-3000 B.P. "brackets the earliest evidence of morphological changes reflecting domesticated status in all three seed crops [chenopod, sumpweed, and sunflower] brought under domestication... in the East." In addition, Cucurbita also was used by both Middle and Late Archaic groups (Chomko and Crawford 1978). Therefore, among some Late Archaic groups in the upper Ohio River Valley, the cultivation of natural and domesticated taxa occurred during a critical juncture in the development of prehistoric agricultural systems. An example of the composition of a Late Archaic floral component was documented at Cold Oak Rockshelter in Kentucky. Gremillion and Ison (1992) document the evidence for the use of nuts and seeds by the prehistoric occupants of this shelter (Figure 44). The floral assemblage from the Terminal Archaic occupation (2930+70 B.P. and 2830+60 B.P.) included acorn, chestnut, hickory, and walnut nutshell; domesticated Iva annua achenes; a sunflower achene; Cucurbita rind; and goosefoot, erect knotweed, maygrass, and ragweed seeds (Gremillion and Ison (1992:122). Aquatic resources were important to Late Archaic economies among groups that lived along the coast and the major rivers of the Southeast, (Ward and Davis 1999). Sassaman (1993) notes that shellfish constituted a principal food source, although turtles were also important resources. Deer and a variety of other terrestrial fauna were also consumed. Sassaman (1993:120-121) also cites evidence that marine resources were utilized along the Middle Savannah River. Such resources likely included anadromous fish that were obtained by inland populations, although it is possible that marine resources were acquired through trade with coastal populations. Between 2800-2400 B.P., at the end of the Sub-Boreal climatic episode and the onset of the Sub-Atlantic climatic episode (2760-1680 B.P.), another change occurred in the vegetation of the region. This appears to have little to do with the onset of the Sub-Atlantic climatic episode, which was a period of climatic warming (Fletcher et al. 1993). Instead, interpretation of pollen diagrams strongly suggest that prehistoric Native American groups apparently were engaged in land-clearing activities for agricultural purposes around 2800 B.P. in Tennessee (Delcourt et al. 1986), by 3000 B.P. in Kentucky (Delcourt et al. 1998), and by 2400 B.P. in Mississippi (Whitehead and Sheehan 1985:134). At Gallipolis Locks and Dam, the pollen record after 2200 B.P. is characterized by relatively high values for Ambrosia (ragweed), which Fredlund (1989:21) interprets as further evidence for agricultural activity. The evidence for land-clearing by prehistoric Native American groups often is documented in the pollen record in one of 2 ways. Delcourt and Delcourt (1985:21) base their argument for land clearing on "...high percentages of ragweed (Ambrosia type) pollen, along with pollen of herbs indicative of disturbed, open ground (Chenopodium types [goosefoot], Iva [sumpweed], Portulacaceae [purslane], Plantago spp. [plantain], and Rumex [dock])". In Kentucky, changes in forest composition and an increase in wood charcoal abundance and in size of wood charcoal remains in Late Holocene deposits at Cliff Palace Pond appears to be indicative of land-clearing activities (Delcourt et al. 1998). Plant associations were again relatively stable until the Neo-Atlantic Climatic episode ca. 1250 B.P. The distribution of Late Archaic Savannah River sites exhibit different patterns in the Appalachain Summit. In some areas, sites with Savannah River components are located along ridges and high gaps and saddles in mountainous areas, but in the broad river valleys sites are located in flood plain contexts near quartzite outcrops (Purrington 1983:126-127). Multicomponent Archaic sites that include Early, Middle, and Late Archaic components have been excavated in Watauga County, adjacent to the project area (Arey et al. 1996). Radiometric data of 3640+/-60 B.P. was obtained on wood charcoal from a Late Archaic hearth (Arey et al. 1996:ii). A Savannah River component was excavated at Warren Wilson on the Swannanoa River in Buncombe County (Keel 1976). Several Late Archaic occupations were identified during a survey of Watauga County, adjacent to the project area (Purrington 1975). In the North Carolina Piedmont, such points became smaller through time. A variety of scrapers, drills, grooved groundstone axes, hammerstones, netsinkers, polished atlatl weights, and stone mortars are also common (Ward and Davis 1999:64). Subsistence data from the Late Archaic period is lacking from Piedmont sites. However, the location of base camps, such as Doerschuk, Lowder’s Ferry, and Gaston, with middens indicate a preference for living along major waterways. A variety of animals, including fish, turtles, migratory birds, white-tailed deer, bear, and small mammals, would have existed in such an environment. Also, wild fruits, nuts, and berries would have been gathered on a seasonal basis (Ward and Davis 1999:67). Extensive research in South Carolina has used a broad database to characterize Late Archaic settlement patterns. Piedmont sites of this period are situated in a variety of environmental settings, but upland locations are typically small, diffuse, lithic scatters reflective of short-term extraction sites. Riverine sites are larger in both size and artifact density. During the occupations of these larger camps, resources found in riverine settings were utilized extensively. A variety of upland environments was also visited during forays from residential camps to obtain necessary resources (Sassaman and Anderson 1994:135). In the Sandhills region of South Carolina, sites of the period are located in all environmental settings and include a variety of types, such as large residential sites along river valleys, upland habitation sites, and limited-activity sites in valley and upland locations (Sassaman 1988). Price (1992) points out that it is not clear whether these patterns existed in the Piedmont. Sassaman (1993), however, summarizing data from the Middle Savannah River Valley, notes that evidence from the region suggests that sites along the river were occupied from spring through fall. During the occupations of these larger camps, resources found in riverine settings were extensively exploited. Winter habitations were most likely located in upland zones and probably contained smaller groups of occupants than were present at the sites in the valley. A variety of upland environments were also visited during forays from residential camps to obtain necessary resources. These data suggest that settlement patterns observed in the Sandhills also prevailed in the Piedmont. Within the project area, archaeological sites with Late Archaic occupations (N=33) have been identified in Burke, Caldwell, Gaston, Iredell, McDowell, and Mecklenburg counties. Archaeological sites with Late Archaic occupations (N=30) have been identified in the project area in the following South Carolina counties: Chester, Fairfield, Kershaw, Lancaster and York. 3.1.3 Woodland Period (5000-500 B.P.) Until 950-850 B.P. (i.e., the onset of the Pacific climatic episode), the post-Hypsithermal development of the eastern deciduous forest continued. However, at this time a significant climatic reversal took place, one that is documented by the appearance of a vegetational discontinuity (a shift in the relative percentages of co-occurring taxa) in the pollen record (Bernabo 1981; Gregg 1975; Wendland and Bryson 1974). Increased penetration by Pacific air masses during summer months may have resulted in drought conditions and higher temperatures. Bernabo (1981) states that the period from 950-600 B.P. was the warmest time during the past 2000 years. These conditions persisted until the Sub-Boreal climactic episode (400-100 B.P.). The onset of the Neo-Boreal, a period of cooler and moister conditions, is evident in pollen records beginning ca. 500-400 B.P. (Kline and Cotam 1979). Cool and moist conditions probably peaked between 350-200 B.P. (Bernabo 1981). A warming trend began around 200 B.P., and by the mid-nineteenth century, a warmer and drier climatic regime had been established. At the time of Euroamerican settlement of the project area, grasslands occurred in glades and along river bottoms (Jurney et al. 1948:9). Regional differences in material culture that gained expression in the Late Archaic, are even more apparent in the pottery assemblages from Woodland occupations at archaeological sites. Beginning in the Early Woodland period, it is apparent that human groups in North and South Carolina shared similar lithic and ceramic technologies, but it also is clear that there was growing stylistic differentiation among pottery wares from the same cultural period. In many instances, where groups were located in a physiographic provinces determined the character of their material culture assemblages, as well as the structure of their societies and economies. This regionalization along physiographic divisions is reflected in the discussions below of the Early-Late Woodland periods, as the focus shifts from the Appalachian Summit, to the Piedmont, and to the Coastal Plain.
Early Woodland Period (3000-1500 B.P.) A number of sequential cultural complexes or phases have been distinguished for the Early Woodland period (3000-2000 B.P.). In North Carolina, the Early Woodland phases include the Swannanoa, the Badin, and the Yadkin phases, although the exact chronological relationship between them is not well understood (Ward and Davis 1999:84). In South Carolina, the phases include the Stallings Island phase, the Thoms Creek phase, the Refuge phase, and the Deptford phase (Trinkley 1990). In North Carolina, as well as South Carolina, only a few sites have been excavated and there is little information on settlement patterns or subsistence (Anderson and Joseph 1988; Trinkley 1990:16; Webb and Leigh 1995). In North and South Carolina, diagnostic Early Woodland projectile points include Swannanoa Stemmed, Plott Short Stemmed, and Gypsy Stemmed as well as a number of triangular forms (Anderson and Joseph 1988). Additional artifact types include soapstone pipes, boatstones, bar gorgets, biconcave mortars, and manos (Sassaman et al. 1990:12). The overall settlement and subsistence pattern from the Late Archaic period changed very little during the Early Woodland and Middle Woodland periods; however, there were a few significant innovations. Bow and arrow technology completely replaced the use of the atlatl. Village sites in the Piedmont are small, and suggest occupation either by small groups or of short duration (Sassaman et al. 1990:13). Early Woodland settlement is characterized by residential camps located in riverine environments. Limited-activity camps were located in the adjacent upland zones. From the coast, to the inner Coastal Plain/Sandhills, to the Fall Line, Early Woodland sites often are found on low sandy ridges near water, on Carolina Bays, and along streams and rivers (Kreisa et al. 1997:24). In the Blue Ridge and foothills of western North Carolina, Swannanoa and Watts Bar wares represent the initial introduction of ceramic technology (Ward and Davis 1999). As Lafferty (1981:307) points out, there is confusion as to what is Watts Bar pottery and what is Swannanoa." It is suggested that the Swannanoa phase dates to 2650-2250 B.P. in the Appalachian Summit (Purrington 1983:132). Swannanoa sherds have been recovered in securely dated contexts at the Wheeler Site, in Virginia that have been radiometrically dated to 2840+/-70 B.P. and 2640+/-90 B.P. (McLearen 1994:52) and to ca. 2900 B.P. in northeastern Tennessee (Ward and Davis 1999:142). Swannanoa vessels have crushed quartz temper or coarse sand temper and Swannanoa varieties include Fabric Impressed, Cordmarked, Simple Stamped, Check Stamped, and Plain and vessels forms include large amphora and simple bowls, usually with straight or vertical rims (Keel 1976:50). Pottery for the Early Woodland in the North Carolina Piedmont includes Badin and Yadkin wares, and the sequence was defined first at the Doershuk Site in the North Carolina Piedmont (Coe 1964; Davis 1996). These wares exhibit influence from the north and south but lack clear association in terms of style and stratigraphic location (Ward and Davis 1999:97-98). Badin sand-tempered ware dates to 2400-2000 B.P. at several sites in North Carolina and is found in association with Badin triangular points (Blanton et al. 1986; Webb and Leigh 1995:28-29). Yadkin consists of cord-marked and fabric-marked varieties, as well as new surface treatments: check stamping, linear check stamping, and simple stamping which derive from Deptford wares of South Carolina and Georgia (Ward and Davis 1999:83). Yadkin sherds are tempered with crushed quartz and have been recovered in association with Yadkin triangular and Yadkin eared points (Blanton et al. 1986; Coe 1964). Yadkin pottery has been dated to 2580-2170 B.P. in Sumter County, South Carolina (Blanton et al. 1986:167). In the Coastal Plain of South Carolina, Early Woodland wares include Thoms Creek/Refuge wares, sand-tempered pottery with dentate stamped varieties, found in association with Swannanoa, Plott, and Pigeon types of formed hafted bifaces (Haile Gold Mine 1993:8). Deptford ware (3050-1350 B.P.) is a check-stamped fine-to-coarse sandy paste pottery. To the north and west, a few sherds of Badin ware, including cord-marked, fabric-impressed, net-impressed, and plain varieties, have been recovered with Badin triangular points (Trinkley and Campo 1999:14). In Sumter County, South Carolina, Badin sherds have been dated to ca. 2300 B.P. (Webb and Leigh 1995:28).
Middle Woodland (2000-1300 B.P.) Middle Woodland settlement patterns in the Piedmont are not well documented. According to Trinkley (1990:21), it is unclear whether the riverine focus of Early Woodland populations continued or whether Middle Woodland populations made greater use of interriverine areas. Anderson and Joseph (1988) suggest that Middle Woodland sites indicate residential mobility, and that populations may have moved on a seasonal basis to take advantage of specific resources as they became available. Sassaman et al. (1990:13) state that settlement included base camps, positioned to maximize access to diverse resources, and limited-activity extractive camps that were occupied for short periods. Middle Woodland subsistence strategies continued the reliance on wild foods. There is no clear evidence of the use of cultivated plants. Food production was intensified during this period and supported locally concentrated population aggregates. Large-scale storage is also evident (Anderson and Joseph 1988; Sassaman et al. 1990:13). The Middle Woodland (2100-1300 B.P.) occupation of the site consists of the Pigeon Phase and the Conestee Phase. Pigeon pottery is characterized by crushed-quartz temper, check stamping, the use of tetrapodal supports, and a sheen on the vessel surface (Ward and Davis 1999:146-147). Ward and Davis (1999:146) also point out that a "pure Pigeon component has not yet been isolated on a site" in North Carolina. While Swannanoa ware exhibits similarities to wares form the north, Pigeon phase pottery and the use of carved wooden paddles that were used to decorate the surfaces of vessels exhibit ties to Deptford wares of the south (Ward and Davis 1999). Connestee Triangular formed hafted bifaces and Connestee Plain, Cordmarked, and Simple Stamped pottery are hallmarks of the latter. Connestee phase occupations occur throughout the Appalachian Summit (Keel 1976; Purrington 1983; Schroedl et al. 1990). This very fine-to-medium sand-tempered ware has been recovered in Middle Woodland (1750-1150 B.P.) contexts in North Carolina, Georgia, and Tennessee (Keel 1976:Appendix; Purrington 1983:137; Ward and Davis 1999:146-150). Connestee phase groups in Tennessee and North Carolina appear to have had contact with Hopewell groups in Ohio, as evidenced in artifacts recovered at Garden Creek Mound No. 2, e.g., anthropomorphic and zoomorphic clay figurines, prismatic blades, copper sheets, beads, Connestee Triangular hafted bifaces, and imported pottery (Keel 1976:117-149). Purrington (1983:138) states that "Connestee ceramics show a marked shift from the strong influences from the south...to a less spectacular series which reflects influences from the west and northwest." The Connestee phase may continue to 950 B.P. in areas of the Appalachian Summit (Ward and Davis 1999:155). In Tennessee, Connestee Ware has been found in association with Candy Creek Ware (Ward and Davis 1999:153). Connestee phase pottery has been recovered at archaeological sites in Lee County and other areas of southwestern Virginia (Egloff 1987:8, 18-19). Types of Connestee Ware include Simple Stamped, Brushed, Plain, Cordmarked, Fabric Impressed, Check Stamped, and Complicated Stamped (Keel 1976:48). Vessel forms include shouldered jars, straight-sided jars, globular jars, and simple open bowls with vertical or flaring rims and some vessels have tetrapodal supports (Keel 1976:49). Vessels are relatively thin walled (Purrington 1983:137; Ward and Davis 1999:151). Some Swift Creek complicated stamped sherds also have been recovered at Connestee sites in North Carolina. Swift Creek is a Middle Woodland ware from Georgia (Ward and Davis 1999:155-156). Middle Woodland settlement patterns in the Piedmont province are not well documented. According to Trinkley (1990:21), it is unclear whether the riverine focus of Early Woodland populations continued, or whether Middle Woodland populations made greater use of interriverine areas. Anderson and Joseph (1988) suggest that Middle Woodland sites indicate residential mobility, and that populations may have moved on a seasonal basis to take advantage of specific resources as they became available. Sassaman et al. (1990:13) state that settlement included base camps positioned to maximize access to diverse resources, and limited-activity extractive camps that were occupied for short periods. Late Woodland Period (1300-500 B.P.)
Late Woodland groups in North and South Carolina followed different trajectories in their development. In some areas, an essentially Middle Woodland lifeway continued until the Mississippian period, while in other portions of the states, Late Woodland groups developed complex social systems and agricultural economies (Trinkley and Campo 1999:15). In some portions of the project area, Late Woodland cultures persisted to the time of European contact in the sixteenth to eighteenth centuries. In other areas, Late Woodland culture was subsumed into the South Appalachian Mississippian Tradition (see below). In the Appalachian Summit, Cane Creek wares consist of a relatively large number of plain vessels, all without the tetrapodal feet characteristic of Middle Woodland wares (Ward and Davis 1999:157). While there are similarities to the Connestee phase wares of the Middle Woodland, the artifact assemblage is similar to those recovered at Uwharrie Phase sites in the Piedmont of North Carolina (Ward and Davis 1999:158). Several Late Woodland phases that are part of the Piedmont Village Tradition, based on diagnostic ceramic wares, have been identified in the Piedmont of North Carolina (Ward and Davis 1999). All of these groups were located to the north and east of the project area. These include the Uwharrie phase (1150-750 B.P.), the Haw River phase (950-550 B.P.), the Dan River phase (950-500 B.P.), the Donnaha Phase (950-500 B.P.), the Hillsboro phase (550-350 B.P.), the Early Saratown phase (500-350 B.P.) (Ward and Davis 1999). These phases appear to be the remains of Siouan-speaking groups ancestral to the tribal groups encountered by European explorers, and reflect even greater regionalization within physiographic provinces. For example, the Dan River and Saratown phases of the central and northern Piedmont may represent the remains of "peoples ancestral to the Sara Indians," while the Hillsboro phase of the north-central Piedmont may be related to the Eno, Shakori, and Occaneechi tribes (Ward and Davis 1999:99). In the southern Piedmont, the Pee Dee culture was identified first at the Town Creek Site in Montgomery County, North Carolina. The Pee Dee occupation is represented by a palisaded village that included a habitation area, a central plaza, and a temple mound (Ward and Davis 1999:123-125). A total of 563 burials were excavated at the site. Pee Dee components have been identified at village sites and show that the Pee Dee culture covers a much larger period than that evidenced by the Town Creek Site. A significant change from earlier periods was the introduction of maize agriculture, which was the most important component of the diet of the inhabitants at the Leak Site. Subsistence was supplemented by more traditional natural resources (Ward and Davis 1999:131-132). Pottery consists of bowls and jars with complicated stamped surfaces as well as vessels decorated with the filfot-cross design (Ward and Davis 1999:126). Based on the pottery and the character of the village/mound complex, it appears that Pee Dee culture is a localized manifestation of the Southern Appalachian Mississippi Tradition. In South Carolina, Trinkley (1990:21-22) suggests that little change in adaptations occurred between the Middle Woodland period and the development of the South Appalachian Mississippian complex. Thus, the Late Woodland period may be considered an extension of the preceding era. Trinkley suggests that the Piedmont may have represented a buffer between competing groups, or it was used for resource procurement by different groups, but inhabited by neither, as was the case during historic times. Prehistoric activity in the region would thus be represented by relatively sparse archaeological materials (Trinkley 1990:24). Trinkley, citing Anderson and Joseph (1988), notes, however, that the Cartersville and Connestee pottery types that are typically assigned to the Middle Woodland period may have persisted into the Late Woodland. Thus, sites ascribed to the earlier period may actually represent later occupations. Anderson and Schuldenrein (1985:720) suggest that the first evidence of intensive utilization of floodplain settings appears during the Late Woodland. Such occupation is marked by pits, hearths, posts, and scatters of shell. Trinkley (1990) states that few indications of agriculture are known from this period in South Carolina, and if they were available, domestic plants comprised an insignificant proportion of the subsistence base. In fact, based on analysis of sites in the Piedmont, Coe (1964:51) indicated that agriculture was not practiced in the region prior to AD 1000. Settlement in the Piedmont appears to have been concentrated along tributary streams, and reflects little utilization of the interriverine areas (Goodyear et al. 1979). Sites of the Late Woodland period were small and widely dispersed (Coe 1964). There are 340 sites with Woodland components in the project area. There are 312 sites in North Carolina, and Woodland occupations have been identified in every county. Site types include habitation, village, mounds, and lithic workshops. In contrast, only 28 Woodland occupations have been identified in South Carolina, and most of the sites contain Woodland, Mississippian, and Catawba components. 3.1.4 Mississippian Period (950-500 B.P.)
During the Mississippian period, complex, chiefdom-level societies developed in the southeastern United States. This period is characterized by large village sites located on floodplains, as well as earthen mounds, settlement hierarchy, evidence of social stratification, and an economy based on agriculture. In addition to these extensive sites, a second category of low-density sites is evident that represents specialized procurement or hunting locations. Within the project area, the South Appalachian Mississippian Tradition is the regional manifestation of Mississippian societies. South Appalachian Mississippian sites have been identified as far north as Lee County, Virginia, as far west as Knox County, Tennessee, as far south as Oconee County, South Carolina, and as far east as McDowell County, North Carolina (Dickens 1976:16). In the Appalachian Summit, Mississippian sites range from small farmsteads to large palisaded villages, often with small sites located in the vicinity of villages with mounds (Ward and Davis 1999). The villages, surrounded by palisades, were located "along major streams and in the tributary valleys, on or adjacent to fertile bottomland soils" with houses in a circular or oval pattern around a central plaza (Dickens 1976: 94-96). Mississippian economies focuses on deer, maize, beans, squash, sumpweed, acorns, hickory nut, walnut, and butternut (Ward and Davis 1999:171). Hallmarks of Mississippian sites include ceramic types that are distinguished on the basis of elaborate decorative motifs and rim treatments. These complicated stamped ceramics contrast with the plain, cordmarked, fabric-impressed, and simple stamped ceramics that characterized the preceding Woodland period (Anderson 1989). Anderson (1989:115) notes that, in addition to signaling Mississippian occupations, the rim treatment of Mississippian pots constitutes a sensitive chronological attribute. He points to a progression of unmodified to collared rims, to rims with rosettes or punctuation, and finally to applied and pinched rim strips. This sequence prevails through most of the region. In the Appalachian Summit, the Pisgah phase represents the local manifestation of the Southern Appalachian Mississippian Tradition, and it characterizes the "climax of Mississippian influence in the Appalachian Summit" (Keel 1976:45). The Warren Wilson Site (Buncombe County) and the Garden Creek Site (Haywood County) have been the focus of excavations that have produced the information that defines the Pisgah phase (Dickens 1976; Keel 1976; Ward and Davis 1999). Pisgah ceramics have been recovered at archaeological sites in North Carolina, South Carolina, Tennessee, and Virginia. The presence of town-and-mound complexes, shaft-and-chamber burials, Mississippian pottery, and other aspects of Mississippian material culture indicates that Mississippian cultural influence in the area was significant and that it can not be viewed as a mere cultural veneer adopted by local groups. Lithics from the period are characterized by the small, isosceles Pisgah Triangular point. Other implements include microtools, gravers, perforators, drills, scrapers, groundstone celts, pipes, discs, and a shell industry that included gorgets, ear pins, beads, and dippers (Purrington 1983:142). While most pottery has fabric-impressed, cordmarked, or smoothed surface treatments, some Pisgah surface treatment appears to be derived from the rectilinear stamped pottery of the Connestee phase (Dickens 1976:13; Purrington 1983:143). Pisgah ware usually has a fine-to-coarse sand temper and the basic vessel form is a globular jar with an everted rim (Dickens 1976:273-274). The rims of vessels often are thickened and "are highly decorated with parallel rows of short diagonal punctuation" (Egloff 1987:12), which is a hallmark of Pisgah ware (Dickens 1976:178). This form of rim treatment has "no precedent in western North Carolina or the surrounding area," but similar forms are present in Indiana, New York, and Ohio (Ward and Davis 1999:166). Pisgah Plain consists of the same vessel forms, but sherds have a smooth exterior surface (Dickens 1976:185-186). Pisgah phase habitation sites consist of small farmsteads and relatively large village/mound complexes, usually located in flood plain contexts, while logistical sites are located in a variety of environmental contexts (Dickens 1976; Purrington 1983). Burials were in or near houses in a simple pit, in a chamber, or in a shaft and chamber grave (Dickens 1976:102-103). The Qualla phase (after 650 B.P.) in the Appalachian Summit is the localized expression of the Lamar culture of the southeastern United States, e.g., northern one-half of Georgia and Alabama, South Carolina, eastern Tennessee, and one-third of North Carolina. In North Carolina, Qualla is divided into Early (500-300 B.P.) and Late (300-100 B.P.). Sites are located in the Little Tennessee and Hiwassee drainages to the south and east of the project area, whereas Pisgah sites are located to the east of the Tuckasegee drainage (Ward and Davis 1999:178-180). This is the homeland of the Cherokee (Dickens 1976). During the late Qualla Phase, the Cherokee were relatively isolated until the early eighteenth century (Ward and Davis 1999:267). The Cherokee fought against the French, but they also remained allies of the British during the Revolutionary War. As a consequence, they were subjected to repeated attacks, and beginning with the Treaty of Hopewell (1785) and "culminating with the Removal of 1838, each new treaty between the Cherokee and the newly formed United States cost them more and more of their mountain homeland" (Ward and Davis 1999:268). In South Carolina, the new sociopolitical structures may have been imposed locally by elites immigrating into the region from the west (Sassaman et al. 1990:15), although the transition between the Late Woodland and Mississippian periods is not well understood (Anderson 1989). Many of the larger sites have been recorded within the Fall Zone environment, including Hollywood and Mason's Plantation mound groups near Augusta, Georgia; Ft. Watson on the Santee River in South Carolina; Mulberry Mound near Camden, South Carolina and near the project area; and Town Creek along the Pee Dee River in North Carolina (Benson 1994:27; Gunn and Wilson 1993:20). Settlement patterns exhibit hierarchies of site types consisting of mound centers, villages, hamlets, and isolated farmsteads (Sassaman et al. 1990). Villages were located on terraces and levees of major drainages. Isolated hamlets also occurred along the region's watercourses (Anderson 1989:114). Occupation of the interriverine settings has not been fully documented for the Mississippian period; however, some small camps have been noted at the headwaters of streams in the Piedmont. As increasing population pressure resulted in warfare between aligned native groups, villages were fortified with palisades. Mississippian culture represents the foundation for protohistoric Cherokee groups met by European explorers and traders during the mid-sixteenth century. Currently, no recognizable artifact assemblage fits the period from AD 1450-1600, which supports the hypothesis that the Middle Savannah River Valley was abandoned during this period (Anderson 1990:445). The economic basis of these developments involved intensive maize agriculture (Anderson 1989:113). Sassaman et al. (1990) note that maize was being grown locally prior to the implantation of Mississippian-influenced culture. Hunting and the gathering of wild foods supplemented the supply of domesticated foods (Price 1992). At the southern reaches of the project area, Mississippian sites occur frequently on the Wateree River, with nine major mound sites around Camden (Stuart 1967, 1970). DePratter and Judge (1990:56-58) have defined six phases of Mississippian occupations on the Middle Wateree, each defined by particular pottery varieties. The Belmont Neck phase (38KE6)(750-700 B.P.) is linked with the Etowah Mississippian of Georgia and is marked by complicated-stamped vessels without rim strips, but notched rims, and concentric circles or concentric curvilinear surface decorations. The Adamson phase (38KE11) (700-650 B.P.) shares certain traits with Savannah II Mississippian of the South Carolina Coast, and is manifest in complicated stamped, plain, simple, vessels with notched and punctated rims. The filfot cross is the major decorative element, and is found on large and small globular vessels and shallow bowls (see also Stuart 1967). The Town Creek (600-550 B.P.) is the localized expression of Lamar culture. Pottery decoration is similar to that of the Adamson phase, but includes complicated stamped surface decorations, segmented and punctated applique strips, rosettes, and nodes. All of these traits have direct parallels with the Lamar series and Savannah series of Georgia (Kreisa et al. 1997:23). The McDowell phase (38KE12)(also known as the Mulberry Site)(600-500 B.P.) also is tied with the Lamar culture, but the pottery has complicated stamp designs with a larger and a bolder filfot cross. The Mulberry (500-400 B.P.) is another localized expression of the Lamar culture. Vessels are decorated with complicated stamping and Lamar-like incising, as well as segmented and punctated applique strips and shoulder decorations. The final expression of the Lamar culture in the area occurs in the Daniels phase (400-275 B.P.). It is believed that this phase represents the remnants of Mississippian groups that had been debilitated by disease. Vessels are characterized by thick vessel walls, poorly executed stamping, and wide applique rim strips. These later Mississippian occupations in the Wateree Valley may represent the remains of the Cofatachqui chiefdom(Elliott and O’Steen 1987:10). Other Lamar sites in and near the project area include: Harrison Mound now submerged; Eagles Nest; Boykin; and other mounds that are submerged beneath the lakes (Stuart 1967, 1970:24). Mississippian Sites have been identified in North Carolina in Burke, Catawba, Iredell, McDowell, and Mecklenburg counties. In South Carolina, a total of 11 sites have been identified in Chester, Fairfield, Kershaw, Lancaster, York counties.
As a result of the failure of the Native American uprising in the Yamassee War, the Catawba Nation was formed in the eighteenth century from several groups that once formed the Cofitachique confederation (Harris 1987:2). These groups were encouraged to settle there by government to serve as buffer at northern border of the colony, and included the Kadapaus, Esaws, Sugarees, Waterees, Wisacks, Congarees, Santees, Saponis, Cussoes, Peedees, Yamassees, Coosas, Enos, Occaneechis, Keyauwees, Chowans, Nachees, and Cheraws. While the groups maintained separate identities, they were bound by trade, marriage, and adoption. In 1759, a smallpox epidemic killed over 50 percent of the Catawba. As a result, groups left the Sugar Creek villages and settled at Pine Tree Hill (later, Camden, SC) and then moved to the upper Catawba at 12 Mile Creek (Harris 1987:3). The villages at 12 Mile Creek were destroyed by the British during the Revolutionary War, so the remaining groups then moved to lands that had been given to them by treaty in 1763. It was during the eighteenth century that the pottery trade developed among the Catawba (Harris 1987:4). 3.2.1
Introduction The following narrative depicts the general historical development of the project area. More specific, local histories have been published for nearly every county within the project area. Comprehensive architectural surveys have also been conducted in many of the counties within the North Carolina portion of the project area. The results of these surveys have been published and contain detailed accounts of the architectural development within each county. In South Carolina, published historic preservation plans and architectural inventories from the mid to late 1970s concentrated on identifying resources that possessed potential NRHP significance. While some of this detailed information has been included in the following historic period context, this narrative is intended to give a broad overview of the major trends in the development of the project area. The discussion is divided into seven historical periods (Table 3).
Research for this context was conducted within the files of the North Carolina State Historic Preservation Officer (SHPO), Raleigh and Asheville (Western Regional Office). Additional resources on the historical development of North Carolina were obtained at the Charlotte Public Library and the North Carolina State Library, Raleigh. Both general history and the SHPO site files were reviewed at the South Carolina Department of Archives & History, Columbia. Information pertaining to the development of hydropower facilities within the project area were obtained from Duke Power Company’s archives in Charlotte. Two published histories, North Carolina Through Four Centuries (Powell 1989) and South Carolina: A History (Edgar 1998) provided broad overviews of historical development within the two state area. 3.2.2 Colonial Period, 1540-1764 The areas of upland South Carolina and middle and western North Carolina were first explored by European Americans in the early-to-middle sixteenth century. The explorations of Spaniards Hernando de Soto, 1540-1542, and Juan Pardo, 1566-1567, penetrated the Catawba-Wateree valley and continued through to the Appalachian mountain range and into present-day Tennessee (Davis 1999; Hudson 1990; Hudson et al. 1984; Smith 1997). Both the French and the Spanish had attempted settlements on the coast of South Carolina, but both attempts had failed. In 1587, the English settlement of Fort Raleigh was established on the North Carolina Coast, but by 1590, the fort had been abandoned with the fate of the settlers left unknown. The company became known as the Lost Colony. Jamestown in Virginia, established in 1607, was the next attempt by the English to settle the coast. Not until 1670, with the establishment of Charles Towne (Charleston), were the English successful in settling the area of present-day South Carolina. The port town served as a base for explorations to the interior while trade with the native populations sustained the settlement. In 1701, explorations into western North Carolina from Virginia led by Englishman John Lawson also drew the interests of the colonists towards the western edge of the frontier. The primary obstacle to settlement of the western lands in both North and South Carolina was the presence of Native American populations. Siouan-speaking peoples, known generally as the Catawba, lived east of the Blue Ridge and into the Piedmont region, while the Iroquoian-speaking Cherokee dominated the western mountainous regions and part of the western Piedmont (Powell 1989). Hostilities between the Native and European groups, however, would persist into the eighteenth century and included the coordinated Yemassee and Cheraw attack on the English settlements during 1715-1718. Fort Dobbs, constructed in 1755 near the Yadkin River north of present-day Statesville in Iredell County, North Carolina, played an important role in providing a refuge for settlers in the North Carolina backcountry during the war. The combination of war, introduced disease, and other negative influences, greatly reduced the Native populations by the mid-eighteenth century. With the cessation of the French and Indian War (1754-1763), and the signing of the Treaty of Paris (1763), the Piedmont and western areas of the Carolina were considered safe for European settlement and Fort Dobbs was abandoned in 1764. Although fur traders were the first to enter the territory, permanent settlers into the midlands and uplands of South Carolina and the Piedmont and mountains of North Carolina arrived as early as the 1730s, drawn by the abundance of cheap, fertile land. These settlers came from the coastal settlements, as well as from the colonies to the north. Large numbers of Scots-Irish, German, and English settlers traveled from Pennsylvania and Virginia along an old Native trading path, later known as the Great Wagon Road, that traversed the Valley of Virginia, into the North Carolina Piedmont, into the lower Catawba Valley and beyond. Travel into the mountains of western North Carolina, an area not heavily settled until after the American Revolution, was generally by foot or horseback since few of the rivers were navigable into the region. The Catawba River connected the area with South Carolina, while the Dan River flowed to Virginia. Residents of the project area in South Carolina had access to tributaries to the Santee and Pee Dee rivers, but the Catawba River was not navigable above the falls at Rocky Mount (now Great Falls). The period between 1740 and 1760 was a time of heavy immigration into these inland regions. With the increase in population, smaller counties were created from larger political divisions to provide for limited local governmental functions. The Carolinas had been granted by Charles II to the eight Lords Proprietors in 1663 (a revised charter was issued in 1665), and the singular county of Albemarle was established. In 1710, North and South Carolina were made separate provinces, each with a royal governor. At that time, North Carolina consisted of seven precincts (changed to counties in 1739), while South Carolina, originally divided into four counties, adopted a parish system in association with the Anglican Church (Corbitt 1987; Stauffer 1994). While the parishes served as election districts, legislative and judicial powers remained in Charleston. Within the project area, Mecklenberg County, North Carolina, created in 1762, was the only individual county established during this period. All other areas were encompassed within larger divisions. Each immigrant group brought with them a distinctive culture and traditions, including religion. In general, the Scots-Irish settlers were Presbyterian, while the German settlers were Lutheran. Many English and Welsh settlers were either Quaker or Baptist. These ethnic groups tended to settle in the same areas, and often influenced the assignment of municipal positions (Mintz and Smith 1998). No recorded architectural resources within the project area date to this time period. The initial constructions of the settlers were temporary shelters that would have been replaced, or would not have survived the elements over the years. Many early settlers moved into log houses that were built by Native American inhabitants. These dwellings were built adapting techniques learned from European traders (Bishir et al. 1999). 3.2.3 American Revolution, 1765-1789 The first immigrants into the upper Catawba valley tended to settle along the river and creeks around which small settlements, trading posts, and fords later developed. For example, the Sherill family settled the west bank of the Catawba River in 1747, near a point later known as Sherill’s Ford in Catawba County, North Carolina. The Sherills may also have brought the first slaves into the region (Hargrov |