AN ASSESSMENT OF THE SPATIAL EXTENT OF A PREHISTORIC EARTHWORK AT THE DEWEY’S KNOLL SITE (NYSM 2490) IN RIPLEY, NEW YORK USING GIS, SOIL MORPHOLOGY, AND ORGANIC LOSS ON IGNITION by ANDREW M. JAOUEN Bachelor of Science, Mercyhurst University, 2012 A Thesis Submitted to the Graduate Faculty of Mercyhurst University in Partial Fulfillment of the Requirement for the Degree MASTER OF SCIENCE ERIE, PENNSYLVANIA 2016 DO NOT CITE IN ANY CONTEXT WITHOUT PERMISSION OF THE AUTHOR Andrew M. Jaouen, Mercyhurst University, Erie, PA 16546 (ajaoue01@lakers.mercyhurst.edu), 2016 Mercyhurst University Department of Anthropology/Archaeology We hereby approve the thesis of Andrew M. Jaouen Candidate for the degree of Master of Science _________________________ ____________________________ Allen Quinn Professor of Archaeology, Advisor _________________________ ____________________________ Nicholas Lang, PH.D. Professor of Geology _________________________ ____________________________ Lyman Persico, PH.D. Professor of Geology, Whitman College ________________________ ____________________________ Anthropology Department Chair  TABLE OF CONTENTS ABSTRACT 1 CHAPTER 1: OBJECTIVES 3 The Role of Geoarchaeology to this Study 4 PART I: LITERATURE REVIEW 6 CHAPTER 2: FORMATION AND GEOLOGIC CONTEXT OF THE SOUTH EAST LAKE ERIE BASIN 6 Pre-Glacial and Glacial Geomorphological Development of Lake Erie 6 Geologic History of the Lake Erie Basin 9 Glacial Lake History of Study 9 Post Glacial History and Lake Phases 10 Early Lake Erie 12 Early Lake Erie (discharge from Lake Algonquin) 14 Second stage Early Lake Erie 15 Middle Lake Erie 17 Modern Lake Erie 21 Formation and Context of the Southeastern Lake Erie Basin 21 CHAPTER 3: CULTURAL HISTORY OF THE SOUTHEASTERN LAKE ERIE BASIN AND SURROUNDING AREA TO THE RIPLEY SITE 23 Socio-cultural Models 24 The Early Late Woodland Period (A.D. 1000 1300) 27 The Terminal Late Woodland Period (A.D. 1300 – 1500) 32 Pre-Contact Neutral and Huron-Petun 37 Contact Period (A.D. 1500 – 1650) 39 Historic Period (A.D. 1640 – 1656) 44 Post-Contact Neutral 44 Post-Contact Wenro 46 CHAPTER 4: KNOWN HISTORY OF THE ERIE 48 Cultural History of the Erie 48 Potential Erie Sites in Niagara Frontier Area of New York 52 Potential Erie Sites in Western Pennsylvania and Western New York 53 History of the Erie Presence in Ohio 56 Ethnohistorical Information on the Demise of the Erie and Huronia 57 CHAPTER 5: IROQUOIAN EARTHWORKS AND PALISADES 63 Earthworks and Palisades for Defense 63 Earthworks and Ritual 71 CHAPTER 6: SOIL MORPHOLOGY 74 Soil and Soil Texture 74 Soil Formation 77 Soil Horizonation 77 Master Horizon Designations 77 Soil Color 79 Soil Structure 80 Buried Horizons and Paleosols 82 Sedimentary Strata 85 Soil Sampling and Stratigraphic Prospecting Strategies 86 CHAPTER 7: THE RIPLEY SITE 89 Stratigraphic Context and Depositional Environment 89 Climate 90 Ripley Site Soil 90 Post-Depositional Processes and Bluff Erosion at Ripley 92 History of Investigation 97 The Ripley Earthwork 102 Current Archaeological Research 108 PART 2: SITE ANALYSIS 112 CHAPTER 8: METHODOLOGY 112 ArcGIS Methodology 112 Field Methodology 112 Laboratory Methodology 116 Beckman Coulter LS 13 320 Laser Diffractions Particle Analyzer Analysis 117 Organic Loss on Ignition (LOI) 118 Statistic Methodology 120 CHAPTER 9: RESULTS 123 ArcGIS Results/ Objective 1 123 Soil Core Maps/Objective 3 127 Characterization of Auger Samples 130 Quantitative Texture Analysis Results 131 Sieve Grain Size Analysis Results 139 Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer Analysis Results 141 Logic Test Results 150 Results for the Organic Loss on Ignition Analysis (LOI) 153 CHAPTER 10: DISCUSSION 161 Map Discussion 161 Soil Core and Horizon Discussion 163 Texture Analysis Discussion 165 Sieve Analysis Discussion 166 Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer Analysis Discussion 166 Logic Test Discussion 168 Organic Loss on Ignition Analysis Discussion 169 CHAPTER 11: CONCLUSION 171 ACKNOWLEDGEMENTS 174 REFERENCES CITED 175  LIST OF FIGURES Figure 1:Map of Ripley Site in Relation to Lake Erie (Google Earth 2015). 5 Figure 2: Basic outline of the geologic history of the Great Lakes (Google Images 2014). 6 Figure 3: Cross Section of Modern Lake Erie (Herdendorf 2013, Figure 2). 8 Figure 4: Bathymetry of Modern Lake Erie showing some of the place names referenced in this paper: A- Detroit River Delta, B- Pelee Passage, C-Cedar Point and Sandusky Bay, D- Conneaut Bank, E- Presque Isle Bank, F- Norfolk Morraine, G- Buffalo Ridge, and H- Niagara River outlet (Herdendorf 2013, Figure 1). 8 Figure 5: Southern portion of the Great Lakes Region, showing the water bodies in the Lake Erie Basin at the outset of the Holocene Epoch, ~12,000 years ago, following the opening of the Niagara River Outlet once the Wisconsin ice sheet had retreated from the Lake Erie Basin. (Herdendorf 2013, Figure 3). 13 Figure 6: Early Lake Erie was bypassed by Upper Lake drainage ~ 12,000 years ago, establishing isolated (endorheic) lakes in the eastern and central basins. A connection of these lakes via the Pennsylvania Channel is uncertain (Herdendorf 2013, Figure 4). 14 Figure 7: Early Lake Erie expanded with discharge from Lake Algonquin. Note that Lake St. Clair-Detroit River flow is directed through Pelee Passage. For a brief period, Lake Erie may have drained eastward via a Niagara River Outlet (Herdendorf 2013, Figure 6). 15 Figure 8: Glacial retreat opened the North Bay Outlet, lowering the Upper Lakes ~ 10,300 years ago and halted drainage to the Lake Erie Basin (Herdendorf 2013, Figure 10). 16 Figure 9: Glacial retreat to the north isolated Early Lake Erie, recreating an endorheic basin. A connection of the eastern and central basins is shown via the Pennsylvania Channel (Herdendorf 2013, Figure 9). 17 Figure 10: Middle Lake Erie continued to rise by isostatic rebound at the eastern end of the lake prior to the return of Upper Lakes Drainage (Herdendorf 2013, Figure 11). 18 Figure 11: As the North Bay Outlet was raised by isostatic rebound, Lake Nipissing drainage began to flow to Lake Erie as well as continue to flow to the St. Lawrence River (Herdendorf 2013, Figure 12). 19 Figure 12: Isostatic rebound closed the North Bay Outlet ~4,700 years ago, returning all Upper Lakes Drainage to Lake Erie. Deltas formed in Lake St. Clair and western Lake Erie during this time and the water level in the Lake Erie Basin rose to a high stand. This high water level stage formed ~4,700 years ago as drainage from the Upper Lakes returned to the Lake Erie Basin. Erosion of the Niagara River Outlet channel eventually established the current lake level (Herdendorf 2013, Figure 13). 20 Figure 13: Cultural chronology of the Southeastern Lake Erie Basin and surrounding areas (Sullivan et al. 1996, Table 3.1. 1Brose 1976a; 2Lantz 1989; 3Ritchie 1965, Stothers 1977, White 1961, 1976; 4Schock 1974, 1976. 27 Figure 14: Nicholas Sanson Map from A.D. 1650. Note the Presence of "N. du Chat" in the Center of the Map. 50 Figure 15: Examples of Iroquois Palisaded Villages (Google images 2014). 69 Figure 16: Ternary diagram for soil texture. 75 Figure 17: Natural Soil Horizon Column for the Ripley Site (UC Davis Soil Resource Web 2013). 91 Figure 18: Percent Organic Matter, sand, and Clay with Depth at the Ripley Site (UC Davis Soil Resource Web 2013). 92 Figure 19: Map of Lake Erie Indicating Areas of Erosion Study. Table Depicts Recession Rates for Map Locations in m/yr. (Geier and Calkin 1983). 94 Figure 20: Photograph of the Ripley Site’s Shoreline Facing West. 95 Figure 21: View of the Cliffs at Northeastern Part of the Village Site in 1907 (Parker 1907:477, Figure 1). 96 Figure 22: Bird's Eye View of the Ripley Site Facing Southeast. Note: Height of the Cliffs (Bing Maps 2015). 97 Figure 23: Arthur C. Parker's 1907 Hand Drawn Map. Note: Trench Excavations across the Earth Ring (Parker 1907). 99 Figure 24: Map Depicting the Approximate Location of the RAP Excavations (Harding 2014, Figure 14). 101 Figure 25: Ripley Site Map and Features from All Previous Excavations. Note: Earth Ring in Northern Portion of Site (Sullivan, 1996, Figure 4.12). 102 Figure 26: Cross Section of Soil Beneath Obliterated Earth Ring (Parker, 1907:518, Figure 18). 104 Figure 27: Parker's Excavation Notes of the Earth Ring and Associated Pit Features. Note: Excavation Trenches Cut Through the Earth Ring Area (Parker 1907, Plate 4). 105 Figure 28: Conklin's Hand Drawn Map that was redrawn from a Copy on File at the Rochester Museum & Science Center (Sullivan et al. 1996:50, Figure 4.9). 107 Figure 29: General view of the Ripley Site facing North. Note: Colored Pin Flags Indicating Artifact Locations on the Surface. 110 Figure 30: View Looking Northwest Over Dewey Knoll (Parker 1907:471, Plate I). 111 Figure 31: General View of the Ripley Site Facing East. Note the Total Station Used to Map Core Locations from Datum. 114 Figure 32: View of the Author Augering for Soil Samples. 115 Figure 33: Example of Equipment Used to Take Soil Samples. 116 Figure 34: Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer. Note: Core Samples on Cart to the Left. 118 Figure 35: Lab setup for Organic Loss on Ignition (LOI) Analysis. 119 Figure 36: Samples in Furnace Ready to be Burned for LOI Analysis. 119 Figure 37: Burned Samples in Desiccator After Analysis. 120 Figure 38:Map Depicting Shoreline Erosion Since 1898. Note: * Indicates the Use of Topographic Maps. 124 Figure 39: Map Depicting Shoreline Erosion Since 1898. Note: * Indicates the Use of Topographic Maps. 124 Figure 40: Map Depicting the Approximate Location of the Earth Ring Based on Parker’s 1907 Map and Shoreline Erosion. Note the Red Circles Represent Locations Where the Cores Were Taken In Relation to the Primary Datum. 125 Figure 41: Overlay of Parker's 1907 Excavation Notes on Current Base Map. Note: Trenches Through the Previously Documented Earth Ring Location (Google Earth 2015; Parker 1907) 126 Figure 42: Soil core map for AJ1-AJ6. Note: Arbitrary Levels, Munsell Colors, and Pedostratigraphy are Documented. AJ3 is Only 60 cm Deep Due to Rock Impasse. AJ 6 is 7.5 m between AJ4 and AJ5. All Munsell Designations are Wet. These Cores are from West to East. 127 Figure 43: Soil Map for Cores AJ7-AJ12. Note: Arbitrary Levels, Munsell Colors and Pedostratigraphy are noted. The Void in AJ10 was Most Likely Caused by Bioturbation. All Munsell Designations are Wet. These Cores are from East to West. 128 Figure 44: Soil Map for Cores AJ13-AJ18. Note Arbitrary Levels, Munsell Colors and Pedostratigraphy. All Munsell Designations are Wet. These Cores are from West to East. 129 Figure 45: Core Map Depicting Horizons and Texture Types for AJ1-AJ6. Note: AJ3 is Only 60 cm Due to Rock Impasse. AJ6 is 7.5 m between AJ4 and AJ5. These Cores are from West to East. 132 Figure 46: Core Map Depicting Horizons and Texture Types for AJ7-AJ12. Note: Large Void in AJ10 is Probably Caused by Bioturbation. These Cores are from East to West. 133 Figure 47: Core Map Depicting Horizons and Texture Types for AJ13-AJ18. These Cores are from West to East. 134 Figure 48: Ternary Diagram for soil textures in Cores AJ1-AJ6. Red-AJ1, Green-AJ2, Blue-AJ3, Yellow-AJ4, Black-AJ5, Orange-AJ6. 135 Figure 49: Ternary Diagram for Soil Textures in Cores AJ7-AJ12. Red-AJ7, Green-AJ8, Blue-AJ9, Yellow-AJ10, Black-AJ11, Orange-AJ12. 135 Figure 50: Ternary Diagram for Soil Textures in Cores AJ13-AJ18. Red-AJ13, Green-AJ14, Blue-AJ15, Yellow-AJ16, Black-AJ17, Orange-AJ18. 136 Figure 51: Graph Depicting Sand Percentage in Weight for 0-25 cm in Each Core. 140 Figure 52: Graph Depicting Percentage in Weight for Bottom of Each Core. 140 Figure 53: Graph Depicting Percentage in Weight for Top and Bottom of Each Core. 141   LIST OF TABLES Table 1: Wentworth Scale for Particle Grain Size Textures (Google Images 2015). 76 Table 2: Horizon, Observations/Notes, and Texture Type per Core and Depth. 136 Table 3: Excel Logic Test Results. Note: The Green Highlights Indicate Those Levels that have an Average Particle Size Smaller than Above and Below It, the Yellow Highlights are Levels that have a Larger Average Particle Size Above and Below It. 150  ABSTRACT The Ripley Site (NYSM 2490) is an Erie Nation earthwork site located in Chautauqua County, New York on the bluffs above southeastern Lake Erie. The site is also known as Dewey’s Knoll and this location has been the center of investigation for numerous archaeological excavations of varying professionalism and intensity since the early 20th century. A circular or crescent-shaped earthwork stood on the site, but it was leveled due to farming activity in the 19th century. Numerous archaeologists have noted an earth ring at this site, but it is not currently present on the modern surface. Due to these circumstances, this project was designed to attempt to relocate the earthwork. The first portion of this study is a literature review to assist the author with a better understanding of the geologic formations of the site and nearby Lake Erie, indigenous groups that lived around the Southeastern Lake Erie Basin, Iroquois earthworks and earth ring use, soil morphology, and previous work conducted at Ripley. The second part was a scientific analysis to attempt to locate the Ripley Site earth ring. The objectives of Part II of this study was to 1) Review the current shoreline of the Ripley Site and determine the nature and extent the site has been lost to erosion; 2) Determine the formation and stratigraphy of the site using soil coring; 3) Determine the extent of the earthwork (if possible); 4)To attempt to identify the earthwork signature using stratigraphy, grain size analysis and organic loss on ignition testing. Shoreline erosion analysis using ArcGIS indicated that ~12 m of shoreline has been lost since 1951. In addition, the data extracted from numerous soil analyses on 92 soil intervals in 18 soil cores helped to determine that the Ripley earthwork is more than likely no longer present at the Ripley Site due to erosion, trenching, plowing, excavation and other means of site disturbance. The analyses conducted in this study include a semiquantitative mapping of color and textures of each core, a sieve analysis, a Beckman Coulter LS 13 320 Particle Size Analyzer analysis, a Logic test to sorth through anomalies in the soil, and an Organic Loss on Ignition (LOI) analysis. it was determined that the site has an abundance of disturbed soil with the occasional anomalously thick Ap horizons in some locations, it is at risk of eroding into Lake Erie, and that much of the site has potentially already eroded into the Lake. The particle sizes and organics were what would typically be found with depth for a glacial till bluff. This important information only added to the author’s conclusion that the earth work was already excavated or eroded. Using this study as an example, it is possible that other sites along Lake Erie are at risk of being destroyed or have already fallen into Lake. The author believes that the archaeological community needs to be aware of this situation. With over 90% of this earth ring gone, and an unknown amount of the original Ripley Site eroded into Lake Erie, it is important to emphasize that we could lose important archaeological data that can help us to understand the Ripley Site better.  CHAPTER 1: OBJECTIVES The Ripley Site (NYSM 2490) is an Erie Nation earthwork site located in Chautauqua County, New York on the bluffs above southeastern Lake Erie (Figure 1). The site, also known as Dewey’s Knoll, has been the center of investigation for numerous archaeological excavations of varying professionalism and intensity since the early 20th century. A circular or crescent-shaped earthwork stood on the site that was destroyed due to farming activity in the 19th century. This earthwork was documented in Arthur C. Parker’s work Excavations in an Erie Indian Village and Burial Site at Ripley, Chautauqua Co. N.Y., in field notes from an avocational archaeologist named Jordan Christensen who excavated portions of the site in 1957 and in a map drawn by another avocational archaeologist Carlton Conklin who worked on the site in 1962. With so many archaeologists noting the presence of the earth ring, and it not being present and visible on the surface, this project is an attempt to relocate the earthwork. The objectives of this research include: Review the current shoreline of the Ripley Site and determine the nature and extent the site has been lost to erosion; Determine the formation and stratigraphy of the site using soil coring; Determine the extent of the earthwork (if possible); To attempt to identify the earthwork signature using stratigraphy, grain size analysis and organic loss on ignition testing.  The Role of Geoarchaeology to this Study Geoarchaeology is a multidisciplinary field that applies earth science and geological concepts and methods to solve archaeology problems (Waters 1992). Three main aspects of geoarchaeology that are especially important include: 1) site formation processes, 2) landscape reconstruction, and 3) stratigraphy (Waters 1992). The first two aspects are crucial to Objectives 1, 2 and 3 above, to determine how much of the site has been lost to erosion, to determine the extent of the earthwork, and to determine the site formation. The other aspect, stratigraphy, also applies to Objective 3, to determine the natural stratigraphy of the site, and Objective 4, to identify the earthwork signature by using soil and sediment stratigraphy. Archaeological sediments can provide a great deal of information to interpret cultural activities (Stein and Farrand 2001). There are four stages in the life history of sediment: its “source, its transport history, the environment of deposition, and post-depositional alterations” (Stein and Farrand 2001:10). Cultural and natural processes can be reconstructed from the history of a sediment and to accomplish this, attributes from each sediment are analyzed and compared to a control found off-site that was minimally altered by cultural or human activities (Stein and Farrand 2001:10). The general sediment source can be determined by analyzing texture and composition and comparing possible sources. The cause of transport can also be determined by both texture and composition (Stein and Farrand 2001:12). To investigate the environment of deposition, both textural and compositional information can be structurally employed. At last, post-depositional changes are identified by noting evidence of disturbances and vertical horizons (Stein and Farrand 2001:14). All of these observations can be obtained through excavations, natural exposures, trenches, or soil cores. To understand how a site forms, one must examine and consider three processes: the cultural process or behaviors responsible for the original formation of the archaeological site, the cultural practices that alter or obscure the original behavioral signatures (including the action of modern landowners, looters, archaeologists, or any people post-dating the original site occupation), and natural processes (Bartley 2006:4; Stein 2001). The original culture process creates a pattern of artifacts in space, and the latter two process types destroy, alter, obscure, or sometimes preserve that original pattern (Bartley 2006:4; Stein 2001). This project is concerned with the latter two types of processes, while the first type will be considered only by the principal investigators of the site, except for the source of construction material for the earthwork. To understand the formation of the Ripley Site, it is important to understand the geologic formation and history of Lake Erie itself. Figure SEQ Figure \* ARABIC 1:Map of Ripley Site in Relation to Lake Erie (Google Earth 2015). PART I: LITERATURE REVIEW CHAPTER 2: FORMATION AND GEOLOGIC CONTEXT OF THE SOUTH EAST LAKE ERIE BASIN Pre-Glacial and Glacial Geomorphological Development of Lake Erie A series of ice contact glacial lakes had developed within the Lake Erie Basin during the retreat of the Laurentide Ice Sheet (LIS). These lakes had a complex history that was controlled by multiple factors including isostatic rebound, ice margin position, and the elevation of outlet sills. These proglacial lakes ended when the Niagara outlet opened, draining these lakes into the Ontario Basin 4,000 years ago (Figure 2) (Coakley and Lewis 1985). Figure SEQ Figure \* ARABIC 2: Basic outline of the geologic history of the Great Lakes (Google Images 2014). Lake Erie has had the longest and most complex glacial and postglacial history of all of the Great Lakes (Herdendorf 2013:24). There is evidence from detailed bathymetric (National Geophysical Data Center 1998; Holcombe et al. 2003, 2005) and other data including seismic/sedimentological information (Lewis et al. 2012) that have shown that the lake-floor and shoreline have features that are indicative of former shorelines (Figure 4). This bathymetry demonstrates the rise and fall in lake levels has changed the paleogeographic outline of water bodies within the basin over time (Herdendorf 2013:24) (Figure 4). Lake Erie today has a surface area of 25,657 km2 (9,906 mi2) and a volume of 484 km3 (116 mi3). It is 388 km (241 mi) long, but at the widest point, it is only 92 km (57 mi) (Herdendorf 2013:24). At a maximum depth of only 64 m (210 ft), Lake Erie is the shallowest of all the Great Lakes and is the only one whose bottom does not reach below sea level (Bolsenga and Herdendorf 1993). Today, the mean water level is only 174 m (571 ft) above mean sea level (AMSL). Based on topography, Lake Erie is divided geomorphically into three distinct basins: the western, central, and eastern (Figure 3). The western basin, which lies west of the tip of Point Pele, Ontario to Cedar Point, Ohio is the smallest and shallowest, with most of the depths between 7 and 10 m (23 and 33 ft). Here, a number of bedrock islands are situated in the western basin and form a partial divide between the western and central basins. The bottom of the western basin is flat except for several steep-sided islands and shoals in the eastern part (Herdendorf 2013:24). The eastern basin is fairly deep and bowl-shaped, with a considerable portion of the bottom below 30 m (100 ft). The deepest at 64 m (210 ft) is located about 13 km (8 mi) east-southeast off the tip of Long Point, Ontario. This “deep” basin comprises of 24 percent of Lake Erie’s total area and 32 percent of its total volume. The glacially deposited ridge north of Erie, Pennsylvania contains a notch known as the Pennsylvania Channel, which provides a subsurface connection for water circulation between the central and eastern basins. Figure SEQ Figure \* ARABIC 3: Cross Section of Modern Lake Erie (Herdendorf 2013, Figure 2). Figure SEQ Figure \* ARABIC 4: Bathymetry of Modern Lake Erie showing some of the place names referenced in this paper: A- Detroit River Delta, B- Pelee Passage, C-Cedar Point and Sandusky Bay, D- Conneaut Bank, E- Presque Isle Bank, F- Norfolk Morraine, G- Buffalo Ridge, and H- Niagara River outlet (Herdendorf 2013, Figure 1).  Geologic History of the Lake Erie Basin Under Lake Erie lies a middle Paleozoic formation of sedimentary rock composed of dolomite, limestone, sandstone, and shale (Larsen and Schaetzl 2001). These sedimentary rocks were deposited roughly 300 to 400 million years ago under a variety of conditions that range from tropical barrier reef habitats, deltaic environments, associations with mountain building episodes, and tectonic plate collisions that took place in the east (Herdendorf 2013). General uplift started long-term erosion that resulted in the development of stream valleys and a drainage system that runs north to south along the lake. During the Cenozoic, continental glaciers further sculpted this valley system by overriding the Niagara Escarpment and down cutting deeply in the shale at the eastern end of the lake, moderately deep in the central portion, and shallow in the limestone/dolomite bedrock at the western end (Herdendorf 2013:26). This process formed the distinctive three basins that make up Lake Erie (Figure 3). During the most recent glacial advance also known as the Wisconsin stage, ice extended as far south as the Ohio River around 21,000 years ago. After, the ice margin receded in pulses with several ridges of glacial moraines deposited under what is now the bed of Lake Erie (Lewis et al. 2012). End moraines occur at the junctions of Lake Erie’s basins, which further separates them from one another (Herdendorf 2013). Lake Erie owes most of its creation and geomorphic character to changes that have been cause by Pleistocene glaciers. As these ice sheets paused, moraines made of glacial till, clay, and gravel built up at the ice margins. Large glacial lakes formed from the earlier moraines and the ice front for 2,000 years beginning around 14,000 years ago. Glacial Lake History of Study The glacial lake stages have been studied as far back as the early 19th century. The first study conducted took place in Northeastern Ohio by Colonel Charles Whittlesey whose research interest was on the glacial shorelines. Whittlesey described a series of sand terraces and beach ridges along the south shore of Lake Erie at altitudes from 21 m (68.8 ft) to 62 m (203 ft) above the present lake (Totten 1985). He determined that the terraces represent “ancient beaches,” and that the ridges were “sub-marine deposits”. Further north, Schooler (1974) compiled past studies of beach ridges in the Pennsylvania portion of the Erie Basin, discovered, and mapped a vast amount of beach ridges and glacial lake features using field study and point survey elevations on beach ridge-tops. Totten (1985) described a set of sand and gravel ridges within the generally flat topography of the Lake Erie Basin as beach ridges. Stable lake levels that occurred from 50 to 300 years in duration created these beach ridges and sand dunes. Totten (1985) in Ohio, Calkin and Feenstra (1985) in New York, and Barnett (1985) in Ontario also utilized fieldwork and point survey elevations of beach ridge-tops to document beach ridges. Post Glacial History and Lake Phases The Laurentide Ice Sheet (LIS) around 14,400 YBP dammed and created the oldest proglacial lake which was Lake Maumee I. The Defiance Moraine marked this Lakes position. At an elevation of 244 m (800 ft) AMSL, near Fort Wayne, Indiana stood the outlet for this lake and it had drained westward into the Wabash River Valley (Calkin and Feenstra 1985). Glacial Ice margin retreat opened the Grand River Valley outlet at an elevation of 232 m (761 ft) AMSL, which initiated Lake Maumee II around 14,200 YBP. A later readvance of the Huron Lobe closed this outlet, which raised sea level and created Lake Maumee III around 14,000 YBP. This large lake drained via the Imlay channel at a present elevation of 238 m (780 ft) AMSL. There is varying debates whether Lake Maumee II or III came first due to the unsure nature of the readvance of the Huron Lobe (Calkin and Feenstra 1985). The retreat of the Huron Lobe reopened the Grand River Valley outlet at an elevation of 216 m (708 ft) AMSL to create Lake Arkona at around 13,800 YBP. This stage filled parts of the Erie and Huron Basins. After some time, this outlet eroded by down cutting of the channel that dropped Lake Arkona’s outlet elevation to 213 m (698 ft) AMSL around 13,700 YBP, then later to 212 m (695 ft) AMSL around 13,600 YBP. These drops in water level can be referred to as Middle Lake Arkona II and Lowest Lake Arkona III (Calkin and Feenstra 1985). Succeeding lake phases could not reoccupy levels north and east of the lake outlet due to isostatic rebound. Even though Maumee II and Arkona drained via the same outlet, their strandlines deviate in elevation to the northeast. During the Mackinaw Interstade, an eastward drainage from the Lake Michigan, Lake Erie and Lake Huron basins occurred via an independent low lake in the Lake Ontario basin known as Lake Ypsilanti around 13,400 YBP. This lake drained into the east through the Mohawk River Valley. This is seen in the Erie Basin by thick deposits of oxidized shallow water or shore sand south of Cleveland at 201 m (659 ft) – 206 m (675 ft) AMSL (Calkin and Feenstra 1985). These deposits are overlain by a thick sequence of silt and clay deposited by Lake Whittlesey and underlain by gravel deposited by a river that graded this lake phase in the Erie basin at 166 m (544 ft) AMSL in elevation (Herdendorf 2013). At some point, the Erie Lobe readvanced and covered eastern outlets to the Ontario Basin, which diverted overflow back to the west. This meltwater exited through the Ubly Channel at an elevation of 226 m (741 ft) AMSL around the Huron Lobe. This lake known as Lake Whittlesey covered the Arkona beaches at around 13,200 YBP. The outlet emptied into Lake Saginaw, which is found in the westernmost part of the Huron Basin (Calkin and Feenstra 1985). As the Port Huron Ice Lobe retreated, the Grand River Valley was reoccupied and eroded to a depth of 210 m (688 ft) AMSL by meltwater. The opening of this new channel formed Lake Warren I at around 13,000 YBP. Downcutting over time of this sill to an elevation of roughly 208 m (682 ft) AMSL created Lake Warren II around 12,900 YBP (Calkin and Feenstra 1985). As the ice melted, a new outlet opened at or near the Mackinac Straights with an elevation of 201 m (660 ft) AMSL. This lake phase known as Lake Wayne was formed at around 12,800 YBP. This level immediately preceded the Warren II level, draining into the Mohawk River Valley of New York (Calkin and Feenstra 1985). Near after, at around 12,700 YBP a minor readvance of ice covered the Mackinac Straights outlet, which raised lake levels to 207 m (679 ft) AMSL, thus draining into the Grand River channel outlet. This is known as Lake Warren III and was the last glacial advance over the Erie Basin (Calkin and Feenstra 1985). As the LIS retreated around 12,600 YBP, Lake Grassmere formed from the opening of the Mohawk River outlet in New York at an elevation of 195 m (639 ft) AMSL. As this outlet lowered, Lake Lundy formed at an elevation of 189 m (620 ft) AMSL. Eventually Early Lake Algonquin formed from Lake Lundy at 184 m (603 ft) AMSL around the same time. Early Lake Erie The glacial lake stages in the Lake Erie Basin ended roughly 12,000 years ago when the Wisconsin ice margin withdrew to the east to allow the lake level to be controlled by the sill at the Niagara Escarpment (Herdendorf 2013:26). Simultaneously, ice had shifted back between Lake Huron and Lake Ontario Basins, opening a northern drainage (Kirkfield Outlet) so that the Upper Lakes (Lake Algonquin) no longer emptied into the Lake Erie Basin (Figure 5). This eliminated the direct glacial influence in the Lake Erie Basin and began a Holocene history of Lake Erie. Forsyth (1973) described in his research that a catastrophic flood of water went over the escarpment; incised a channel in the moraines and bedrock, which resulted in a low water stage (Figure 6) (Herdendorf 2013). Figure SEQ Figure \* ARABIC 5: Southern portion of the Great Lakes Region, showing the water bodies in the Lake Erie Basin at the outset of the Holocene Epoch, ~12,000 years ago, following the opening of the Niagara River Outlet once the Wisconsin ice sheet had retreated from the Lake Erie Basin. (Herdendorf 2013, Figure 3). Figure SEQ Figure \* ARABIC 6: Early Lake Erie was bypassed by Upper Lake drainage ~ 12,000 years ago, establishing isolated (endorheic) lakes in the eastern and central basins. A connection of these lakes via the Pennsylvania Channel is uncertain (Herdendorf 2013, Figure 4). The Niagara River Outlet, still depressed by glacial loading was around 50 m (164 ft) below the present level of Lake Erie (Holcombe et al. 2003; Lewis et al. 2012). Known as Early Lake Erie, this low stage had an elevation of 120 m (394 ft) AMSL and consisted of a small lake in the eastern basin and a smaller unconnected lake in the central basin. Early Lake Erie (discharge from Lake Algonquin) A re-advancement of the glacial margin north of Lake Erie blocked the Kirkfield Outlet roughly 10,400 years ago. Because this outlet was blocked, the Early Lake Erie began to release from Lake Algonquin (Lake Huron and Michigan Basins) via the Port Huron Outlet through the newly formed St. Clair River-Lake and the St. Clair-Detroit River system (Calkin and Feenstra 1985). By about 10,400 YBP this additional inflow created a lake that consisted of a marshy western basin through which extension of the river flowed via the 1) Pelee Passage, a shallow central basin lake that flowed to the east through the Pennsylvania channel which cut through the Norfolk Moraine and 2) a deeper eastern basin, which may have drained to the east over the Niagara Escarpment through the Niagara River (Figures 4 and 7) (Herdendorf 2013). It is debated on what the remaining Early Lake Erie did. Lewis and others (2012) believe that Early Lake Erie remained an isolated Lake, but Holcombe and others (2003) determined that for a brief period of time around 10,400 YBP that Early Lake Erie rose above the Niagara River sill by virtue of Upper Lakes drainage. Figure SEQ Figure \* ARABIC 7: Early Lake Erie expanded with discharge from Lake Algonquin. Note that Lake St. Clair-Detroit River flow is directed through Pelee Passage. For a brief period, Lake Erie may have drained eastward via a Niagara River Outlet (Herdendorf 2013, Figure 6). Second stage Early Lake Erie During this stage of Lake Erie, the flow into western Lake Erie was disturbed 10,300 YBP when the North Bay Outlet was released from Lake Algonquin to the St. Lawrence River by deglaciation (Figure 8). This lowered the level in the Lake Huron Basin for several thousand years and stopped drainage to the Early Lake Erie (Larson and Schaetzl 2001). For about 5,000 years after this, the drainage bypassed Lake Erie. This stoppage of over 90 percent of Lake Erie’s inflow created stagnant conditions (Figure 8). This was most likely due to climate changes which lowered precipitation rate and lead to increased evaporation. The chance that Early Lake Erie resided within a closed lake basin was originally proposed by Lewis and others (1999, 2012) and was shown by new bathymetric data compiled by the National Geographic Data Center (1998) that revealed former shorelines that are now submerged under the current Lake Erie (Holocombe et al. 2003). Figure SEQ Figure \* ARABIC 8: Glacial retreat opened the North Bay Outlet, lowering the Upper Lakes ~ 10,300 years ago and halted drainage to the Lake Erie Basin (Herdendorf 2013, Figure 10). Figure SEQ Figure \* ARABIC 9: Glacial retreat to the north isolated Early Lake Erie, recreating an endorheic basin. A connection of the eastern and central basins is shown via the Pennsylvania Channel (Herdendorf 2013, Figure 9). As isostatic rebound at the eastern end progressed, rising water flooded the moraine at a low spot known as the Pennsylvania Channel, forming a single lake in the two basins (Figure 9). An alternative mechanism for forming a single lake involves the filling of the central basin by major south shore tributaries to the point that it flooded and expanded the Pennsylvania Channel connecting these basins (Herdendorf 2013). This infilling of tributary water and isostatic rebound permitted Early Lake Erie to expand. Middle Lake Erie About 10,000 YBP the rising water in the lake basin began to slow down, level off, and around 7,500 YBP it ended at an elevation of around 145 m (476 ft) AMSL. This was followed by a very slow rise in elevation for the next 2,000 years. Deemed the “Middle Lake Erie” stage by Hartley (1958), he presented a relevant argument for low stages at 25 m (82 ft) below present Lake Erie based on evidence from test cores; while Coakley and Lewis (1985) used radiocarbon dates and contours on the glacial till to determine that the level was at least 30 m (100 ft) below the current lake level. This stable period of lake level was probably due to decreased precipitation and increased evaporation during the Xerothermic or Hypsithermic Interval (Forsyth 1973; Phillips 1989; Shane et. al 2001) and this counterbalanced isostatic uplift. Because of this, for about 5,000 years (roughly 10,300 to 5,300 YBP) Lake Erie was a closed basin (Figure 10). Uplift of the North Bay Outlet began to raise the levels of the Upper Lakes, known more formally as Lake Nipissing around 5,400 YBP (Larson and Schaetzl 2001). Lake Erie slowly raised but remained remote (Figure 11). Figure SEQ Figure \* ARABIC 10: Middle Lake Erie continued to rise by isostatic rebound at the eastern end of the lake prior to the return of Upper Lakes Drainage (Herdendorf 2013, Figure 11). Figure SEQ Figure \* ARABIC 11: As the North Bay Outlet was raised by isostatic rebound, Lake Nipissing drainage began to flow to Lake Erie as well as continue to flow to the St. Lawrence River (Herdendorf 2013, Figure 12). This Middle Lake Erie stage ended around 5,300 YBP when drainage from the Upper Lake returned to the Lake Erie Basin because of continued postglacial uplift around North Bay, Ontario. This ended the Upper Lake’s drainage into the St. Lawrence River and created the Lake Nipissing stage in the Lakes Huron-Superior-Michigan Basin (Lewis 1969; Calkin and Feenstra 1985). The major influx of water from the Upper Lakes, plus more humid climatic conditions, sharply increased water levels in Lake Erie (Figure 11) and gave way to the formation of a large, now submerged delta in western Lake Erie at the mouth of the ancestral Detroit River (Herdendorf and Bailey 1989). Low water features that previously formed in Lake Erie such as Presque Isle Bank, Conneaut Bank, and the Norfolk Moraine were flooded at this time. This changed the water circulation patterns and the altered littoral drift initiated the formation of large spits of land such as Presque Isle in Erie Pennsylvania and Long Point, Ontario (Herdendorf 2013). Deposition of the delta in Lake St. Clair also had taken place at this time of around 5,300-3,600 YBP and the radiocarbon dates for the lake clays that underlay the pre-modern St. Clair River delta show that formation of the delta began during Lake Nipissing time around 7,300 YBP (Kaszycki 1985), and not during a Lake Algonquin time of around 10,400 YBP as suggested by other researchers (Flint 1957). Lewis and others (2012) have documented shoreline features that indicate Lake Erie rose to a high stand about 3-4 m (10-13 ft) above its present level at around 4,700 YBP (Figure 12). Erosion of the bedrock in the Niagara River Outlet caused the lake to fall to its current level at around 3,500 YBP. Figure SEQ Figure \* ARABIC 12: Isostatic rebound closed the North Bay Outlet ~4,700 years ago, returning all Upper Lakes Drainage to Lake Erie. Deltas formed in Lake St. Clair and western Lake Erie during this time and the water level in the Lake Erie Basin rose to a high stand. This high water level stage formed ~4,700 years ago as drainage from the Upper Lakes returned to the Lake Erie Basin. Erosion of the Niagara River Outlet channel eventually established the current lake level (Herdendorf 2013, Figure 13).  Modern Lake Erie When Lake Erie reached its current level around 3,500 YBP, as coastal erosion occurred, it delivered beach sand to the littoral zone, the massive spits of Long Point and Presque Isle were nurtured and new splits such as Point Pelee, Ontario and Cedar Point, Ohio formed. Penecontemporaneously, barrier beaches were formed across mouths of most of the estuarine tributaries. Today, human construction projects along the coast have modified these natural landforms. This often accelerates erosion rates to adjacent reaches of shoreline; however, Lake Erie has now reached a stable water level (Herdendorf 2013). Some features still exist to this day. For instance, the stream adjacent to the Ripley Site has been downcutting since the Holocene. Formation and Context of the Southeastern Lake Erie Basin The spatial extent of the current study includes an area in western New York, known as the Southeastern Lake Erie Basin. The Allegheny Plateau and the Lake Erie Lowland are the two physiographic provinces that are predominant in this area and the major distinction between these zones is in the elevation, relief, geologic history, and bedrock (Sullivan et al. 1996). Their geologic history and the resistance of the bedrock predominantly cause the elevation differences between these two locations, where it is mostly made of upper Devonian sedimentary shales (~400 MYA) (Beckerman 1981; Shepps et al.1959:4). The older Devonian Shale is predominantly present in the Erie Lowland plains and under Lake Erie, while the sandstone bedrock under the Allegheny Plateau is Mississippian and Pennsylvanian in age (Beckerman 1981). The entire Allegheny Plateau in New York was glaciated except for a small portion found in southwestern New York and these glacial processes were the major cause for the formation of the basin and region itself as mentioned in detail above. During the last (Wisconsin) glaciation, circa. 11,000 YBP, a large glacier covered this region (as mentioned above). As outlined above, glacial processes are responsible for many of region’s landforms. A series of moraines along the lakeshore mark the retreat of the Wisconsin ice margin and are considered to be Late Cary in age (Shepps et al. 1959:32-44). In southwestern New York these moraines include the Lavery, Lake Escarpment, Clymer, Findley, Kent, and Defiance moraines. The Lavery, Defiance, and Lake Escarpment moraines are found close to the Allegheny Plateau and Lake Erie lowland divide. Valleys carved by glaciers cause the topography of the landscape to trend south-southeastern. This trend causes the rivers to flow west-southwest and into the Allegheny and Mississippi watershed (Shepps et al. 1959). In terms of modern political boundaries, the southeastern shores of Lake Erie are found in the western-most section of New York State and Pennsylvania. These counties in this region include Chautauqua, Cattaraugus, Erie, and Niagara in New York and Erie, Crawford, and Warren counties in Pennsylvania. The northeastern corner of Ohio is also part of the region and includes the counties of Cuyahoga, Geauga, Lake, and Ashtabula. In New York, the Niagara Frontier falls in the upper northwest corner and is made up of the Erie and Niagara counties, which lies distinct from the Chautauqua/Allegheny area of the southwest. To understand the context of the Ripley Site it is important to understand the cultural developments of the region.  CHAPTER 3: CULTURAL HISTORY OF THE SOUTHEASTERN LAKE ERIE BASIN AND SURROUNDING AREA TO THE RIPLEY SITE In the Southeastern Lake Erie Basin, occupation by Native American groups is focused on a time period dated from late Prehistoric to European contact or early historic periods (A.D. 1000-1650). The designations Early Late Woodland, Terminal Late Woodland, Contact, and Historic are the terms that are generally used by anthropologists in the areas around western New York, northwestern Pennsylvania, and northeastern Ohio to make sense of different cultural developments between groups. In the Southeastern Lake Erie Basin region, archaeologists use two different chronologies to describe dates relating to Native American occupations. Mayer-Oakes’ (1952) and Ritchie’s (1965) chronologies place the beginning of the Late Woodland Period in A.D. 1000 and consists of a time span until European Contact around A.D. 1650 (Figure 13). In their chronology, the early Late Woodland and Late Woodland time periods are separated by the year A.D. 1300 due to a change in subsistence patterns (see below). Since northeastern Ohio also falls within the Southeastern Lake Erie Basin Region, it is important to note that the chronology sequence used in the state is similar as the Pennsylvania and New York chronology. However, this places the start of the Early Late Woodland Period in A.D. 500 and the Terminal Late Woodland begins in A.D. 1000 (Sullivan et al. 1996:17). Since the Ripley Site is only about 80 km (50 mi) from the Canada border, it is important to include chronology information regarding the Iroquoians of Southern Ontario. Richard MacNeish (1952) and James Wright (1966) established a chronological framework based on ceramic serialization and the direct historic approach (Warrick 2000:420). The mostly widely accepted chronology is Princess Point (A.D. 500-1000), Early Iroquoian (A.D. 1000-1300), Middle Iroquoian (Uren phase: A.D. 1300-1330; Middleport phase: A.D. 1330-1420), and Late Pre-Contact (A.D. 1420-1534). Socio-cultural Models A number of different socio-cultural evolution models have been applied to the later prehistoric period of the Southeastern Lake Erie Basin. These address the variable lengths of occupations, sociopolitical organization, and subsistence strategies in the Woodland time period (Sullivan et al. 1996:17). All of these models situate for later prehistory around an increase in population, reliance on maize horticulture, and sedentism. Engelbrecht (1987) used the historic Huron as an analogy to suggest that the prehistoric and early historic groups of the Southeastern Lake Erie Basin had permanent villages but moved them at frequent intervals. Reuse of the sites is not implied in the model, but movement of sites every 10-20 years is suggested and he cites soil and wood depletion as the most likely causal factors for these moves (Engelbrecht 1987). In addition to this, the northern Iroquois relied on a matrilocal kinship system that also helped to define village movement. Snow (1996) has argued that the origin of matrilocal residence among the northern Iroquois is to be found in migration of the ancestral Iroquoian agriculturist into already inhabited areas of New York and Southern Ontario. Matrilocality did not evolve gradually with agriculture as females assumed greater responsibility for subsistence production, but rather it was a sudden development in response to migration by the ancestral northern Iroquoian groups (Snow 1996). Besides matrilocality, the Iroquois socio-political organization is generally tribal and characterized by a lack of institutionalization of ascribed leadership positions, reliance on village councils, and an importance of relationships between kin groups for intra- and inter-societal interactions (Trigger 1981). Broad categories are often useful to understand these concepts, but models that are more precise are needed to understand differences between societies and for a clear understanding of sociopolitical variation. Village-movement models also incorporate sociopolitical organization (Trigger 1981). In the region of northwestern Pennsylvania, the village movement models derived from trait-based distinctions among late prehistoric occupations in the region’s differing physiographic zones (Lantz 1989). With this in mind, numerous existing late prehistoric cultures have been proposed, and each have to do with a differing geographic locale such as the Lake Erie Plain, upper Allegheny River Valley, Allegheny Plateau, etc. While a tribal organization is implied, this model suggests the use of differing settlement and subsistence patterns for site use, since for example, more fishing stations would be located near the Lake Erie Plain and would differ from those in the Allegheny uplands. The northeastern Ohio model is based on settlement-subsistence models focused on seasonal use of plant and animal resources (Brose 1978b). These Ohio models suggest differing levels of population movement/mobility and multiple occupations over long periods of time. There is also an assumption in these models that a number of roughly similar groups and societies, generally tribal, occupied segments of the Southeastern Lake Erie Basin. More recently, Brose (1994) has proposed a subsistence/settlement pattern model for the Whittlesey of northeastern Ohio, which stresses an economic strategy of utilizing seasonally broad but limited resource catchment areas, and the persistence of small, resource procurement locations (Brose 1994; Brose 1976b). The earliest Whittlesey Phase (A.D. 1000-1250) subsistence and settlement pattern is characterized by small, seasonally occupied hamlets of two to five single-family homes whose residents were dependent upon hunting, fishing, and the gathering of nuts and seeds supplemented by incipient squash and maize cultivation. Later (A.D. 1250-1450), many of these easternmost Whittlesey populations seasonally coalesced into small villages on high river bluffs (Brose 1994:175). Although maize and squash were prominent at this time, the harvesting of fish, elk, raccoon, white-tailed deer, turkey, and wild plant resources remained important. The final Whittlesey Phase (A.D. 1450-1600/1620) included a settlement/subsistence pattern marked by large agricultural villages on high bluffs overlooking arable floodplains (Brose 1994:176). These villages occupied for much of the year, include 10-15 homes often surrounded by palisaded ditches and embankments. Dispersed upland hunting camps also continue to be used, especially in the winter, while spring and fall fishing camps continue to be utilized along the lakeshore (Brose 1994:177). In Southern Ontario, the village movement is often debated and the origin of the Iroquois in this location is not agreed upon. James Wright (1966) adapted the “in situ” hypothesis in his work on the Ontario Iroquois and extended Iroquoian continuity in the Northeast back several thousand years into the Laurentian Archaic times (Wright 1984). The in situ development of southern Ontario Iroquoians has become close to an accepted fact and forms the working model of many prominent Iroquoian archaeologists (Warrick 2000). However, some archaeologists question the in situ hypothesis. David Stothers’ study of Princess Point suggests that the origins were outside of Ontario, either in Ohio or Illinois (Smith and Crawford 1997; Stothers 1977:152-153). He considers Princess Point to have been ancestral Iroquoian and related to Owasco in New York. Dean Snow (1996) has also argued against the in situ model by stating that Northern Iroquoians migrated into their homelands around A.D. 600, from the Appalachian area of Pennsylvania. According to Snow (1996), the immigrant Iroquois speakers represented on the Princess Point sites in the Grand River Valley displaced or absorbed native Middle Woodland Algonquian speakers and had arrived with a full repertoire of Iroquoian traits such as palisaded long house villages, maize agriculture, and matrilocal residence/matrilineal descent systems. Dean Snow (1996) and others (Bursey 1995; Fiedel 1999) support the Iroquoian migration theory based on the archaeology, demographic, linguistic, and anthropological evidence (Warrick 2000). Figure SEQ Figure \* ARABIC 13: Cultural chronology of the Southeastern Lake Erie Basin and surrounding areas (Sullivan et al. 1996, Table 3.1. 1Brose 1976a; 2Lantz 1989; 3Ritchie 1965, Stothers 1977, White 1961, 1976; 4Schock 1974, 1976. The Early Late Woodland Period (A.D. 1000 1300) Sites in this time period are generally small and found on the lake plain and uplands (Sullivan et al. 1996:17). It is not known how much the natives relied on horticulture at this time and if the appearance of corn around A.D. 1100 helped increase sedentism (Ritchie 1969). The regional population of Early Late Woodland groups had a specific pattern of spatial organization and they tend to be distributed by geographic location. Lantz (1989) suggested that separate groups occupied the lake plain and plateau river valleys. He believed that the Allegheny Iroquois, who were located on the plateau, were mostly located in the upper reaches of the Allegheny River Valley and into the Cassadaga Valley. The Lake Plain Iroquois extended along the lake plain from the northeast corner of Ohio to New York (Sullivan et al. 1996). Lantz (1989:2) proposed that both groups were in place from roughly A.D. 940 to A.D. 1525, and both of these Iroquoian groups were Erie. Late Woodland mobility has been characterized by seasonal occupations between the lake plain and uplands (White 1963a). Marian White (1963a) believed that semi-sedentary seasonal movement for early Late Woodland villages was the most ideal. White’s idea was also supported by Brose’s village models (Brose 1976a), which were used exclusively in northeastern Ohio for the Riverview and Fairport Phases, two phases that range from A.D. 1000 to A.D. 1350. Brose recognized three different types of sites used for seasonal movement with his model and they included; a summer horticultural village, mid-summer campsites, and mid-winter campsites. Smaller villages are generally found along the lakeshore and near streams with secondary confluences. The summer camps are generally found near lacustrine environments while the winter camps are mostly found in uplands or beach ridge environments (Brose 1978b: 103-104). The Portage and Martin sites in the New York Niagara Frontier are two sites that date to the Early Woodland or Terminal Late Woodland time periods. The Portage Site was dated to around A.D. 1000 and is located on the Niagara River (Widmer and Webster 1981:51). The site is assumed a fishing camp based on faunal assemblages and dates to what is called the Hunter’s Home Phase (Ritchie and Funk 1973:119) which is a Late Middle Woodland phase with similarities to later cultures (Sullivan et al. 1996). On Grand Island, near the Niagara River in Erie County, New York is the Martin Site. Early Late Woodland pottery from this site exhibits a combination of those found in Ontario and Owasco of central New York (White 1976:112). This evidence suggests that this site may be a transitional phase; therefore, archaeologists have included this site in the Grand River Focus of the Princess Point Complex, a phase of the Niagara Frontier from around A.D. 1000 to A.D. 1300 (Figure 13) (Stothers 1977:27). The site was probably used as a continuous occupation fishing station due to the high amounts of marine faunal evidence discovered (White 1963a). In the Chautauqua/Allegheny area of New York, the Early Late Woodland (A.D. 1000 – 1300) is referred to as the Allegheny Phase (Figure 13) (Schock 1974). These sites are classified as non-farming, small, and found along the lake plain where they are probably seasonal fishing camps, except for the Westfield Site which is an earthwork site about 2.5 km away from the Lake Erie shoreline and relatively close to the Ripley Site (Guthe 1958; Widmer and Webster 1981:78). Archaeological evidence suggests that farming had probably taken place there. In Pennsylvania, horticultural villages in the Allegheny Valley, such as the Kinzua site (A.D. 1200) are found along floodplains (Dragoo 1977; Widmer and Webster 1981:78). This site was excavated extensively by the Carnegie Museum and turned up numerous structures and artifacts (Dragoo 1977; Schock 1974). The Allegheny Iroquois occupied this region of the Upper Allegheny River Valley for at least seven centuries, and merged from a Middle Woodland base. Analysis of ceramics indicate they were in contact with the Mead Island Culture and Monongahela groups to the south, Fort Ancient to the west, the Whittlesey focus of northeastern Ohio, and may have even absorbed some northern Iroquois ceramic traits with those living on the Lake Erie Plain. The general outlines of Allegheny Iroquois or Erie villages were small, compact and circular, and enlarging through time from 40 m to over 100 m in diameter (Lantz 1989:4). During the early period around A.D. 1000, houses were rectangular with six-meter perimeters (Lantz 1989:4). Two hundred years later by about A.D. 1200, the houses lengthened to average 6 m by 16 m and increasingly grew larger as time went on. These homes contained deep storage pits, near protected entrances; while surface hearths were often found between sleeping platforms (Lantz 1989). By around A.D. 1100, extensive interaction with the Lake Plain Erie is documented through ceramics by the appearance of the Sceiford Plain, Corded and Incised, early Ripley Plain, and Ontario Horizontal types. By about A.D. 1300, the ceramics indicate interaction with the Pickering Branch of the Ontario Iroquois Tradition as defined by Wright (1966). The Mead Island people were the dominant culture downriver from the Allegheny Iroquois territory in the area of Warren, Pennsylvania. It is not known if there had been any change in village patterning through time because complete sites have not been excavated. They were unique in that their homes were circular; resembling those of Monongahela; but inside each of these there was a large cylindrical storage pit (Lantz 1989). The Mead Island site had a 78 m occupation zone enclosed within a stockade and a double palisade trench. The ceramics from this location most likely are associated with those from Fort Ancient and Whittlesey of Ohio, but there is evidence of interaction with the Allegheny and Lake Plain Iroquois (Erie). Unlike the Allegheny Iroquois upriver from them who primarily preferred floodplains, the Mead Island group selected locations either on flood plains or on glacial terraces near major drainages (Lantz 1989:9). Another site, known as the Griswold Site, represents one of the many small lake plain sites found in Pennsylvania that are either fishing camps or small summer horticultural camps (Widmer and Webster 1981). The Griswold Site is found west of Erie, Pennsylvania on a small stream. The ceramics found at this site have been dated to roughly A.D. 1200. Although the soil at this site is a sandy loam and ideal for horticulture, environmental evidence and paleo-botanical data has not supported these notions. In contrast, one site that is known for its evidence of horticulture on the Lake Erie Plain and geographically near the Griswold Site is the Sceiford Site. This site is only 1.5 km south of the lake near the New York/Pennsylvania border and contained several circular structures as well as one large semi-rectangular structure (6-8 m on one side). Radiocarbon dates place occupation of this site from A.D. 900 to A.D. 1100 (Lantz 1989:14). The presence of maize, black walnut, an abundance of ceramics, lithics, and a plethora of faunal remains suggest long-term occupation (Sullivan et al. 1996). The ceramics at Sceiford infer links to the north with the Glenn Meyer of the Ontario Iroquois Tradition as well as some interaction with Iroquois populations east of the Genesee River (Lantz 1989:5). Lantz believes this site may be an early branch of Erie because the site is located only 8 km to the west of the Ripley Site and near the East 28th Street Site (Carpenter et al. 1949), which were both designated as Erie villages by MacNeish (1952). In Ohio, sites in the Riverview Phase (Figure 13) have only intermittently been excavated and are not well known. The Kurtz Site, located in the alluvial bottomland is one example of a Riverview horticultural village (Widmer and Webster 1981:62). The upland Avon Plant site is another, but is only thought to be a winter hunting camp (Widmer and Webster 1981:62). The Fairport Harbor Phase (Figure 13) is much more significantly recorded and is diagnostic through the ceramics and lithics. The ceramics at this time are grit tempered and cord marked to the rim, sometimes including oblique tool impressions or decorative circular elements. The lithics are described as broad based triangular points and with few other tools. The Fairport Phase at the South Park Site (mentioned below) represents a summer occupation (Brose 1973). Subsistence indicates a hunting and gathering strategy that may be supplemented by maize cultivation. Long-term occupation at the site may not have occurred during Fairport times because there was no consistency in orientation of the structures or any domestic feature like other sites of the Fairport Phase (Widmer and Webster 1981:69). The Fairport Harbor Site is another possible Fairport Phase horticultural village. This site seemed to represent a summer village, but may be a spring fishing camp (Widmer and Webster 1981; Brose 1976a: 39). The Walnut Tree Site was interpreted as being a winter hunting camp and has dwellings with evidence of storage for maize in the winter (Pratt 1979). The Conneaut Fort site was also designated as a winter camp (Brose 1978b), but Widmer and Webster (1981) noted the presence of certain species of bullfrog that pointed to a more permanent occupation because these frogs hibernate. This site dates back to A.D. 1340, and it is not clear whether this site should be placed in the Fairport Phase, the following Greenwood Phase, or a combination of both. The ceramics at this site fit better into the Fairport Phase and are more closely similar to those from western New York. This site is most likely a transitional site. The Terminal Late Woodland Period (A.D. 1300 – 1500) A change in settlement villages was the predominant form of societal organization around A.D. 1300. In New York and Pennsylvania, these villages tend to be discovered in the lake plains and upper lake plateau areas (Sullivan et al. 1996). Small, palisaded, and defendable hilltop sites with earth rings were generally found in the plateau region and these are more commonly found than lake plain sites during this time period (Johnson, Richardson III, and Bohnert 1979; Sullivan et al. 1996; Widmer and Webster 1981:79). According to Sullivan and others (1996:21), the internal compositions of upland sites are not well known. Like in the Allegheny Plateau, the village sites that have been documented along the lakeshore have been described as being small, palisaded, and include earth rings (Cheyney 1859; Edson 1875; Edson 1894; Larkin 1880, Thomas 1894; Turner 1850). It is important to note that it is probable that there were more lakeshore sites at one time but were lost due to lakeshore erosion, and thus has affected the archaeological knowledge that is present (Widmer and Webster 1981:79). Parker noted many “camps and villages” along the lakeshore from Cattaraugus Creek in New York to the border of Pennsylvania (Doty 1925). This masters thesis shed some light onto erosion rates of bluffs along Lake Erie and it hopefully may convince others to salvage or protect the remaining lakeshore sites. By 1300 A.D., interpretations from archaeological data indicate a more sedentary lifestyle and an increase in year-round occupation of villages. White (1963a) and Brose (1976) both suggest an increase in horticulture and a reduction in seasonal movement. Pollen profiles that date to this time indicate a climatic shift that started in A.D. 1300. This probably had an influence on subsistence patterns as a shift to warmer and moister conditions may have resulted in greater maize yields and perhaps subsequently facilitated population increases (Brose 1977). By this time, villages were predominantly relying on a mixture of the three sisters (corns, beans, and squash) as well as heavily hunting various game such as whitetail deer, turkey, elk, bear, etc. In New York, the village movement model (White 1978a) and the geographic location of a site is the primary criterion when used to date sites to A.D. 1300-1500. When one sees clusters of sites present, they are assumed to represent movements of one or more villages. It is important to note that sites with no European trade items but with ceramics complementary to those of the Contact Period are assumed prehistoric. White (1961, 1978a) recognized two arrangements of village movements in the cluster of sites in the Niagara Frontier. On the eastern side, she noted that the Goodyear, Newton-Hopper, and Simmons are located within 2.4 km (1.5 mi) of each other along terraces of Buffalo Creek; all were in defensible locations. The western-village movement is represented through (from Early Late) Buffam, Eaton, Green Lake, Ellis, and Kleis (Figure 13). In each of these sequences, the earliest sites of Buffam and Goodyear were right before contact. The Buffam site that was located in South Buffalo has no historic material in association with the village and it is the only site in the cluster that had an earth ring and ossuary burial; two traits that disappear by contact times in the Niagara Frontier Region. The little known Webster Site is near and possibly related to the Goodyear Site. In addition, the Nursery Site and the Harris Hill Site are located only 16 km (9.94 mi) north of Goodyear and are possibly small Iroquois villages dating to the 1400s. These might represent a village movement sequence. Another important sequence to note is represented by five sites that were documented by E.G. Squier in 1851 near Clarence, New York. Four of these sites contained small earthworks, between 0.2 and 0.6 ha large (0.49 and 1.48 acres), and closely spaced together (White 1976:124-127). One of these sites known as the Henry Long Site, had an occupation area that included a single long house, a palisade, and a ditch (White 1963b). The ceramics at this site suggest a date of around A.D. because they were similar to those found in the Uren and Middleport phases of the Middle Ontario Tradition. For a long-time, the Henry Long Site was the only one of Squier’s five sites that was found by archaeologists. The Christensen Site was possibly re-identified in a modern survey and is possibly a Middle Ontario Middleport Phase site (Rozenwig 1983). In this area near Clarence, a number of ossuaries have been found and are probably associated with these sites. White (1976:127) suggested that two or more of these communities may have interred their dead into a single ossuary, and she presumes these inhabitants of these sites to be ancestral Erie. In the New York Chautauqua/Allegheny area, the village movement model was also applicable. Schock (1976) believed there were four village-movement sequences in the area that he called the Chautauqua Phase (A.D. 1300-1525) (Figure 13). This phase is noted by the presence of shell-tempered ceramics, non-decorated vessels, small triangular points, and located on elevated knolls with palisades. A minimum of 15 sites in southwestern New York appear to represent this phase or possibly earlier related sites (Schock 1976). These sites lie within a minimum of a four-village movement sequence and include one near Clear Creek, Chautauqua County, New York; the other near central Cattaraugus County; a third near the Lake Erie coastal plain but a few miles inland from Lake Erie, and the fourth being on Cassadaga Creek, Chautauqua County, New York near Gerry and Sinclairville. Besides the Cattaraugus Site cluster that is located on a floodplain (mentioned below), the other Chautauqua Phase sites are in defensible, hilly positions, overlooking major stream valleys (Schock 1976; Widmer and Webster 1981). No villages of this phase have been found on the lakeshore. Schock (1976) believed that the Cattaraugus Creek clusters of sites are all related to a single community are near a floodplain, or terrace for this creek. Silverheels and Highbanks are two of these sites that are found near the Cattaraugus Reservation which contain historic period and earlier Iroquois artifacts. It was believed that the prehistoric portion of the Silverheels Site was possibly related to the Lake Erie Plain village sequence, but that this site was also possibly an historic burial site for those from the High Banks (Schock 1976:108). In all, Schock believes that the Chautauqua Phase sites appear to have been influenced more by cultures to the south than by the Iroquoian groups in the Niagara Frontier or in central New York (Schock 1976:106). Similar to New York, Pennsylvania has a few village sites located near the Lake Erie Plain. These pre-contact sites are more plentiful in the Allegheny and French Creek Valleys. According to Dragoo (1977), excavations at Onoville in this region revealed longhouses, numerous small circular buildings, and a palisade. Other sites include the McFate Site and Wilson Shutes Site. Widmer and Webster (1981) have noted a change in the settlement and subsistence patterns for the area of northeastern Ohio in the Late Terminal Woodland. By around A.D. 1400, there was a change in seasonal movement and a more year-round occupation of villages (Brose 1976a). While White proposed increased horticultural in New York at this time, the same cannot be attributed to the northeastern Ohio area. One explanation is that there was a possible climate shift beginning at about A.D. 1300-1350, which could have affected the subsistence patterns. A shift from cool and dry to generally warmer and moister conditions may have increased yields of maize and promoted a general reliance on horticulture. Evidence for this shift comes from pollen profiles (Brose 1977). Three types of sites are known in Ohio for this time period of the Greenwood Phase (A.D. 1350-1500), which was proposed by Brose (1977) (Figure 13). These include small seasonal intermittent economic camps, small spring horticulture villages, and winter-spring camps divided into lakeshore fishing and upland hunting. The villages are usually located in defensible positions near farmable soil. Alternatively, Widmer and Webster (1981) have proposed a settlement model of permanent, year-round encampments and villages, as well as seasonal campsites. The most excavated site of this phase was the South Park Site by Brose (1973). The artifacts here are the same as the Fairport Phase except the projectile points are more associated with Madison rather than the Levanna-type that is common in New York and Pennsylvania and earlier phases. Other sites of this phase include Eastwall, Conneaut Fort, Seibert, Greenwood Village and Tuttle Hill. Located on a bluff near the Cuyahoga River, the Siebert site was only limitedly investigated. Being only 40 km from Lake Erie, no structures or palisades were found. The Ahlstrom, Greenwood Village, and Tuttle Hill sites are small agricultural villages, while the Eastwall site was quite large (Brose 1977; Brose 1978a; Widmer and Webster 1981). The Conneaut Fort Site (Brose 1976b) is a hilltop site with an earthen enclosure covering nearly 0.8 ha (1.97 acres). Other Greenwood Phase sites include Bennett (Brose 1977), Carey Hill (Brose 1976b), and Rogers #1 and #2 (Brose 1977). Pre-Contact Neutral and Huron-Petun In southern Ontario, a group known as the Neutral have been documented. Longhouses were prevalent on their villages and their growth surged during the 15th and early 16th century (Warrick 2000). This architectural growth was probably due to a demographic increase throughout this time. However, by the early 17th century, longhouse lengths dramatically decreased with an average of only 30 m (98 ft) for Neutral homes by A.D. 1580 (Lennox and Fitzgerald 1990:445). Reasons for this change could be attributed to a switch in village government, from dominant matrilineages to clan segments, enabling more accepting longhouse membership (Warrick 1996:20). In correlation to the growth of longhouses in the 15th and early 16th centuries, pre-contact Neutral as well as the local Huron-Petun sites expanded to cover large areas of 4.0 ha (9.98 acres) or more. Around A.D. 1400-1450, the defensive setting of settlements on high ground by forks in streams as well as multiple palisades in Ontario Iroquoian and St. Lawrence Iroquoian sites may indicate that warfare was probably the major factor behind such growth in settlement sizes (Trigger 1985). A number of reasons could have contributed to warfare; however, the presence of St. Lawrence Iroquoian pottery and abundance of Onondaga chert tools in the late 15th century Huron-Petun sites in the Toronto region (Williamson and Robertson 1998) suggest a peaceful interaction with Neutral (trade) and St. Lawrence Iroquoians (marriage exchange) (Warrick 2000). In the 15th and 16th centuries, interaction between Iroquoians and Algonquians in Ontario was both aggressive and peaceful. On the belligerent side, the Neutral appear to have been engaged in a war with the Central Algonquian, known as the Western Basin Tradition in extreme southwest Ontario (Warrick 2000). In the late 15th century, multiple-row palisades and earthworks surrounded Neutral villages. Earthworks were evident at Clearville, Southwold, and Lawson and there is evidence at all of these sites of shell-tempered Western Basin pottery. Fifteenth century Western Basin sites were also encircled by earthworks such as the Parker and Weiser sites near Sarnia (Murphy and Ferris 1990:257-259). Hostilities seem to have increased during the 16th century, indicated by the withdrawal of the Neutral from the Chatham and London regions to lands east of the Grand River by A.D. 1550 (Lennox and Fitzgerald 1990:438). In contrast to the Neutral, the Huron-Petun experienced peaceful existence with the Algonquian neighbors to the north. An increase in trade between the Huron-Petun and Shield Algonquians beginning in A.D. 1450 can be inferred from village site distributions in which the Huron sites gradually appear further north closer to their northern neighbors (Warrick 2000). Contact Period (A.D. 1500 – 1650) Notable and distinct changes in settlement patterns occurred in the region by A.D. 1550. There seems to be a shift toward selecting lower elevations but still easily defensible positions. Being near good agricultural soil was continually important and another climate change probably contributed to this factor. This climate shift, named the Neo-Boreal episode, was a time of cool, moist conditions that could have made higher elevations unsuitable for horticulture, while the milder conditions along the lake might have been more favorable. Tightly packed and fortified villages began to appear in the archaeological record during this time. The widespread use of longhouses and palisades suggest that there could have been an increase in warfare, competition, and population pressures, therefore groups would have to be more closely associated (Sullivan et al. 1996). It has been proposed that this increase in warfare was probably due to the climatic shift, food resourcing, and population pressure by nearby tribes (Widmer and Webster 1981:76). In the Pennsylvania region, the plateau valleys were abandoned and the sites became increasingly situated on the edge of the plateau valley near the Portage Escarpment overlooking the Lake Erie Plain. Most of these sites are located on defensible terrain along minor streams. There seems to have been a general abandonment of sites around the contact period in this region. The sites in the Pennsylvania area of the Allegheny Plateau were abandoned and even those sites on the Niagara Frontier shifted toward the Lake Erie Plain. Interpretations of the archaeological record seem to suggest that these abandonments occurred at the same time as distinct changes in settlement patterns occurred. As mentioned, more sites have been documented near the lake and near lower elevations than in the previous Terminal Late Woodland Period. This subsistence shift is probably due to the need to have access to better agricultural soils that were located near and along the lakeshore (Sullivan et al. 1996). Another hypothesis for this population shift could have been due to the climate (as mentioned above). The Little Ice Age, a period of cooler and moist air flow with average temperature 1- 2 °C lower than it is today (2015) possibly made the uplands unsuitable for farming and horticulture, therefore those sites near the lake plain would have appeared to be much more suitable because of the milder weather conditions (Johnson, Richardson III, and Bohnert 1979). Analysis conducted on Lake Erie drumfish otoliths dated from archaeological sites in this area indicate lower summer temperatures from A.D. 300-700 (Johnson, Richardson III, and Bohnert 1979). The archaeological evidence of the region supports this population shift as many sites that date to this time period are found along the lake plain, and sites found in the upland plateau regions seem to be abandoned (Widmer and Webster 1981). In Ohio, archaeological evidence suggests a reduction of villages in this time period as well, and archaeologists believe that it is probably for the same reasons. Brose (1978b) has suggested that the absence of European trade goods on sites of this time period in Ohio is companionable with a population decline before 1650. All of the sites in Ohio, like its Pennsylvania and New York counterparts, were found to be in defensive positions near streams; however, those sites in the northeastern portion are located on steep bluffs and not near the lake plain (Brose 1977:33). Widmer and Webster (1981:74) suggested that the Ohio sites did not occur along the lakeshore because of the rise in sea level and erosion of the shoreline; it is unknown as to why the Ohio area has different village patterns than the Niagara frontier. As more Europeans were making contact in the Southeastern Lake Erie Basin Region, the Niagara Frontier became more important to the regional economy because it was rich with trade goods such as beaver furs. However, the spatial location was not ideal because of its distance from Dutch, British, and French settlements. With this in mind, trade probably had to flow through neighboring groups such as the Huron and Seneca (Sullivan et al.1996:23). European trade items started to appear as early as the early 16th century, but the exact time is unknown and this influx of European goods slowly increased as trade routes became more secure over the region (Sullivan et al. 1996:23). The later sites of the two-village sequence that was postulated by White for the Niagara Frontier region in New York date between A.D. 1540-1650. Both of these village communities tended to move in the same direction and only 13-16 km apart, which suggests they may be socially-politically connected. It has been documented that contemporaneous Seneca, Mohawk, and Onondaga communities were separated by around the same distances, suggesting this distance was not random, but some kind of unidentified requirement by inhabitants of the region. Mentioned earlier, the site of Buffam on the western village sequence of the Niagara Frontier was followed by Eaton, Green Lake, Ellis, and Kleis. The Ellis and Kleis were the most recent in this community (White 1961) and the most diverse in terms of ceramics. These sites produced fine scrapers, which could have been used for the fur trade in this region. In the eastern village sequence, following the Goodyear Site was Newton-Hopper, Simmons, and Bead Hill. The first two were within 3 km of each other in defensible positions on terraces overlooking Buffalo Creek. Bead Hill is south of these sites and was partially destroyed due to land development. The best data from these sites comes from the Simmons Site that was excavated by Marian White in the 1960s as part of a field school. Kleis and Bead Hill sites appear to be the latest of these communities. They are not set on easily defensible terrain and White (1961) noted that a similar change from defensible to non-defensible positions occurred with the Seneca near this area from A.D. 1630-1650. Use of the village-movement model has been met with only little success in the Ripley Site area in the Allegany/Chautauqua region. A major problem in application of this model is the lack of contemporary, large sites. Those who support the village movement model, including MacNeish (1952), White (1961) and Wright (1966) all suggest Ripley materials are very similar to the Goodyear site in the Niagara Frontier. MacNeish (1952) considered Goodyear ancestral to Erie, and Wright (1966) believed that it was contemporary with the Green Lake site. Carpenter (1949) noted “striking parallels” between the ceramics from Ripley and the Erie, Pennsylvania East 28th Street site (more information on the Ripley Site is in Chapter 7). Johnson (Johnson, Richardson III, and Bohnert 1979:79) placed Ripley ten years earlier than the East 28th Street Site, which he believed dated to A.D. 1635. Also in this area, the Silver Heels site had abundant trade material. Some items were probably not directly obtained but were exchanged through groups like the Seneca. Unlike Ripley, this site has Seneca-style ceramics that are also found at the Ellis and Kleis sites in the Niagara Frontier Region (Schock 1974). It is unclear why trade was going on between these groups, but it was perhaps to make relations for beaver trade in the region. In Pennsylvania, the East 28th Street Site had more historical material than the Ripley Site and dates to around A.D. 1630-1645 (Carpenter et al. 1949). However, like Ripley, the East 28th Street Site does not appear to be related to sites nearby. Numerous burials and trade beads were located at this site but it was destroyed in the 20th century to make space for a gravel quarry and eventually land development where the modern streets of Parade and E 28th are located in the city of Erie. The Winter Green Gorge Site is one of the sites that were located on the Portage Escarpment and also near the East 28th Street Site (Widmer and Webster 1981). This site is located on a high bluff above Four Mile Creek and a portion of the earth ring may still be visible. The ceramics found here are known as McFate Incised, which is common to this region of Pennsylvania, and another has been noted as a Whittlesey type (Murphy 1971). In Ohio, similar to the Greenwood Phase settlement model proposed by Widmer and Webster (1981), the South Park Phase (A.D. 1500-1650) settlement system included permanent villages and seasonal camps. These were either small special-purpose campsites located near the villages or aquatic resources (Brose 1976a). The South Park Site is the best-known example of the South Park Phase. Several longhouses with central interior hearths were oriented in a north-south direction and enclosed by a palisade (Brose 1973; Widmer and Webster 1981). The Tuttle Hill Site may also be a South Park Phase village because it is similar in configuration as the South Park Site (Widmer and Webster 1981:75). Other sites include Fairport Harbor, Lyman, Greenwood, and Ambler Metropark #2 sites (Brose 1976a; Brose 1978a). Later Whittlesey (A.D. 1450-1600) ceramics are most associated with the South Park Phase and are tempered with grit shell, and one common type, Tuttle Hill Notched. Numerous investigators noticed a similarity of this type to various Seneca types including Seneca Barbed Collar and Dutch Hollow Notched. Trade goods dating to the contact period were also found at Fairport Harbor (Morgan and Ellis 1943), and date to the South Park Phase as well (Brose 1976a). Historic Period (A.D. 1640 – 1656) Sometime between the years A.D. 1640-1656, a series of inter-tribal wars led to the displacement of several communities that once resided along Lake Erie including the Erie in A.D. 1656 (see the section Ethnohistorical Information on the Demise of the Erie and Huronia below). During this time, the Five Nations of the Iroquois sought to establish control over the region and defeated, dispersed, or incorporated the Neutral, Wenro and Erie tribes that resided in the Southeastern Lake Erie region. The information on any of these groups is rare and it has proved difficult to determine their cultural boundaries (Sullivan et al. 1996). Post-Contact Neutral The Neutral (same group as mentioned above) have been given detailed attention by a handful of authors, and they generally resided near the Ontario Peninsula (Coyne 1895; Fenton 1940; Harris 1896; Houghton 1909; Jones 1909; Lennox and Fitzgerald 1990; Warrick 2000; White 1961, 1972, 1978b). Their confederacy resided near Hamilton, Ontario but many groups were also near the Niagara River around the 1630s (White 1972). Overall, their villages covered an unknown extent east and west, and many placed them in the Niagara Frontier (White 1961). Ethnohistorically, the first to document the Neutral as “la Nation neuter” was Samuel De Champlain. He located them as “two days’ journey from them [Hurons] in a southerly direction [where the Neutrals live] westward of the lake of the Onondagas” (Biggar 1922). This lake was Lake Ontario. Other information about the Neutral country comes from Father Daillon, the Recollect, who visited several of their villages in A.D. 1626. He never talked about the Niagara River, but at one point he mentions a river that could be the Niagara (White 1961). He mentioned that the Neutral wanted to trade with the French in the region but did not know the route. An Algonquin named Yroquet who trapped beaver in the Neutral territory at that time, refused to share the information of the mouth of the “River of the Iroquois” which would take them there (Sagard-Theodat, 1866:798 cited in White 1961). Authors have noted that the “River of the Iroquois” usually referred to the St. Lawrence, but in this particular case, Severance (1917:16) thought it might actually be the Niagara River. It is difficult to understand the Neutrals’ lack of knowledge of the mouth of the Niagara, especially if any of their villages were located near it. However, this reason given by the Neutrals to Daillon might have been intended to satisfy the Hurons present who were trying to prevent the direct trade between the French and Neutral (White 1961). It is possible that the Neutrals’ knowledge of the Lake Ontario-St. Lawrence route to Quebec was limited by their practice of land transportation, as they have been known to be pitiable canoesmen (Fenton 1940:188). Their only enemies were the Central Algonquian in the West (mentioned above) so their knowledge of the eastern geography might have been limited (Biggar 1922; White 1961). There is no conclusive evidence locating the Neutral villages east of the Niagara River in A.D. 1626. A later reference placed the Neutral on the east side of the Niagara River, as well as the west. Father Lalemant (Jesuit Relations) reporting on the journey of Fathers Brébeuf and Chaumont to the Neutral in A.D. 1640 stated that there were about forty villages and the closest was 40 leagues south of the Huron (Jesuit Relations 21:209). This report included information that implied that the Neutral territory included both sides of the Niagara River, and that the village of Onguiaahra was near the river Onguiaahara or Niagara. The name Akhrakvaeronon appears on the Nouvelle France map east of the Niagara River (Steckley 1985). This group may be a tribe that the Huron translate to being “easterners”. Little is known about them, except that the Iroquois defeated them in A.D. 1652. It has been suggested that this was a Neutral group, but the name Akhrakvaeronon does not appear on the Sanson maps (Figure 14) and their location is uncertain (Heidenreich 1988:100). White (1961) has suggested that the Van Son site on Grand Island is a Neutral site and 59 burials in both ossuaries and single interments were found here. The habitation site for this cemetery is unknown, but the Burnt Ship Site only 0.8 km away poses as a possible location. Post-Contact Wenro Another group known as the Wenro were first documented in the Jesuit Relations, which states that they were associated with the Neutral and lived east of them, but not much is known ethno-historically of this tribe (Thwaites 1896-1901:17). Other written accounts suggest the Wenro were more associated with the Erie around southern New York and Pennsylvania, thus far from the Neutral (Thwaites 1896-1901:21). Stothers (1979) made a suggestion that suggests the Wenro were found in southwestern New York, near the vicinity of the Ripley Site. However, it is generally believed that the Wenro joined the Huron Confederacy once displaced out of their territory by the League of Iroquois in A.D. 1638 (Thwaites 1896-1901:21; White 1978). The exact location of the homeland, village number, and population size of the Wenro are uncertain, but the Kienuka and Shelby sites between Neutral and Seneca site clusters just south of Lake Ontario are possible candidates for late 16th century to early 17th century Wenro villages (Engelbrecht 1991:4; Warrick 2008:225; White 1978b: 409). Based on the lack of 17th century Iroquoian sites north of the cluster of Erie sites, south of Buffalo, New York, it is possible that the entire Wenro nation occupied just one village (Engelbrecht 1991; White 1978a). If the inferred association of Genoa Frilled or “frilled” pottery (Hawkins 2001) with Wenro ethnicity is correct (Ridley 1973), the best candidates for 1630s Wenro sites in western New York State are the Ellis, Kleis, and Silverheels sites (Engelbrecht 1991:9; Warrick 2008:225). There is evidence for a Wenro migration to the Wendake based on high percentages of Genoa Frilled pottery on certain sites occupied during the 1640s (Warrick 2008:225). Pottery analysis from Wendat sites found that the frilled pots were locally manufactured and appeared in a few A.D. 1635-1650 village sites that were historically related and geographically close to the village of Ossossane, a Wendat community that offered refuge to the Wenro (Thwaites 1896-1901; 15:159).  CHAPTER 4: KNOWN HISTORY OF THE ERIE Cultural History of the Erie Erie was the name attributed to several tribes culturally and linguistically related to the Huron, Neutral, Five Nations, and other Northern Iroquoian peoples such as the Wenro. The Erie have been referred to by many different names that include Erie, La Nation du Chat, Raccoon Nation, Eriehronon, Rigueronnons, Gentaguetehronnons, and Ehressaronon. All of these were synonyms to the name Erie or were allied groups that fall under the headline of the Erie Nation. The French Jesuits, however, referred to the Erie by the name the Huron gave them of Eriehronon, or as La Nation du Chat (Sullivan et al. 1996:25; Viola 1976; White 1961). While the Nation du Chat is sometimes referred to as the “Raccoon Nation”, it is suggested that it probably translated best to the “Cat Nation” (Wright 1974:69). Erie, as translated from the Huron, is “it is long tailed”, referring to the eastern puma or panther (Hodge 1907:430). Their identification as a Northern Iroquois group is based solely on the seventeenth-century French statements because no Erie words were ever recorded. White (1961:41) believed that the French used the term geographically to refer to all the Iroquoian-speaking people south of Lake Erie. The Erie were defeated and displaced before any Europeans had the opportunity to document accounts of them; therefore, all European documentation regarding the Erie was second-hand (Parker 1907). The general location of the Erie was first published in A.D. 1647-1648 and a similar reference was said to have been written in 1644-1645 by Gendron, a 17th century Frenchman who wrote “This Lake called Erie, was formed inhabited on its southern shores by certain tribes whom we call the Nation of the Cat; they have been compelled to retire far inland to escape enemies, who are farther West” (Gendron 1868:8-9). The Erie inhabited an area from the southern lakeshore of Lake Erie to below Pittsburgh, Pennsylvania in the Allegheny drainage, and from Buffalo, New York west to Toledo, Ohio within the Central Lowlands Province and the Appalachian Plateau Regions (White 1978a). Hoffman (1964) extends their territory as far as Virginia by equating the Erie with the Pocaughtawonauck, Massawomeck, Richahecrian, and Rickohockan His assignment of such a vast area based on weak evidence seems unacceptable and differs greatly with the size of territories identified with the Huron confederacy, the Iroquois League, or the Neutral (White 1978a).. The Jesuit Relations of A.D. 1647-1648 and Gendron (1868) both mention that the Erie resided along the southern shore of Lake Erie. Cartographic evidence for the location of the Erie confirms these accounts (White 1978a). A map titled Novelle France, possibly drafted by Jean Bourdon in A.D. 1641, placed the Erie in this area and illustrates the locations for other native groups prior to A.D. 1650 (Heidenreich 1988). In addition to this, Nicolas Sanson placed the Eriehronon or Erie people between the eastern end of Lake Erie and Lake Chautauqua on his maps that date A.D. 1650, A.D. 1656 and A.D. 1657; however, it is unknown whether they occupied an area that extended near Buffalo, New York (Figure 14). Numerous other cartographers noted the Erie location but these maps were made after the Erie were defeated and dispersed (Bernou ca. 1680 in Delanglez 1938:115; Franquelin 1684 in Delanglez 1943; Du Creux 1660 in Thwaites 1896-1901:1; White 1978a). Figure SEQ Figure \* ARABIC 14: Nicholas Sanson Map from A.D. 1650. Note the Presence of "N. du Chat" in the Center of the Map. The true extent of the Erie is generally unknown because no village of the Erie can be documented from either the literature or maps. The limited archaeological evidence allows a choice across a wide area of western New York, Pennsylvania, and northern Ohio. Various archaeologists have identified Late Woodland sites as Erie across this entire region and even into western Ontario. Most of these identifications are based on assumptions about the location of the Erie that cannot be supported (White 1978a). Usually the best evidence that can be anticipated would be from sites that have European artifacts attributable to European trade of the early 17th century, but linked to earlier prehistoric sites. Gendron (1868:8-9) stated in his work that he believed the Erie were forced to move inland from enemies in the west. Marian White (1972) agreed with this idea due to various site abandonments in the Niagara Frontier at the time. White (1972:62) believed that the Erie were probably forced out of the Niagara Frontier by invading groups of both Neutral and Seneca who sought to exploit the beaver trade in this region, but not enough information is known. This dispersal would have happened between A.D. 1654 and A.D. 1656 (White 1961). Marian White’s 1961 dissertation made a good argument that the Erie resided in the area of western New York and Pennsylvania, extending into the Niagara Frontier Region. There were a number of groups, as mentioned, that fall under the title of “Erie”. It has been noted that the Erie had around” 2,000 or 3,000 warriors” (Thwaites 1896-1901:42). A total estimate of the entire Erie population was 14,500 at the time of their last war in the mid-1600s (Hodge 1907:431). It is possible to infer the sociopolitical structure of the Erie from that of the Huron and Five Nations Iroquois (Tooker 1964). The Huron and the Five Nations were alliances composed of more than a single tribe. Each tribe, as mentioned, consists of one to three main villages, with this being a slightly varying number (White 1978a). It is likely that the Erie had a similar structure. Erie probably referred to a group of tribes or even an alliance, and probably dwelled in one or more villages. The name of a village was often applied to a tribe by the French and presumably the Huron as well. These names may refer to a geographical feature and were not applied to the new community when a village moved to a new location (White 1978a). The names of two Erie villages have been recorded: Gentaienton and Rigue. One of these groups, known as the Gentaguetehronnons, or the people of the village Gentaienton (Thwaites 1896-1901:61), appear in a list from A.D. 1679 (JR 61:195). Their name is roughly translated to “the people who bear or carry a field” (Steckley 1985:12). This tribe was not mentioned before A.D. 1655-1656 and may only have been known because word of their demise had spread. Another allied group, the Rigueronnons, possibly occupied the village site Presently E. 28th Street in Erie, PA. Rigue in Pennsylvania (Thwaites 1896-1901:42). Rigué was an Erie village that was contemporaneous with Gentaienton and was frequently mentioned in historic documentation (Jesuit Relations 42:186; White 1978a). The reference between the earliest and the latest reference is about 25 years, which does not exceed the length of time that a village might remain at one location. Therefore, it is possible that Riguehronnons were a group of people that the French settlers placed under the general headline of the Erie (White 1978a: 412). Potential Erie Sites in Niagara Frontier Area of New York A group of sites in the Niagara Frontier of northwestern New York State region display characteristics of being Erie, and have been identified as the communities of two contemporary villages. The movement of this pair of villages has been traced in a northeasterly fashion from about A.D. 1550 to around A.D. 1635 (Terminal Late Woodland) (see Chapter 3). The majority of these sites have pre-contact earth rings or earth banks that elsewhere date no later than A.D. 1530-1550. Some sites in this sequence northeast of the eastern village sites are probably antecedent and may be earlier around A.D 1175-1200 (White 1978a). Research has shown that estimates would place most of them prior to A.D. 1450 (White 1978a). The latest communities living at Bead Hill and Kleis had moved out of the area by A.D. 1640. The location of these sites and the disruption in their village movement pattern signals their removal from the Niagara Frontier and coincides with the contemporary ethno-historical accounts. These two sites may be identified as Erie, prehistorically as ancestral Erie, and probably exhibit the movement of single tribe (White 1978a). However, none of these sites have been linked to any historic sites and questions of ethnic identity are unsettled. Potential Erie Sites in Western Pennsylvania and Western New York Near the Ripley Site in western New York and northwestern Pennsylvania, the Mercyhurst Archaeological Institute (MAI) has excavated a few potential Erie sites (Quinn et al. 1998). The Orton Quarry Site (36ER243) (Quinn et al. 1998; 2000) was an ossuary situated in a gravel pit around 3.4 km (2.1 mi) east of the boundary of the Borough of North East, Pennsylvania, 2.6 km (1.6 mi) south of Lake Erie, around 9.0 km (5.6 mi) east of the New York State line, and 10 km (6.2 mi) as the “crow flies” from the Ripley Site. Excavations revealed 14 cultural features, including refuse pits, fire pits, mixed/use pits, a bell shaped storage pit, and a conical pit with a posthole. The largest and most revealing feature was the ossuary that contained a minimum of 68 individuals and included skeletal remains from numerous occupations with carbon dates that range from A.D. 1050-A.D. 1580. Another site near Ripley was the Tracy School Site (36ER287) (Quinn et al. 1998). This site was located in Millcreek Township, Pennsylvania just west of the City of Erie. During the spring of 1996, MAI conducted Phase III excavations at a road intersection for the expansion of a school. A total of 13 cultural features excluding post-molds were located, most being single use fire features, as well as 13 post-molds. A total of 4,254 artifacts and ecofacts were recovered, including flaked stone, ceramics, fire-cracked rock, faunal remains, and floral remains. Two lithic points were found and one was identified as a Vosburg, which is ascribed to the Late Archaic Laurentian tradition (Quinn et al. 1998:10). The fire feature at this site was radiocarbon dated to a date of A.D. 1630, which is very close to the date of the East 28th Street Site and only 6.8 km (4.22 mi) “as the crow flies” away. The Elk Creek Site (36ER162) is located in Girard Township, about 27.4 km (17 mi) west of the City of Erie, and 13.5 km (8.39 mi) east of the Ohio state line (Quinn et al. 1998). The site was situated along a modern floodplain for Elk Creek and the primary use of the site was potentially a special use site or satellite area from areas larger that have not yet been discovered. Five cultural features were identified, including two fire pits, two post molds, and a midden. Four of these features occurred within the large excavation block at the eastern boundary of the study area. A total of 5,861 artifacts were recovered from the Elk Creek site that included flaked stone, prehistoric ceramics, and historic artifacts. The ceramics were in poor condition and no individual vessel typology could be applied to them. Most of these sherds however were grit tempered with quartz, or granitic gneiss. The site was given a date of A.D. 1470. These three sites span the entire Late Woodland period (1000-1650), and Orton Quarry alone provides evidence for most of this time span. Based on 14C dates, Orton Quarry was the earliest (A.D. 1050 and A.D. 1170) and expanded to near the Historic Period of about A.D. 1550. The majority of the Orton Quarry dates range from A.D. 1290-A.D. 1430. The principal use of this site appears to correspond to the initial use of the Tracy School site, which has dates around A.D. 1340 (Quinn et al. 1998). The later occupation of Orton Quarry falls within the same time period as the Elk Creek site, which conformably overlaps the upper date range of the final use of the Orton Quarry site. Although this area was habited continuously throughout the Late Woodland, none of these specific sites was (Quinn et al. 1998). Based on the data, it was suggested (Quinn et al. 1998) that during the Late Woodland the Elk Creek and Tracy School sites were special activity locations. Quinn and colleagues believed that the parent community for Elk Creek lies directly west of the site on the uplands above the mouth of the creek, while the village associated with Tracy school may lie north of the site near Presque Isle Bay. Orton, with its apparently thrice-used ossuary (regarding its size and abundance of features) is clearly a location of a series of much more substantial habitations than the Elk Creek or Tracy sites. With these three sites in mind, the ethnic identity of the populations remained unclear. While it clearly appeared that the Whittlesey ceramic and ethnic boundary lies somewhere west of Elk Creek and that the Lake Erie Plain have traditionally assumed to be proto-Erie or Erie, there is little data in Pennsylvania to support this (Quinn et al. 1998). Other potential Erie sites in this area and near the Ripley Site include the Peacock Site at the mouth of Chautauqua Creek, Chautauqua County, New York; the Barber Site (36ER308) at the mouth of Twentymile Creek in eastern Erie County (Quinn 2003); the Reese Site (36ER63) at the mouth of Twelve Mile Run, Erie, PA; the Sommerheim Park Site (36ER155) located on the Lake Warren strandline overlooking Presque Isle Bay, Erie, PA; the Griswold Site (36ER56) located in Millcreek Township on the eastern bank of Wilkins Run ca. 1.2 km south of Lake Erie [see The Early Late Woodland Period (A.D. 1000-1300) Chapter 3]; the Westfield Site, Westfield in Chautauqua County, New York which was very close to Ripley and also had an earthwork present; East Wall (36AB40) at the mouth of Conneaut Creek, Ashtabula County, Ohio (Johnson et al. 1979:79); Burning Springs (White 1961:129); and the Sceiford site (Lantz 1989)[see The Early Late Woodland Period (A.D. 1000-1300) in Chapter 3]. History of the Erie Presence in Ohio West of the New York-Pennsylvania line are Late Woodland sites that correspond with the Whittlesey tradition. Brose (1976; 1978) has indicated that Whittlesey probably does not last into historic times and is probably not Erie. Brose (1994) has giving three distinct date ranges to the Whittlesey Tradition: The earliest Whittlesey Phase (A.D. 1000-1250); the Middle Whittlesey Phase (A.D. 1250-1450); and the final Whittlesey Phase (A.D. 1450-1600/1620) (See Chapter 3). Archaeologically, the Erie have been placed in the northern Ohio territory since the first explanations were being offered for Ohio’s prehistoric occupants by Fowke (1902) and Shetrone (1919) (as cited in Bush and Callender 1984:33). It is important to note that these interpretations were made before any scientific excavations took place (Greenman 1935; Morgan and Ellis 1943). Greenman (1935) made the first attempts to tie the Erie Indians into Ohio archaeology. He places the Reeves site of northeastern Ohio into the Iroquoian Aspect of the Upper Mississippian classification and notes that the occupation points mostly to the Erie (Greenman 1935:9). This classification was based on the location because it seemed to fall in the general area where the Erie were believed to reside. Morgan and Ellis (1943) indicate in their excavations of the Fairport Harbor Site that they must relate to the Whittlesey Focus and even claim Erie affiliation. This information was also not perfect because it was based on the “facts” that the Erie Indians spoke the same language as the Huron (Wyandot), lived in the Southeastern Lake Erie Basin, they tilled soil, and historic artifacts relating to early 17th century were found. Richard Morgan (1952) continued the argument for the Whittlesey Focus in which he cites both himself in Morgan and Ellis (1945) and Greenman (1935). He indicates it is possible they are Erie Indian sites because the Erie were destroyed as a functioning group by 1654, and the presence of historic materials dating to the early 17th century would make sense. However, like many of their contemporaries, Greenman, Morgan and Ellis, and Morgan all relied on the assumption that the Erie occupied the southern shore of Lake Erie, with no substantial archaeological evidence. Fitting (1964) attempted to understand the Whittlesey Focus and developed a chronology for the most prolific of the Whittlesey sites (Fairport Harbor, Reeves, Tuttle Hill and South Park, all in northern Ohio). All of the historic data that was found by Morgan and Ellis, Greenman, and Morgan were either dated incorrectly or out of context because they were found on the surface. Fitting (1964) suggested that there were two distinct groups that could be identified through ceramic assemblages from the sites. He suggests that earlier Algonquian speaking groups (as seen by the Fairport Harbor assemblage) was replaced by a later group that could have been the Erie. Brose (1976) in contrast, had identified that the Whittlesey Focus was a continual occupation in the northeast Ohio region that can be dated from A.D. 900. He sees no replacement groups as proposed by Fitting, and suggests that there was occupation of the South Park Site into the 17th century. Brose (1976) argues that by the Late Whittlesey Focus, archaeological evidence claims this area to be Algonquian groups and not of Erie occupation. It is generally accepted today that the Whittlesey Focus is made of Algonquian groups. Ethnohistorical Information on the Demise of the Erie and Huronia In the 1640s, the Erie began to receive a lot of hostility from the Five Nations when they made an effort to destroy their historic rivals and enemies, the Huron. The ethnohistorical accounts derive from oral tradition and most of this information is not written fact, but as an anthropologist these accounts are useful to better understand this Nation. With their location being above Toronto, Canada, the Huron had first contact with the French. The Huron and Algonquin both had been fur and trade partners with the French ever since 1609 when Samuel de Champlain aided the Huron war party to fight Mohawks near Lake Champlain. Since this event, the Five Nations Iroquois had a grudge against the French and Huron. Within twenty years, the Iroquois obtained firearms from English, French, and Dutch traders. With these weapons, the Five Nations became a huge force in the Indian vs. Indian frontier in the Northeast (Lupold 1975). The Huron believed they were the Wendat (in their own language it was Sendat). The English changed the name and named them Wyandot, however, the French referred to them as a loose generation of tribes that includes the Tobacco tribe and the “Huron de la Nation Neutre”, as Hurons. Champlain referred to the Huron as the “Good Iroquois” to distinguish them from the hostile tribes. To the west and southwest of the Huron country, lay the land of Tionontati, Tobacco Huron, or Tobacco Nation. Further south and mostly westward from the Niagara River were the villages of the Neutral, with whom the Erie were often grouped. All of these groups belonged to the Huron-Iroquois linguistic stock, but all were rivals. By 1615, the Huron were extremely powerful and successful agriculturalists who were famous for their tobacco. Huronia consisted of at least 18 fortified towns and their population has been estimated to around 30,000 (Parker 1907:528). Plagues, war, and famine reduced these numbers for the Huron, Tionontati, and Neutrals, until 1641 when their strengths dipped below 20,000 and maybe even as low as 10,000 members. The Jesuits among the Huron continued to stir up an anti-Iroquois sentiment and by 1639, the Huron burned 113 captured Iroquois. Seneca and Mohawk war parties had violated a League truce, which was negotiated by the Huron, and by the winter of 1648 they raided Huron villages. This small army of 1,000 men fell on Huronia and many Hurons fled. Estimations suggest 10,000 Huron were killed. Their prisoners were tortured and the Huron group remained enemies to the Iroquois for a century. The Iroquois on a path for power in the region to control the fur trade fought a series of destructive wars against various neighbors. In 1649, they defeated the Tobacco People, the Neutrals in 1651-1656, and finally the Susquehanna in the 1670s. This long period of war weakened the Iroquois, but they regained strength by adopting huge numbers of the defeated tribes into their own group. The Iroquois continued to be a strong tribe while many fugitive Hurons sought refuge with the Erie. The Erie provided them with comfort, food, and shelter but many chose to remain individuals and operated in small bands with calling themselves the Wendat (Wyandot). In May 1653, an Onondaga captain on a peace visit to Montreal informed the French that they would not attack the French but were planning to wage war against the Ehriehronnons (Lupold 1975:45). According to the nineteenth century romanticized view of the Erie-Iroquois war, it is believed that the war was started due to the Iroquois beating the Erie in an intertribal athletic contest in lacrosse but this seems unlikely. One of the earliest accounts of this event appeared in a July 1845 issue of the Buffalo Commercial Advertiser. Many writers for this publication continued to repeat this story of how the Iroquois-Erie War began. Apparently, the Erie seemed to have won the friendship of the Seneca, and invited them to a game of lacrosse. They challenged the Seneca for 3 years before a challenge was finally accepted where 100 men from each nation competed (Lupold 1974:46). This athletic competition took place near Buffalo, New York and belts of wampum, jewels, bands, moccasins and other goods were wagered on the outcome. The Erie lost in every aspect of this competition. Not only did they lose lacrosse, they also lost a footrace and wrestling match. Frustrated, the Erie chief ordered those who lost be put to death. With this outcome, the Erie decided to plot revenge. They planned to attack the Seneca who resided near Seneca Lake in Geneva, New York, but a Seneca woman who lived with the Erie betrayed them to the Seneca. The Five Nations gathered a huge army of 5,000 warriors and met the Erie in battle near the Genesee River where the Erie realized they were facing the entire Iroquois Confederacy. The Iroquois proved invincible, and placed 1,000 reserve warriors to the rear of the Erie and ambushed them forcing the Erie westward. It is also said that the Erie had allies come as far west as the Mississippi and fought the Seneca in Buffalo but this notion is unknown. It seems ideal that the cause of war was that the Erie were overly aggressive and that they had accepted the Five Nations sworn enemies into their tribe (Huron and other northern neighbors). These factors, along with the Iroquois thirst for fur lands and hunting grounds is enough of an explanation for a cause for war. By 1640, the Iroquois exhausted their fur lands and attempted to get other tribes north and south of Lake Erie to trade with them. The Iroquois wanted to be the center of fur trade. By 1653, the Erie sent a peace delegation of 30 ambassadors to the Seneca capital of Sonotouan to renew a peace treaty. During the course of these negotiations, a Seneca man was killed in the argument with an Erie representative. To retaliate, the Seneca killed all Erie ambassadors except five who escaped (Hunt 1912:101). This incident sparked a full-scale war between the Iroquois and Erie. The Erie responded to this by raiding a Seneca town and ambushing an Iroquois war party and killed 80. In 1654, they captured an Onondaga chief and put him to death by burning him (Hodge 1907). The Iroquois then invaded the Erie territory and headed for the Erie towns that were inland. They carried their canoes and may have ascended Cattaraugus Creek in New York or gone to the headwaters of the Allegheny River. The Jesuit Relations does not say the exact number of men, but the Iroquois Army was anywhere from 700-1,000 warriors, with the Erie being close to 2,000. Most of the Erie retreated to their principal town of Rique (mentioned above) near the present City of Erie, Pennsylvania and as far west as Conneaut Creek. The Iroquois surrounded the Erie in the summer of 1654. The Iroquois attacked the palisaded fort, hurling firebrands that set the town on fire. They suffered many casualties as they were met with a shower of poisoned arrows that the Erie could discharge rapidly. These poison arrows are usually attributed to the Erie in written accounts, but there is no evidence suggesting they actually existed and they are probably just a myth. The Erie had very few firearms, while the Iroquois had an abundance. To counter-act the arrows, the Iroquois sieged the palisaded fort by using their barked canoes as shields over their heads as they advanced. When near the palisade, they placed the canoes upright and scaled the walls by using their canoe as a ladder. The Onondagas massacred the Erie and “wrought such carnage among the women and children that blood was knee deep in certain places” (Hodge 1907:431; Thwaites 1959:181). After the inhabitants of Rique were killed, captured or dispersed the Iroquois rested and remained there for up to 2 months. The Erie tribal unity was shattered, but conquering the Erie was not easy. Hyde (1962) said it required almost 4 years to finish them off. As late as 1657, Onondaga warriors were still bringing captives from Erie County, New York and Montreal, Canada. Of the approximately 15,000 Erie, only 600 surrendered. William D. Ellis (1966) believed that the last stand of the Erie and Iroquois took place on an open plain below Copley Swamp near Akron, Ohio. Ellis believed that the Iroquois used Conneaut as a point for their invasion and that the Erie’s last stand was probably somewhere in the Cuyahoga Valley. This was William D. Ellis’s opinion and to the author’s knowledge the Erie probably have not gone this far south.  CHAPTER 5: IROQUOIAN EARTHWORKS AND PALISADES Earthworks and Palisades for Defense Earthworks are found throughout the eastern United States, but there is a particular concentration in the Southeastern Lake Erie Basin in western New York State including the Niagara Frontier and southern Ontario (Ramsden 1990; Squier 1851; Sullivan et al. 1995). Most of the earthworks in this vicinity are found near streams and elevated swamps, appear to be on hilltops, and usually are either oval, crescent, and even rectangular shaped (Engelbrecht 2009; Mainfort and Sullivan 1998; Squier 1851; Sullivan et al. 1996; Sullivan et al. 1995; Thomas 1889). It is generally assumed that the soils for construction were obtained locally and the earthwork formed by mounding basket loads of dirt. Earthworks often vary in size, and areas within the embankments are usually less than 1.5 ha (3.9 acres) but some as large as 6.39 ha (15.8 acres) have been documented (Squier 1851). In addition to area, the heights of them also varied but tend to not exceed 1.5 meters. These earthworks often follow the natural contours of the ground across various types of undulating terrain (Sullivan et al. 1995). One earthwork site known as the Oakfield Site near Oakfield in Genesee County, Ontario (White 1961:52-53) had an earth embankment surrounding the area of the site and was still clearly visible in the 1960s. At the time, it was impossible to determine the full extent of the ring but the area enclosed by the earthwork was still 1.17 ha (2.9 acres). Estimates from Squier’s (1851) map and White’s (1961:53) observations of the total area indicates that at one time the area enclosed by the earthwork ranged from 2.1 to 3 ha (5.2 acres to 7.6 acres), but probably more toward 2.1 ha (5.2 acres). Squier and Houghton both mentioned the presence of a palisade, but presented no evidence (Squier 1851:65; Houghton 1909:364). The topsoil at this site averaged about 10 cm (4 in) in thickness and in many places had eroded off completely, exposing the clay subsoil. The ditches outside the embankment from where the soil could have been removed for the banks had become partially filled with soil, some of which had eroded from the bank, and the remainder was dumped refuse. Refuse was also found in pits on the site (White 1961:53). Today, the earthworks are often plowed over, destroyed, or at risk of being removed due to land development. Earthworks can also naturally degrade over time due to post-depositional natural and cultural processes. This event typically leads to widening and flattening of the bank, and ditch infilling due to simple downslope transport of fine grained materials that make up the earthwork itself (O’Neal et al. 2005). This landscape smoothing process can lead to a change in the appearance of an earthwork over time, decontextualize artifacts and sediments, and complicate reconstructions of time and labor in construction (O’Neal et al. 2005). In the 19th century, it was originally proposed that these earthworks were used as defensive structures or fortifications (Engelbrecht 2009; Parker 1907; White 1961). White (1961:54) for example, took steps toward model building in the Niagara Frontier region based on the assumption that the earthworks were defensive in function. She proposed a correlation between the enclosures and flat defenseless terrain, with an absence of these structures in hilly, defendable situations. Stothers (1977) believed they were used for defense as they were often constructed along cultural boundaries, while Stothers and Graves (1983:119-120) expanded on this notion and suggested that the enclosures form lines of demarcation between opposing cultural groups. They also make distinctions between “earthwork enclosures” and “earthwork structures” (linear embankments), but they associate both types of features with defense. Iroquois oral tradition states that fighting was common. According to Parker (1916:17), “Feuds with other nations, feuds with brother nations, feuds of sister towns and feuds of families and of clans made every warrior a stealthy man who liked to kill”. Fighting involved social and political units of varying scale. Therefore, it is not surprising that an Iroquois concern with defense can be inferred archaeologically at both the regional and village level. The formation of Iroquois nations is reflected in the location of contemporaneous villages near one another (Engelbrecht 2003:112-126). It is assumed that the members of these villages came to one another’s defense in the event of an attack. The development of large palisaded hilltop communities typical of the sixteenth-century Iroquois is commonly interpreted as a defensive response to widespread warfare (Engelbrecht 2009:179). Thus, it is likely that a concern with defense can also be seen on a smaller spatial scale. A feature such as an earthwork can be considered to have a defensive property if it serves to restrict physical access (Engelbrecht 2009:179). The location of many Iroquois sites on a hilltop or point of land with a steep bank on one or more sides provides well-known examples. A palisade typically protected accessible routes of entry, channeling movement through a narrow entryway. The palisade was often the first and most important line of defense, but enemies were often still able to gain entry into a village (Engelbrecht 2009:179). When Europeans first encountered Iroquois palisaded villages, they often referred to them as “castles”. Reason for this was first; a village on a hill surrounded by a 6-9 m (20-30 ft) high palisade would have appeared imposing even if it were not made of stone in the same way as a walled European castle. For example, Peter Ramsden (1990) has suggested that palisade fortifications in Southern Ontario were not used as a physical defense mechanism, but rather as a symbol for defense. He suggests that the palisades would be an intimidating demarcating line between what was happening inside the fortification and outside of it, and thereby deter oncoming intruders. Engelbrecht would agree with this statement because he makes the connection that both European castles and the Iroquois palisaded villages were built for defense and restricted access to the interior. However, unlike Engelbrecht, Ramsden believed they were not primarily functioning as a defense mechanism because wooden palisades are not very strong and could easily be penetrated by an oncoming force. The third comparison Engelbrecht (2009) states in his work is that both European castles and Iroquois palisaded villages housed people. With these 3 ideas in mind, the term castle used by the Europeans was completely reasonable. Iroquois palisades evolved over time, becoming increasingly more effective as defensive structures (Engelbrecht 2009). However, the diameter of palisade post molds can vary within a site, and over time, these slightly increased (Prezzano 1992:242; Ritchie and Funk 1973:363). The heights of the palisades also varied greatly, even within the same community, but it is believed that they were somewhere near 4 to 10 meters in height (Prezzano 1992). Samuel De Champlain observed a few palisades and specifically one around an Iroquois village he attacked in 1615 and by stating it was around 9.1 m (30 ft) high. He also mentioned that there was a triple palisade around a Huron village with a height of around 10.7 m (35 ft) high (Grant 1959:283, 292 [1907]) In some cases, a ditch was dug in front of the palisade, with the dirt piled at the base, creating what remains today as an earth ring (Engelbrecht 2009). After many years of these palisades decomposing, the earth remains the same height and this is commonly seen as being the earthworks that are found on sites today. Gabriel Sagard (1939:91-95 [1632]) observed sizeable trees placed at the base of a palisade to strengthen it with interwoven branches and saplings between the upright posts to create a wall 2.4 or 2.7 m (8 or 9 ft) high. Palisades may also have served secondary functions as snow fences or windbreaks or served various symbolic functions, but their sturdiness and the obvious effort involved in their construction suggest that defense was their primary function. Ramsden (1990) has suggested that larger palisade fences were typically constructed around times of increase global cooling and that the need for “windbreaks” or snow fences might have necessitated their construction. Since palisades and earth rings are commonly found near the lake where there are typically higher winds and heavier snowfalls, his hypothesis has credibility. Palisades with more than three rows of posts can sometimes be found during excavation. However, there are no ethnographic descriptions of this type of palisade, and a palisade of four or five rows of posts would have probably been unnecessary (Engelbrecht 2009). Gary Warrick (1988) believed that village occupation duration might be assumed from the number of post molds observed along a longhouse wall; many of these suggest decay of the original posts and repairs using new posts (Jordan 2007:245; Warrick 1988). The presence of more than three rows of palisade posts suggests rebuilding. This is an entirely different situation however, from that of multiple palisade rows resulting from village expansion or contraction. In a palisade, it was typical to have only one or two gateways into the village (Figure 15). These were often passageways that were designed by overlapping palisade lines that required people to walk in single file. These could be easily closed off. An entryway similar to this has been described for the Huron, and examples have been recovered archaeologically from a number of sites. For example, Van den Bogart observed a relatively wide main entryway of 1.1 ft (3.5 m) into an Oneida village, but a second entryway was only 0.6 m (1.97 ft) across (Gehring and Starna 1988:12 as cited in Engelbrecht 2009). While palisades serve as the first line of defense for a village, they were not always successful. Even with a sturdy palisade, there was always the possibility that the enemy might gain entry into an Iroquois village and surprise attack. In the Jesuit Relations (Thwaites 1896-1901:29:253) of 1646, there is accounts of two Huron sentries who fell asleep and were killed by Iroquois warriors. Abler (1970:28) discusses that there are numerous situations in the Jesuit Relations and Allied Documents of enemy warriors gaining entry into the village. For the Huron, a warrior would gain great prestige if he successfully entered into an enemy village at night and captured or killed someone (Engelbrecht 2009). “While the Iroquoians seem to have gone through much trouble to build these palisades, to the modern reader it seems almost useless because of their reluctedness to use sentries at night” (Abler 1970:28). In situations where an enemy was within the village, secondary interior defense and village layout would have become critical. Often the restricted area of many hilltop village locations would have encouraged community nucleation (Engelbrecht 2009). It is also possible that the placement of houses within a village was motivated by a need for defense. Sagard (1939:92 [1632] in his work, mentions that an open space was often left between the palisade and Huron longhouse to facilitate defense (Figure 15). It has been argued that in the Draper Site in Ontario, houses were placed to create defensible corridors and reduce access to the center of the village in the event a palisade was breached (Engelbrecht 2003:92, 2009:182; Williamson 1978). Figure SEQ Figure \* ARABIC 15: Examples of Iroquois Palisaded Villages (Google images 2014). The assumption that earthworks were all defensive in function is an integral part of several models of regional cultural dynamics that propose a chain reaction, throughout late prehistory, of increased population and competition for prime agricultural soils, brought about as a result of increased sedentism and an increase in horticulture (Brose 1976a; 1978b: 580; Johnson et al. 1979; Widmer and Webster 1981:71; White 1963a). There are a limited number of studies that have offered definitive interpretations regarding earthwork site functions. According to Sullivan and others (1995) the general acceptance of the idea that all earthen embankments form defensive fortifications associated with palisades persisted even though the available evidence does not support a consistent association of the earthworks with posts. Jones and Jones (1980:69) compiled information about eighteen earthen embankments in central and western New York that have been investigated for palisades. Ten sites yielded evidence of post molds or intact poles; while the remaining eight had no evidence, seven of these were located in western New York. The Ripley Site was one of these and it mentions that there was evidence for an earthwork, no ditch, and no evidence for a palisade. There is very little information about the features within and around the embankments (Sullivan et al. 1995:118). There is virtually no information about the spatial arrangement of features such as pits, caches or post molds and no analyses besides a simple description (Sullivan et al. 1995). Squier (1851:12) and Cheyney (1859) both report large pits or caches in the enclosures, and Squier (1851:13) states that carbonized corn was found in many of these pits. Parker (1907) reported evidence of clay pipes, faunal remains, and some plant materials found within pits or caches within the Ripley Site. This may suggest they were used for the storage of ceremonial materials, but no analysis has been done to infer this conclusion. William Green (1997) and Sullivan conducted an in depth analysis of pit features at the Ripley Site and only determined that Ripley is unlike others nearby and that Ripley might be special purpose because of the high percentage of little-used tools and “ceremonial items”. Sullivan has noted the presence of tobacco phytoliths on the site that may also point, at least in part, to a pattern of ceremonial use as tobacco was often an important item in Iroquois religion. However, this is not substantial evidence to conclude that the earthworks function as ceremonial centers. Descriptions of earthworks often mention burials and sometimes ossuaries are reported near enclosures (Schock 1976:94), but the mortuary practices associated with these sites have not been a specific source of inquiry. Squier (1851:13) notes that the sites of “ancient lodges or cabins,” marked by deposits of cultural debris, could be traced within many enclosures. Marian White (1963) discovered one longhouse-type structure measuring 20 m by 6 m (65 ft by 19 ft) which contained a corn-filled pit within a small earthwork at the Henry Long site in Erie, County, New York. She assumed a domestic function for this structure, but the recovered artifacts and specimens have not been analyzed. It was common practice for the Iroquois to store foods in pits or caches over the winter months, and this may represent such as case. Earthworks and Ritual While earthworks were often associated with the remains of a palisaded village, it has also been proposed that they hold ritual and/or ceremonial importance. Cheyney (1859) and Guthe (1958) both agree that while the embankments were possibly used for fortifications, some were probably used for rituals and not for habitation. Cheyney suggests that an earthwork was used for fortification only if it was in an ideal location in need of defense, such as lake plains or areas of low elevation (Cheyney 1859). He also believed that many of the embankments he interpreted as being fortifications lay on a cultural line that would have divided the Iroquois and hostile “Appalachian groups” (Cheyney 1859). In 1958, almost 100 years after Cheyney, Guthe (1958) revived this idea while working in the Chautauqua/Allegheny area. His work preserved the dichotomy that earthworks served either defense or ritual needs. According to Guthe (1958), after excavating numerous sites, the number of artifacts found within the earthworks did not seem to correlate to the amount of hours of construction the earthworks must have taken (Guthe 1958). Guthe was attempting to point out that if sites took such a long time to construct, then they would have been occupied more often. Theoretically, the longer the site was occupied the more items one may find there. Although Guthe’s conclusion is poor, he believed that those sites with fewer artifacts were used for ceremonies. Erection of monuments, such as ceremonial enclosures, also has been linked to competition for critical resources between small-scale segmented societies, especially when mortuary ceremonialism was involved (Chapman 1981; Renfrew 1977:200-206). Maintenance of formal disposal areas for the dead can be linked with development in relatively uncentralized societies of more complex kinship structures and groups, including territorially based descent groups, as a result to control critical, and scarce resources (Chapman 1981). The symbolic presence of mortuary facilities legitimizes the claim to control and use of the important resources. The associated mortuary ritual also can serve as a mechanism for social integration and maintenance of group identity (Sullivan et al. 1995:119). Further, there seems to be a link between level of investment in mortuary facilities and increase in competition (Chapman 1981). An example of this function of mortuary ritual from the Great Lakes area is the Huron Great Feast of the Dead. Ceremonies such as this involved individuals from a number of villages and served to symbolize and cement tribal unity through periodic placement of the dead in formal ossuaries (Heidenreich 1978:374-375, Trigger 1990). Both the defense and ritual interpretations of the earthworks are inconsistent with general models of “tribal” societies like those believed to have inhabited the Great Lakes region prior to European contact. The argument can be made that defensive structures, monuments, and special disposal areas for the dead all can be linked to pressures brought about by a sedentary lifestyle and increased plant cultivation. These types of site/features possibly could co-occur within a single cultural setting (Sullivan et al. 1995). Sullivan (1995:119) states that points such as this not only suggests archaeologists may have made poor interpretations of sites containing earthworks, but that these interpretations may have influenced reconstructions of cultural dynamics. In southwestern New York, interpretation of all earthworks as defensive in nature may have increased archaeological perceptions of the level of pre-contact intertribal conflict, presumably related to critical agricultural resource. The fact that the function of earthwork sites is loosely understood makes gaining understandings of both the defensive behavior and the density of population significant problems. However, while these notions do raise concerns, defense and support for palisades generally seems to be the best interpretation for these structures because of their size, consistent location on hilltops, and location on cultural boundaries.  CHAPTER 6: SOIL MORPHOLOGY Soil and Soil Texture The term soil has many definitions and varies slightly between disciplines. Soil is unconsolidated material at or near the Earth’s surface that is made of organic and inorganic constituents (Schaetzl and Anderson 2005:9). To geologists and geomorphologists, soil is a natural entity that is caused by weathering at the immediate surface of the earth of sediment and rock; it acts as a medium for plant growth, and is affected by various environmental factors such as climate, fauna, flora, topography, and location. Soil develops from “parent material”, or the weathered bedrock and debris that have fallen onto the landscape (Holliday 2004:3). The make-up of soil consists of varying amounts of organic matter, sand, silt, and clay; all of which are specifically designated names based on particle grain size. Soil morphologists use a ternary diagram to identify the soil texture type based on the various percentages of sand, silt, and clay (Figure 16). Soil texture particle size divisions are based on the USDA divisions for the fine earth fraction (clay through sand), and on the USGS divisions for larger particle sizes. Another scale to base this on is known as the Wentworth Scale (Table 1). Sandy texture classes can be further described by taking into account the predominant sand particles. You can name the sand within a texture size class e.g. (very fine, coarse sandy clay, etc.), or one can name the sand size after a texture class in parenthesis e.g. sandy clay loam (fine to medium sand). Determining soil texture in the field can be a daunting and confusing task for beginners, but texture can also be determined in a laboratory setting. Exact percentages of sand, silt, and clay can be determined in the laboratory by various methods and this project utilizes a Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer for this purpose (more mentioned below). Figure SEQ Figure \* ARABIC 16: Ternary diagram for soil texture. Table SEQ Table \* ARABIC 1: Wentworth Scale for Particle Grain Size Textures (Google Images 2015). Soils are laterally extensive across the landscape and form across various landforms from a variety of parent materials and they vary in a predictable manner because of changes in erosion, deposition, drainage, fauna, vegetation, and age of the landscape (Holliday 2004). Soils differ as the microclimate and macroclimate varies; this predictable variability is referred to as the “constancy of relationships” and is unique to soils amongst geomorphic phenomena (Brewer 1972). This important defining characteristic of soil in a buried context allows it to be traced over paleotopography in three dimensions. However, individual layers of sediment would be confined to particular depositional environments (Holliday 2004:3). Soil Formation Hans Jenny in his book Factors of Soil Formation outlined the importance of a technique he called Cl.O.R.P.T (Jenny 1941). This approach equally weighs five factors that are all responsible for soil formation: s = ƒ [Cl.O.R.P.T]) wherein soil (s) is a function (ƒ) of the factors of climate (Cl), organisms (O), relief (R), parent material (P), and time (T). All of these contribute to unique characteristics in the process of the soil formation. Jenny suggested that soil is one element of the larger environmental system, and the best way to understand soil formation is to break it down into multiple components (Vogel 2002). Soil Horizonation Soil formation is the direct result of water, wind, temperature and organisms acting on sediment through time. As precipitation occurs, it filters through the sediment, leaching organic material and fine mineral inorganics with it. Plant roots displace sediment, draw water upward, and leave organic material and sometimes voids in their place as they decompose (Vogel 2002). Various insects and animals burrow in the soil, moving sediment and large particles, leaving tunnels that sometimes fill in with other organic material. These processes, and others, lead to the formation of soil horizons or pedostratigraphy. Master Horizon Designations Soil horizons are divided as follows according to Gregory Vogel (2002) and the National Soil Survey Center (2012): O: Often the top horizon found at a site and made up of leaves, twigs organic debris. Often known as the humic layer. A: This is directly below the O horizon and the soil is often stained with very degraded organic material. The downward movement of water through this horizon moves the finer clay particles and weathered minerals to the layers below. The zone of leached clays, minerals and organic accumulation is the A horizon. Animal burrows are very common in this layer. Ap: An Ap horizon generally occurs on sites of plowing, grazing, or other anthropometric-induced activities that have altered the surface. Regardless of the mineral or organic content, it is termed Ap horizon or plow horizon. The lower boundary of an Ap horizon is often abrupt (Vogel 2002:12). E: These horizons only sometimes develop below an A horizon that was leached of clay and other minerals, but does not contain organics. These are usually found in dryer climates. E horizons generally consist of highly resistant minerals such as quartz, and in return, they appear more of a lighter gray in color. Generally, the boundary between an E horizon and below may be gradual or abrupt (Vogel 2002:12). B (Bt and Bw): B horizons have been altered by weathering on the surface, but mostly contain clays and minerals, and no organics. B horizons are the accumulation of minerals from the A and E horizons. Sometimes clay particles form clay skins in the soil structure, when this occurs with heavy concentration of clay; it is termed a Bt horizon. If the horizon does not contain clay skins, or a strong accumulation, it is a Bw horizon. C: If the soil is deep enough where weathering does not affect it all the way to bedrock, the unweathered broken parent material is referred to as a C horizon. Soil structure is not found in a C horizon. This layer is not affected by weathering, and any stratigraphy present in a C horizon had to have been deposited (sedimentary horizons). Deposition, as mentioned, can occur from glaciers, erosion, gravity, water, wind, etc. Cultural features such as pits, trenches, and post molds may be filled with material from any horizon, but until weathering affects the fill matrix, it is also termed as being part of a C horizon. R: Hard bedrock may be found under any of the horizons described above, and is only known as an R horizon. Soil Color Soil color is often the first characteristic of soil that is observed and this observation is usually before the soil is touched, smelled, or tasted. The color one can examine is actually the clean soil particles or the coatings they have (cutans). Soil color is observed using the Munsell® Soil Color System. This ensures that soil scientists objectively quantify color by comparing the samples to soil color chips. The Munsell System takes advantage of the fact that color is composed of hue, value, and chroma. Hue refers to the chromatic composition of light that emanates from an object. Hue is most commonly represented by the abbreviations R for red, Y for yellow, and YR for yellow-red. The hue ranges from 2.5-10 going from a red to yellow th hues consist of 10R, 2.5YR, 5YR, 7.5YR, 10YR, 2.5Y, 5Y, etc. The factor that most influences the hue is the mineralogy; red soils are hematite-rich, brown are goethite-rich and white soils often contain a high amount of salts or carbonates (Vogel 2002). Value describes the darkness or lightness of the color. Some refer to value as the intensity of the color. This ranges form 0-10 and is displayed along the vertical axis on a Munsell page. As the value changes, the amount of white or black is added to the color: 0 is black and 10 is white. Low color values are darker colors, and high values are very light similar to if it was as exposed to sunlight. Low values imply that the soil is higher in organic carbon and/or wetter. Soil color should always be noted as moist unless a dry color is needed. Chroma refers to the purity, strength or grayness of the color; it is ranked on a scale of 0-10, in increments of two. At a Chroma of 0, all hues converge to a single scale of neutral grays, referred to as N0. Chroma is changed as more gray is introduced into the color. Most soil particles have some degree of soil “paint” on them. The color of a soil particle, soil horizon, or sample is a function of the type and degree of its coatings. Darker colors imply matter that is more organic, while red come from iron bearing minerals. For this fact, color provides important information about particle and ped coatings in soil that in part, contains genetic clues. For example, only in E horizons are soil materials so clean that the color of the primary minerals of the soil come through. Because most soils are dominated by quartz, many E horizons are exposed as white. Darker colors, as mentioned, contain more organics, and red-brown colors are associated with well-drained, upland soils in which the conditions are oxidizing. While many soils are often a variety of colors, some are uniform. Variation in soil-color is usually due to types of different coatings or variations of the coating. “Spots” of one color set in a matrix of another such as mottles, might imply some sort of accumulation process, such as when sodium salts precipitate into a dark soil. For mottled soils, the colors of the matrix and the mottles should be described, and the abundance of the latter noted (Schaetzl and Anderson 2005:17). Soil Structure Soil structure refers to the arrangement of the primary soil particles (sand, silt, clay) into natural aggregates called peds. Aggregates formed by tillage or by other human-induced practices are called clods (Schaetzl and Anderson 2005:18). Soil horizons have their own unique structures, depending on the amount of biotic activity, organic matter, freeze-thaw, wet dry cycles, etc. Soils that lack the pedologic structure are termed massive (fine-textured materials which break along no preferred planes) or single grained (in sands where each grain behave independently). A strong distinguishing criterion of soil aggregates is that they have internal cohesion that prevents being broken up by external forces that continually occur in soil. Humus is often a common “glue” that holds together these peds but often it is the saliva, urine, and other secretions by soil fauna. Many peds are simply fecal pellets held together by body fluids and prior compaction. In clayey soil, they may be attracted by the chemical bonds that clays naturally have. When discussing soil structure, the two most commonly used adjectives to describe it include ped shape and size. The smallest peds are in the near surface A horizons and here granular peds are usually about 1-5 mm in diameter. This granular structure is mostly due to soil biota. These peds tend to be porous, and because they do not fit together tightly, horizons of granular structure are highly permeable and fall apart on exposure. There is a variety of soil structure types. As mentioned, granular and massive are usually the most common in upper horizons. Platy structure consists of thin (< 4 mm thick) platy-like peds that are parallel to the soil surface and are commonly found in E horizons. The mechanism of the platy structure formation is not well understood, but may have to do with freeze-thaw cycles. Lower in the soil profile the peds become larger and more block-like. Angular and subangular blocky structure is very common in B horizons. In an angular blocky structure, the edges facets between them are distinct, while the more common subangular blocky structure has more rounded edges. These peds are often held together by a coating called a cutan which is a material that has been translocated into this horizon by illuviation. This material moves into the B horizon through suspension or solution in the large pores between peds (Schaetzel and Anderson 2005:20). At greater depths, the forces that act to break up the soil matrix into distinct peds not only become less frequent, but they tend to be more vertically oriented. Blocks that are taller than they are wide called prisms or prismatic structure form in these conditions. This is typically found in the lower B horizon and C horizon. In areas where soils have high amounts of exchangeable sodium, the tops of the prisms sometimes appear to be white from an abundance of clean quartz grains; this structure is called columnar. The tops of the prisms become rounded as soil water is forced to run between them because permeability between the peds is slow. Therefore, the edges are worn away and take on a rounded shape, but the bases may still be flat and angular (Schaetzl and Anderson 2005:19). Buried Horizons and Paleosols Soils that formed in the past and that are no longer actively forming today are known as paleosols (Waters 1992:57). Soils cease to form when they are buried to a depth that removes them from active soil forming processes. They can change slightly however, if the biological or climatic regime under which they were formed changes (Waters 1992:57). A buried paleosol is a soil that was buried by sediment and no longer subject to the environmental processes that created it. Buried horizons are very important for archaeological and geoarchaeological work, as these are indicative of past climatic episodes and often were a medium of past anthropogenic modifications (Schaetzl and Anderson 2005). Soil horizons develop slowly, gradually, and over time and a new soil will begin to form on younger sediments that bury the soil if the surface again becomes stable. If the younger sediments that bury the soil are thicker than the zone of active pedogenesis, the buried soil will not undergo further pedogenic alteration (Waters 1992:58). These buried horizons represent land surfaces that were stable and forming over long periods. Some soils take hundreds if not thousands of years to form strong horizons (Waters 1992:57). A concept known as genetic stratigraphy can also be useful for archaeologists, soil scientists and geomorphologists to get a chronostratigraphic framework for the soil. These are divided into Allogenic and Autogenic events. Allogenic events are often long-term weathering events that can be seen in relict paleosols (mentioned below) and represent widespread regional climatic episodes. Allogenic events can help the geoarchaeologists to understand the paleoclimate. Sediment packages without developed soils represent times of rapid deposition also known as autogenic events (Vento et al. 2008; Vogel 2002). Autogenic events are locally developed and often the result of a circumscribed event that is constrained geographically and/or environmentally. These events may correspond to local tectonism down to delta switching, storm scour, or channel avulsion (Vento et al. 2008:2). Colors can also be a clear indicator one has found a buried horizon because the color is often the first to change. However, the appearance of fine or very fine biopores and the presence of a granular structure are much more reliable (Vogel 2002). Buried soils are best recognized in a stratigraphic sequence because they retain their soil texture, structure and diagnostic horizons from the previous climate when they were deposited (Waters 1992:57). Often, the upper parts of buried soils are sometimes missing because they were eroded before they were buried. Usually only a thin layer of soil deposits over a buried soil and this consequently results in the buried soil being still in the zone of accumulation and active pedogenesis. With this phenomenon, the buried soil can undergo post burial changes as the overlying soil deposits “weld” into the upper portion of the buried soil horizon. This process results in a “welded” soil profile (Water 1992:58). If the thin layer of overburden erodes on this soil profile, the buried soil horizon can once again be re-exposed. The re-exposed soil horizon is known as an exhumed Paleosols. On occasion paleosols occur at the Earth’s surface. These are called relict paleosols, and are often found on old landforms that have not been buried or eroded. Relict paleosols are important for the geoarchaeologists because they provide a means of understanding past climatic episodes. Because relict soils remain sub-aerially exposed after pedological changes, much later features affect them. Younger soils impose on top of the relict soil and pedologically change it, similar to the buried soils. This is known as a composite soil (Waters 1992:59). A relict soil develops during the strongest period of soil development to which it was subjected. For instance, if the original soil on a surface were weakly developed, it would be altered and masked by subsequent stronger pedogenesis. Paleosols are important for a variety of reasons and applicable in a number of scientific fields. First, because soils form at the surface, they essentially mirror the topography of the landscape. While some landforms never fully become buried and have relict paleosols, some areas have consistent soil formation and remain stable for long periods but are then later buried by younger sediments. In this case, buried paleosols can help to reconstruct past topography during these episodes of stability. Second, paleosols are important because they indicate episodes of landscape stability that may help to reconstruct paleoenvironment. Archaeologists use paleosols as references to better understand the paleoenvironment and from this, they can better understand how earlier groups of people utilized the landscape. This leads to the third reason; In addition, paleosols are important because they can be used as markers for correlating and reconstructing stratigraphic sequences. For example, by using flood plain stratigraphy on the Brazos River in Texas, the paleosol was an important stratigraphic marker that was used to differentiate late Holocene alluvium from early and middle Holocene sedimentation. Fourth, also important to scientists, the physical and chemical characteristics of a paleosol can be used to reconstruct the climate in which they formed (Holliday 1989). Holliday (1985) used variations in soil profile characteristics from five paleosols on the Lubbock Lake site in Texas to reconstruct temperature and precipitation fluctuations in the Holocene. Finally, vegetation affects soil profile development. By analyzing the various morphologies of paleosols, it can help to reconstruct what types of vegetation were present on them during the period of their formation (Holliday 1990). Holliday (1987) for example, was able to determine that the vegetation of the Southern High Plains was not covered with a forest but rather open grassland. In all, paleosols preserve information about soil formation episodes and help to illustrate periods of landscape stability. Sedimentary Strata Horizontal layers created as sediments are deposited (ex. the layering of gravel or sand in a stream) are not soil horizons, but rather, sedimentary strata. These layers are created as soon as the sediment is deposited and do not change except when altered by surface weathering. When stratification occurs, they tend to follow the Principles of Geology and Nicolas Steno’s Laws (E.g. the Law of Original Horizontality, Law of Superposition, Law of Cross Cutting Relationships, and the Law of Lateral Continuity). Stratigraphic levels can be altered in numerous ways (faulting, folding, burrows, roots, etc.), but do not generally form because of weathering. Stratigraphic units form through erosion, when layers are deposited from either wind, water, gravity, volcanoes, etc. Soil Sampling and Stratigraphic Prospecting Strategies There are a variety of techniques to examine and record soil stratigraphy (French 2003; Goldberg and Macphail 2006:316-334; Holliday 2004:31). Typically, on archaeological sites, the archaeologist records soil horizons and stratigraphic profiles as part of archaeological documentation. Soil horizon designations are typically the first component of a soil description done in the field and often conducted during a Phase I or archaeological reconnaissance, however it is often more beneficial to interpret these horizons in the final stage of analysis as there is more data that can help reinforce inferences made. Having a geoarchaeologist or an archaeologist with a background in geology often proves very useful to understanding site formation and can be important to properly document the site while work is in progress. Besides analyzing profiles during excavation, backhoes are often employed on or near the site as an expedient measure to prospect the natural stratigraphy (Holliday 2004:30). Backhoes however are extremely invasive and can damage culturally sensitive material. A less invasive means of stratigraphic prospecting is through examination of cores extracted from the ground. These cores can be removed by a variety of techniques. The first includes using powered truck or trailer-mounted soil-coring machines that are generally 9 cm in diameter and can penetrate roughly 120 cm deep (Holliday 2004). This allows rapid stratigraphic analysis of stratigraphy in large-scale areas. Another means of coring involves hand-operated augers or soil probes. The most common is called a bucket auger, which can recover samples in small intervals These are typically cylindrical, anywhere from 10-25 cm in length, with a diameter around 7-10 cm. (Holliday 2004; Goldberg and Macphail 2006). Extension rods can be added to the buckets to reach roughly 3 meters deep depending on lithostratigraphy and how deep the bedrock is located. Because it is hand-operated, the bucket-auger is extremely useful in locations where one cannot have access with powered coring instruments or backhoes but one major drawback is that they have the potential to mix the samples. The process of augering can obscure sedimentary and soil structures, depositional contacts, and soil boundaries. Bucket augers provide general information regarding lithostratigraphy or pedostratigraphy, but for micro stratigraphy and detailed stratigraphic mapping, it is not sufficient. Between powered truck augers and hand augers exist portable power drills that are capable of taking cores. The major setback to these is that fuel must be carried, but the costs are less than the truck and trailer powered corers. These are mostly used for coring and analyzing frozen sites, as using a hand-powered auger would be difficult to penetrate frozen soils (Holliday 2004). Although mostly used for geologic testing, coring and augering devices have been employed for an analysis of deeply buried archaeological sites and to recover micro-debitage (McManamon1984; Michlovic et al. 1988; Stafford 1995; Stein 1986). These techniques raise a variety of questions and issues regarding sampling representativeness, but they have been useful for directly finding buried occupation zones (Holliday 2004:33). For example, Michlovic and others (1988) collected samples of a buried A horizon to better understand the site formation. In these soil cores however, they pulled up in situ micro-lithic artifacts that suggested a buried occupation horizon. Although they found these micro-lithic artifacts by chance, they did prove to be useful to understand the site age. To get the best representation of site stratigraphy, a combination of trenching, coring, and augering techniques probably provides the most valuable information for large areas (Bartley 2006). The method to apply varies depending upon the soil type, stratigraphy, soil conditions, thickness of deposits, the size of the area, budget, location, and time (Holliday 2004).  CHAPTER 7: THE RIPLEY SITE Stratigraphic Context and Depositional Environment The Ripley Site (NYSM 2490), also known as Dewey’s Knoll, lies on the Lake Erie bluffs in Chautauqua County, New York, near the town of Ripley. The bedrock here is made of Upper Devonian shale, more specifically the Westfield shale and the Shumla siltstone, which both are members of the Canadaway Formation of the Chautauquan Series (Tesmer 1963; 1966). The Northeast shale member of this formation underlies most of the Erie Lowland province and the Portage Escarpment in this part of the county (Tesmer 1963). Dexterville siltstone and the Ellicott shale of the Chadakoin Formation of the Chautauquan Series underlie the plateau south of the escarpment and along major streams. As one proceeds southward from the escarpment, unconsolidated surficial deposits of Quaternary age stones are present (Tesmer 1963). The Ripley Site lies on these unconsolidated glacial till deposits. Glacial processes are responsible for many of the landforms in this area of New York. In Chautauqua County, these include the Kent, Clymer, Findley, Lavery, Defiance, and Lake Escarpment moraines (Miller 1973; Sullivan et al. 1996). The latter three are those closest to the Ripley Site. These moraines are considered Late Cary in age. Glacial troughs are also notable here. The three closest troughs to the Ripley Site were scoured during the Pleistocene (Sullivan et al. 1996). Today, various creeks are present in the southeast trending valleys that are part of these glacial troughs. South of the Ripley Site, the upland topography trends toward the south-southeast but is crossed by west-southwest flowing creeks, which are confined by the recessional moraines in this area. French and Brokenstraw Creeks drain this area between the Findley and Clymer moraines (Mueller 1963:13). Climate Four controls have an influence of the climate of the Ripley area. These include latitude, air masses, continentally, and the modifying effects from Lake Erie (Beckerman 1981:2-33). Lake effects are those climatic features that primarily come about because of modification imposed by the Great Lakes. These raise the January average temperature by 5 °F and along the shoreline where Ripley is situated the effects are visible. In the summer, the lakes help lower the average July temperature by 3 °F. These effects have a direct influence on the number of frost-free days that is important to agriculture sustainability in the region (Beckerman 1981:2-33). The warmer temperature on the lower air layers causes instability in the atmosphere. The annual precipitation due to this is 34.5 in/ yr. and severe snow conditions in the fall and winter. As most of the winds come from the north and west, the area of Ripley tends to get the moisture that is picked up over the lake and added to the atmosphere. This phenomenon increases the mean number of days with measurable precipitation, but decreases daily rainfall. With this in mind, the summer has less precipitation while the fall/winter has considerably more. Ripley Site Soil This region generally has limey soils, which are formed over undulating terrain of glacial sediment and till parent material (gravelly loamy glaciofluvial deposit over sand) (University of California, Davis 2013). The soil at the Ripley Site and in the Lake Erie Lowland is primarily podzolic forest soils that are relatively young in age and ordered as Inceptisols, or Typic Dystrochrepts (Braun 1950:24-25; De Laubenfels 1966; UC Davis Soil Resource Web 2013). Directly under the site lies the Chenango gravelly loam, which is primarily a gravelly glaciofluvial deposit that was mainly derived from sandstone, shale, and siltstone. This location is on a 0-3 percent slope and is well drained (UC Davis Soil Resource Web 2011). The natural horizonation for the site is typical for locations along this part of Lake Erie (Figure 17). However, archaeological disturbance of these profiles, as well as prehistoric modifications of A and B horizon relationships are likely to be evident in the stratigraphy of the site. The 2C indicates that the site has a discontinuity and represents a soil that has a wind-blown or eolian “cap”. Figure SEQ Figure \* ARABIC 17: Natural Soil Horizon Column for the Ripley Site (UC Davis Soil Resource Web 2013). Figure SEQ Figure \* ARABIC 18: Percent Organic Matter, sand, and Clay with Depth at the Ripley Site (UC Davis Soil Resource Web 2013). Post-Depositional Processes and Bluff Erosion at Ripley The northern portion of the Ripley Site is marked by a wave-cut cliff that erodes at a high rate, and at times, large sections of the cliff erode at once into the Lake, which leads to concern over site preservation (Sullivan et al. 1996). It has been estimated that at least 15.2 m (50 ft) of shoreline has been lost in the last 50 years (Harding 2014). This rate of erosion is unpredictable and varies dependent upon climatic and topographic variables. The majority of the site is possibly still intact and is currently in a plowed farm field that is actively used for corn, wheat, and sunflower crops. The site is plowed and disked bi-annually, which exposes cultural artifacts regularly (Allen Quinn, personal communication 2013). As noted earlier, shoreline erosion has dramatically altered the extent and configuration of the original beach along Lake Erie’s southwestern shoreline. This has raised the question as to whether additional Late Woodland villages located in similar settings as the Ripley Site were submerged or destroyed. Parker believed that the northern extent of the site had eroded due to erosion of the bluff. He also had noted that there were several sites along the mouths of streams east of Ripley; but to the writer’s knowledge, these have never been located again. With these factors and observations in mind, it is only safe to conclude that there is an un-estimated number of large village sites along the shoreline that have been lost due to erosion (Widmer and Webster 1981:3-127). Lake Erie’s shoreline erodes at a relatively high rate (Cross et al. 2007:19). In 1974, the commonwealth of Pennsylvania deployed a team led by Paul Knuth of Edinboro, Pennsylvania to begin an erosion-monitoring program along the western coast of Pennsylvania from Springfield Township to the Borough of North East, Pennsylvania on the Chautauqua County, New York Border (Knuth and Lindenberg 1995). One hundred and thirty pins were placed at a distance landward of the bluff crest every 1.5 km along the shoreline (Knuth and Lindenberg 1995). Since 1982, around every five years, measurements of the distance from these “control points” to the bluff crests have been made. The closest location near Ripley, New York with shoreline data is the Borough of North East, which has a recession rate of about 0.167 m/yr. (0.55 ft/yr.). Even closer to Ripley lies the location of the North East Marina (8.5 km south of Ripley) which reports a long term average recession rate of roughly 0.3 m/yr. (0.98 ft/yr.) over the course of a 100-year period. However, the short-term recession rate for this location in a similar study by Morang and Melton from 1985-2001 presented an average recession rate of 0.2 m/yr. (0.65 ft/yr.) (Morang and Melton 2001). In another study by Geier and Calkin (1983:31), the area directly near Ripley (Reach 10) has a recession rate of around 0.17 - 0.22 m/yr. (0.56- 0.7 ft/yr.) (Figure 19). Along the New York coastline, Geier and Calkin (1983) reported an average recession rate for the 1938-1974 short-term period was 0.10 m/yr. (0.30 ft/yr.) with a measured range of 0.02-0.22 m/yr. (0.06-0.7 ft/yr.) and a standard deviation of 0.05 m/yr. (0.16 ft/yr.). The 1875-1974 long-term period was 0.15 m/yr. (0.5 ft/yr.) with a range of 0.04-0.46 m/yr. (0.13-1.5 ft/yr.) and a standard deviation of 0.08 m/yr. (0.26 ft/yr.). Figure SEQ Figure \* ARABIC 19: Map of Lake Erie Indicating Areas of Erosion Study. Table Depicts Recession Rates for Map Locations in m/yr. (Geier and Calkin 1983). Wave action is the dominant erosional mechanism at the Ripley Site (Cross et al. 2007:18). During times of low lake levels, the wave action tends to be directed toward the deepest layers of sediments. When these lake level rise, the sediments higher on the bluff become impacted. Erosion rates can vary widely depending on the type of material exposed to the wind and wave action. If wave action is directed at the lower bedrock layers, like the case at the Ripley Site, it is likely that less erosion would occur. Bedrock makes up 63 percent of the bluff material on the New York Lake Erie coastline (Geier and Calkin 1983:14). If the wave action is directed toward soils with more sand and clay, as in the case in the area of North East, Pennsylvania, erosion occurs more rapidly. Just because waves are striking the lower bedrock layers does not mean erosion does not occur. Often these waves cause the bluff to undercut, and large chunks of land catastrophically break off and fall into the lake. This may affect an average erosion rate. The Ripley Site has undoubtedly been affected by erosion, and determining an erosion rate for the site is objective #1 of this thesis (Figures 20-22). Figure SEQ Figure \* ARABIC 20: Photograph of the Ripley Site’s Shoreline Facing West. Figure SEQ Figure \* ARABIC 21: View of the Cliffs at Northeastern Part of the Village Site in 1907 (Parker 1907:477, Figure 1). Figure SEQ Figure \* ARABIC 22: Bird's Eye View of the Ripley Site Facing Southeast. Note: Height of the Cliffs (Bing Maps 2015). History of Investigation This site has served as the focus of archaeological investigations of varying quality since the mid-19th century. The site’s location on the Erie bluffs, its close proximity to the shoreline, plethora of artifacts, and the sandy soil are all factors in its archaeological popularity. Despite the numerous investigations at the Ripley Site by both professional and avocational archaeologists, four reports and only a handful of articles of the investigations have been published. Arthur C. Parker published the first report on Ripley where he investigated mostly the northern portion (Parker 1907). After completing one field season at the site, avocational archaeologist Conklin (1989) wrote his report that focused on a handful of burials. The third report was authored by Sullivan (1996) and colleagues, and included an abundance of useful information regarding previous excavations and the most recent analyses conducted. In 2014, a Master’s Thesis project by Victoria Harding out of Indiana University of Pennsylvania (IUP) conducted a spatial analysis of Parker’s 1907 work and the work of Sullivan, Neusius and Neusius of the Ripley Archaeological Project. This project looked at spatial relationships between excavations and came up with a conclusion that the site was multi-component. Prior to Sullivan’s and Harding’s work, Parker’s 1907 report, entitled Excavations in an Erie Indian Village and Burial site at Ripley, Chautauqua Co., N.Y., was considered the major reference for this site. Parker believed the site was an Erie fortified village and cemetery complex dating to the late 16th or early 17th century. His assertions were based on the presence of a few European trade items found as funerary objects and on ethno-historical Jesuit accounts, and cartographic evidence all of which placed the Erie in this vicinity (Parker 1907). Figure SEQ Figure \* ARABIC 23: Arthur C. Parker's 1907 Hand Drawn Map. Note: Trench Excavations across the Earth Ring (Parker 1907). Prior to Parker’s arrival at the site, the first excavations were in 1904 and were conducted by Mark R. Harrington. Harrington directed his research primarily on the western portion of the site where he excavated pit features and burials with no publication to show for his work (Sullivan et al. 1996). In 1906, Parker began his excavation at Ripley where he focused on what he believed to be the “village section” (Parker 1908; Sullivan et al. 1996; Harding 2014). Parker excavated using a series of trenches and noted many post molds, fire pits, graves and even excavated what he believed to be the earthwork. In 1988, Sullivan reopened excavations at this site under the auspices of the Ripley Archaeological Project (RAP) (Figure 24) (Sullivan et al. 1996). She published two volumes on the site, which outlined the excavations, interpretations and findings. The excavations were collaborative, and undertaken by the New York State Museum (NYSM) and Indiana University of Pennsylvania (IUP). Lynne P. Sullivan (NYSM), Phillip and Sarah Neusius (IUP) served as the project’s co-principal investigators. The major goals of the RAP were to develop refined models of late and proto-historic cultural dynamics in the Chautauqua/Allegheny portion of the Southeastern Lake Erie Basin. Investigations of the Ripley Site were important to the larger goals for the region because of shortcomings of Parker’s nearly century-old work and the general scarcity of modern, in-depth studies of sites dating to these time periods in this area. After all of the excavations, analyses on recently collected artifacts were compared to extant artifacts curated in the NYSM. Because of this comparison, Sullivan and Neusius were led to believe that the activities on the site were related only to mortuary activities and ceremony and dismissed Parker’s interpretation that the site was a residential village site (Sullivan et al. 1996). Radiocarbon analyses of charcoal in post molds on the site provided (Sullivan et al. 1995) two dates [710 ± 110 B.P.; cal A.D. 1219-1331(p =.69) and cal A.D. 1344-1394 (p=.31) (Beta 29940); and 620 ± 110 B.P.; cal A.D. 1284-1416 (p =1.00) calibrated at 1 sigma (Beta 29941) (Stuiver and Reimer)]. These dates raise questions concerning the occupation span of the site (Parker 1907, Sullivan et al. 1996). Figure SEQ Figure \* ARABIC 24: Map Depicting the Approximate Location of the RAP Excavations (Harding 2014, Figure 14). The Ripley Earthwork As mentioned, an earthwork once stood on the site but it was leveled in the mid-19th century due to farming practices on the sites location (Parker 1907:519). When the earthwork was discovered it was crescent shaped, however it was probably completely circular at one point in time prior to coastal erosion. Parker depicts it being located at the northern portion of the site and extended toward the lake (Figures 23 and 25). Figure 25: Ripley Site Map and Features from All Previous Excavations. Note: Earth Ring in Northern Portion of Site (Sullivan, 1996, Figure 4.12). A local resident named George Morse made the first account of this earthwork being present in the late 1820s (Parker 1907). In this firsthand account, Morse described the presence of “pot hunting” at the site and mentions the removal of artifacts, the presence of a possible grave/stone cairn that later fell into the lake and the earth ring. Parker in his 1907 report quotes the account of George Morse (Parker 1907:519): “It [the earth ring] was breast high, and covered with a second growth of whitewood woods. All around the circle, several rods from the edge, was the primeval forest, which was cleared away… To be precise, I remember that the ring was not complete, for the two ends, like the letter “C,” touched the lake bank. Since the earliest days, relics have been carted away. When the stumps were pulled and wherever the grub hoe struck, arrows and “skinning stones” would come to light. Sometimes Indian crockery, in pieces as big as your hand or bigger, would be found… My father planted corn there in 1826, and he plowed and dug it [the earth ring] level. There was a stone mound covered with earth there. My brother dug into it, but did not dig deep enough…. Finally, the bank caved off, ([the bank caves off every spring, a good deal), and a part of the mound fell into the water. Then we looked at it, we saw a skeleton exposed under it. Shortly, the entire mound went over into the lake [McKutchen n.d.].” (Parker 1907:519). This account is crucial to the understanding the lack of visible surface remains of the earthwork, and useful for subsequent investigators in identifying the location of the earth ring. Mark R. Harrington, an avocational archaeologist, briefly excavated the site in 1904. When Parker first arrived on the site in 1907, there were still remnants of the earth ring (Parker 1907:518). He believed that the ring probably enclosed the main portion of the village, and he identified the ring as “having a very hard and compact soil, with the occupied soil being covered with a layer of sand and gravel” (Parker 1907:518). The ring was evidently disturbed and intermixed (probably from plowing) but had few signs of modification by human substances such as ash or charcoal (Parker 1907:518). According to Parker’s profile map, it appears that the original ring sediments overlay the original site occupation in some places, (Figure 26). Based on this evidence, it was suggested by Parker that the earth ring was leveled down and the earth was thrown over the occupied soil by plowing. This physical evidence goes hand in hand with Morse’s firsthand account to Parker (Parker 1907:519). Figure SEQ Figure \* ARABIC 26: Cross Section of Soil Beneath Obliterated Earth Ring (Parker, 1907:518, Figure 18). The outline of Parker’s compact horizon was then traced onto a map, which evidenced a crescent shape. Parker believed that the earthwork was originally a full circle, but cliff erosion at the site had removed the northern portion of it (Figure 23 and 25). It was Parker’s belief that the area within the earth ring was the area most modified by the occupation (Parker 1907:519) (Figure 26 and 27). This was based on his observation that the occupation soil within the ring was the most deeply stained and was intermixed with “waste products from native activities” (Parker 1907). Parker found another occupation layer outside the ring that he reports was limited in scope (Parker 1907: 519) (Figure 23, 26 and 27). Figure SEQ Figure \* ARABIC 27: Parker's Excavation Notes of the Earth Ring and Associated Pit Features. Note: Excavation Trenches Cut Through the Earth Ring Area (Parker 1907, Plate 4). Parker mentions that earth rings are found in many places in western New York and were often palisaded. He believed this also applied to the Ripley Site, where he believed the site was located within circular walls of sharpened posts. Because he found evidence for occupation outside the earth ring, he also states in his report that a number of either less cautious or crowded out families probably dwelled outside the walls (Parker 1907:519). Sullivan by contrast suggests that the earthwork was most likely used as a ceremonial enclosure because there was no evidence for palisades and her excavations produced few artifacts (Sullivan et al. 1995). However, it is possible that the 80-year difference in excavations may have led to the site being picked over by pothunters. Local accounts of pot-hunting are common and local folklore states that one of Parker’s field assistants Everett Burmaster would dig up artifacts and sell them in town to pay for his weekends of entertainment (Sullivan et al. 1996:30). Subsequent excavators of the site also encountered remains of the earth ring. Jordan Christensen began excavation of the Ripley Site in 1957 where he encountered the remnants of the earth ring, but his main focus was on the site’s burials (Sullivan et al. 1996:39). The location of the earth ring also appears on Conklin’s map in his 1962 report (Figure 28) (Conklin 1962). Conklin’s information is of particular interest. In his paper, he provides a brief discussion of his interpretations of the depositional processes at the site. He described the upper layer of soil as a yellow sand, ranging from 25.4 – 35.5 cm (10-14 in) deep in most places and he noted that most of the features took place under this sand. Under the sand, Conklin notes a clay layer that he describes as being in the southwest corner of the knoll. In the northeastern part of the knoll, Conklin mentions that the occupation layer was 45.7-50.8 cm (18-20 in) deep with no indication of post molds or refuse where people could have lived (Conklin 1962). On the crest of the knoll the village deposits were only 5-7.6 cm (2-3 in) deep but in other areas they were 25-38 cm (10-15 in) deep with refuse from “top to bottom” (Conklin 1962). Figure SEQ Figure \* ARABIC 28: Conklin's Hand Drawn Map that was redrawn from a Copy on File at the Rochester Museum & Science Center (Sullivan et al. 1996:50, Figure 4.9). Despite this history of excavation, the earth ring and the site are not well understood. Much of the earth ring has been lost due to the previous excavation, land erosion, and plowing of the site, but it still may be possible to locate it, or at least find some remnants of it to reconstruct its location. Although the existing information from the Ripley Site may be flawed due to numerous avocational archaeological excavations; a lack of screening, a lack of provenience, and a general scarcity of technique, large portions of the site remain intact. This has led to new research questions, and new work on the site. Current Archaeological Research In the fall of 2012, the Mercyhurst Archaeological Institute (MAI) began to show interest in the Ripley Site. Upon consultation with the local landowner and various visits to the site, a pedestrian survey was conducted in May of 2013. After artifacts were flagged, surface collections of artifacts using an electronic total station and GPS to determine artifact densities of the site commenced (Figure 29). By the fall of 2013, stratigraphy testing via shovel test probes (STP) was conducted to determine the horizonation and soil formation of the site. This also was conducted to locate any unknown features. The majority of these STPs were outside the current site boundary. Allison Byrnes, Allen Quinn, and David Pedler of MAI completed a spatial analysis of surface artifacts in 2014 that was presented as a paper in the Society for American Archaeology in Houston, Texas (Byrnes et al. 2014). Based on this data, it is believed that this site was a village occupation site, but further documentation will need to be completed. Also at this time, there were two undergraduate students working on documenting Arthur C. Parker’s 1907 camp, and looking for various clay outcrops that Parker noted in his report (Parker 1907). By locating the clay outcrops, it helped to make more sense of the ceramic artifacts that had been found on the site. Excavations in the 2014 field season had taken place on the eastern extent of the site near Young’s Creek. These excavation’s results were presented as a paper at the 80th Annual Meeting for the Society for American Archaeology at San Francisco, California in April 2015 (Byrnes et al. 2015). After digging 18 STPs and observing the geologic history of this location, 787 ceramic sherds and 542 flaked stone artifacts were recovered. The detailed geoarchaeologic data of this location analyzed by Curtis McCoy discovered that the artifact densities and soil profiles were both of a colluvial nature. A comparison of modified and unmodified debitage from the village and terrace locations demonstrates that there is a very slight statistical difference between these assemblages. Therefore, it appears that this terrace does not represent a functionally unique midden locality, but rather an area of secondary deposition of materials eroded from the above village. Most recently in 2015, excavations by MAI has been taking place in the western portion of the bluff. Some artifacts have been found, but work is still in progress in this location. The location of the earthwork was lost since the most recent excavations in that location was in the late 1980s and early ‘90s. Numerous maps and sketches show the earthwork, but none provides geographical coordinates of where the ring actually is. It was then decided to begin work to try to find the defining feature of the Ripley site. Numerous ideas were debated, but it was decided that the best way to conduct this investigation would be to use geologic soil coring to analyze the soil morphology, and map the spatial extent of the earthwork by looking at changes in soil development and horizonation. If buried soils are found, chances are this could be the location of the earthwork. Figure SEQ Figure \* ARABIC 29: General view of the Ripley Site facing North. Note: Colored Pin Flags Indicating Artifact Locations on the Surface. Figure SEQ Figure \* ARABIC 30: View Looking Northwest Over Dewey Knoll (Parker 1907:471, Plate I). PART 2: SITE ANALYSIS CHAPTER 8: METHODOLOGY ArcGIS Methodology The research for this project initially began from December 2012 to February 2013. During this time, a collection of historic aerial photographs and topographic maps were obtained by scrutinizing archives and online GIS databases. Using topographic maps from 1898, 1978, 1986, and 2000, as well as aerial photography from 1951, 1994, 2004, and 2008, the analysis on the Ripley Site bluff erosion was conducted. All maps were downloaded and uploaded into ESRI ArcGIS 10.2 (ESRI 2011). Using a modern base map as a reference, all maps were geo-referenced and overlain on fixed points such as roads, highways, and buildings. The shorelines were estimated and digitized then converted into shape files. The distance between the lines was then quantified to determine the amount of erosion that has taken place. In early 2014, David Pedler of the MAI geo-referenced one of Arthur C. Parker’s maps from 1907 using ESRI ArcGIS based on the data above. This map provided critical information regarding the possible remainder of the earthwork and how much of the site has been lost due to erosion. Based on the information provided on this map, initial determinations on where to take soil cores was then decided and a methodology was developed. Field Methodology Fieldwork for this research occurred from September-November 2014. To start data collection, the possible core locations had to be “tied” into the existing MAI grid. To do this, a Nikon EDM (Total Station) was placed on the 500 E and 500 N datum located in the northwestern portion of the site (Figure 31). Using this instrument, a transect of five cores was placed at 15 m intervals along the northern most extent of the site near the cliff face. Based on David Pedler’s map, it was then decided to place a 6th core in between cores four and five (7.5 meters from each) in order to narrow down a location where the earth ring possibly transected. The cores were then highlighted using pin flags. At a later date, a compass and meter tape were used to emplace core transects AJ7- AJ12 and AJ13-AJ18. Using a 3 ¼ x 10 in AMS soil recovery auger, cores were taken at roughly 20 cm intervals and each was placed into 8 x 10 in artifact collection bags with careful consideration to not disturb the pedostratigraphy (Figures 32 and 33). The sampling strategy applied here is a variation of Incremental Sampling (National Soil Survey Center 2012:8-2). While aim was to take specific sample intervals, various levels of compaction and rock inclusions often affected the sample integrity while coring, resulting in inconsistent sample depths. After removing each interval, a tape measure was placed in the hole to determine the metric depth. The plastic artifact bags were then labeled and placed into brown paper bags and brought to the MAI Geology Laboratory for analysis. A total of n = 92 sample intervals in 18 cores were removed from the site (Figure 40). Figure SEQ Figure \* ARABIC 31: General View of the Ripley Site Facing East. Note the Total Station Used to Map Core Locations from Datum. Figure SEQ Figure \* ARABIC 32: View of the Author Augering for Soil Samples. Figure SEQ Figure \* ARABIC 33: Example of Equipment Used to Take Soil Samples. Laboratory Methodology In the laboratory, each core was mapped qualitatively onto pre-designed soil core sheets and attributes such as Munsell Color, pedostratigraphic depths, textures, and possible horizons were noted. Following this, a 100-gram sample of each soil interval was taken and heated in a crucible at 350 °C for 20 minutes to remove some water content so the sample could easily make it through the sieve. The dry soil weight was noted. The sample was then poured through a set of geologic sieves (-1 , 0 , 1 , 2 , and dustpan) and each sample at each phi size was weighed using a Mettler Toledo Mc303 Precision Balance 320. The percentages of the whole of each phi size were then documented using Microsoft Excel. After this was complete, the overall sum of all phi sizes were collected and the entire percentage of the weight of sands were noted. This same process was done using the dustpan percentage. Beckman Coulter LS 13 320 Laser Diffractions Particle Analyzer Analysis A more detailed analysis on particle size was then conducted using a Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer (Figure 34). Each sample was prepared using Mercyhurst’s standard Operating Procedure that is the following: A 0.3 g sample from each interval bag was taken and placed into a test tube. Around 5 ml of deionized water was added and the sample was placed onto a Fisher Scientific Vortex Genie 2 G-560 to saturate all soil particles. Then, 5 ml of sodium hexametaphosphate (44 g/L solution) was added to the test tube and placed onto a Burrell Wrist Action Shaker on setting five for approximately one hour to separate any clay bonds that may have occurred. Each sample was then run through the Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer using aqueous mode. Each test tube was placed in the auto prep station and the instrument was set to include the following: 1. PIDS data, 2. Auto Rinse, 3. Measure Offsets, 4. Align, 5. Measure Background, 6. Measure Loading, and 7. Enter Sample Info. Pump speed was set to 50, run length was 60 seconds, sonicate during run was checked (to separate particles more), sonicate power was at five, and a box was checked to compute sizes and check during runs. Each run took about 10 minutes and all 92 sample intervals were tested. The machine then output all the dates in a Microsoft Excel file that describes the particle size in microns and the percentage of value. The statistics regarding this will be described in detail below. Figure SEQ Figure \* ARABIC 34: Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer. Note: Core Samples on Cart to the Left. Organic Loss on Ignition (LOI) An analysis on the organic carbon took place using the following methodology. Samples were taken out of the soil interval bags and placed into crucibles (Figure 35). These were then placed in a Barnstead Thermolyne 30400 Furnace and set to 105 °C for one hour to remove water content (Figure 36). After, samples were removed from this furnace and set aside. Smaller crucibles were then weighed individually and the weight was recorded in a Microsoft Excel spreadsheet. Each dried sample was then individually placed into a mortar and pestle, crushed to remove clay bonds, and then ~5 g was placed into the small crucibles and the weight was recorded. Afterwards, these were then placed into a Barnstead Thermolyne 1400 Furnace and set to run for four hours at 400 °C to remove organic content. After four hours, the samples were then placed into a desiccator to cool (Figure 37). After about one hour, samples were weighed. To determine the percentage of organic content, the post-ignition weight was subtracted from the pre-ignition weight and then divided by the pre-ignition weight and multiplied by 100. Figure SEQ Figure \* ARABIC 35: Lab setup for Organic Loss on Ignition (LOI) Analysis. Figure SEQ Figure \* ARABIC 36: Samples in Furnace Ready to be Burned for LOI Analysis. Figure SEQ Figure \* ARABIC 37: Burned Samples in Desiccator After Analysis. Statistic Methodology To understand the grain sizes and stratigraphy from the site, various statistical analyses were necessary. To complete this, the Laser Diffraction Grain Size data was entered into a separate Microsoft Excel Spreadsheet for each core bucket depth and each core. The first step was to multiply the particle size in microns by the percentage of abundance for each occurrence for each sample bucket in each core. This was then added together and divided by 100 to get an average particle size number. This number was then compared on the Wentworth Scale and average particle size was noted. To determine the volume percent for each individual particle size, all particle sizes in microns and the percentage-of-whole data was extracted from the Laser Diffraction Grain Size Excel files and placed into its own spreadsheet. The particle sizes were then copy and pasted into individual columns depending on its Wentworth Scale size (clay- 0.001-0.004, silt- 0.004-0.062 mm, very fine sand- 0.062-0.125 mm, fine sand-0.125-0.25 mm, medium sand-0.25-0.5mm, coarse sand- 0.5-1mm, very coarse sand 1-2 mm). The sum of each of these percentages were then calculated and placed into another Excel spreadsheet. The total clay, silt, and sand percentages were then calculated for each core and graphed by depth using Microsoft Excel. After this was complete, the texture for each core was calculated using the online USDA Soil Texture Calculator. The percentages of sand, silt, and clay were inputted into this calculator and the calculator placed them onto a predesigned ternary diagram. Each corresponding color represents a separate core, but depths of the cores are not indicated on these diagrams. This calculator also indicates the texture type based on the quantitative particle size data that can be found in Table 2. The next analysis was a logic test to search through the numerical data and determine if any anomalies exist in each core. This method would help to locate sand abundance, buried soils, and general differences in the soil that do not fit the trend. Using the average particle sizes and organizing the data by depth, two separate columns were made to indicate whether the average particle size in each corresponding level was smaller or larger then above. If the function indicated an anomaly (assuming particle size is larger or smaller with depth), it placed a 1, if there is no difference it placed a 0. In the final column the sum of 1s and 0s were calculated. If a 2 or 0 was present, the core depth was highlighted as a potential anomaly. To accomplish this, the function for smaller than above was = If (X2<X3, 1, 0), the function for smaller than below was = If (X3<X4, 1, 0), the X is the Excel cell number for the core above or below. After completing this analysis, the data was highlighted and those horizons were analyzed in more depth. The assumption for this test was that the grain sizes will either be larger towards the bottom of the core as it is a till plain, or that the average grain sizes will be consistent throughout the core. The results from this analysis is found in Table 3 below. The final statistical analysis was done by calculating the Organic Loss on Ignition (although this analysis itself is not statistical). As mentioned above, the formula used to find the percent of organic matter was % Organic Matter (%OM) = the post-ignition weight (POW) - pre-ignition weight (PRW) and then divided by the pre-ignition weight (PRW) and multiplied by 100. This was done for each bucket depth in each core. At the end, these were then graphed using a bar chart. The results are found below.  CHAPTER 9: RESULTS ArcGIS Results/ Objective 1 Using the topographic map from 1898, the shoreline was almost 60 m (196 ft) different from the modern 2011 satellite image. When comparing all the topographic maps together, the 1898 map was around 38 m (124 ft) away from the 1951 aerial photograph and the 1978 topographic map shorelines. The 1951 map is roughly 12 m away from the modern shoreline. When comparing a topographic map and aerial image from 1978 and 1951 they both had identical shorelines. All of the newer images (1994, 2004, 2008, and 2011) show no difference in shorelines. Figure 39 depicts Parker’s excavation over a modern base map where the earth ring was potentially located. Figure SEQ Figure \* ARABIC 38:Map Depicting Shoreline Erosion Since 1898. Note: * Indicates the Use of Topographic Maps. Figure 39: Map Depicting Shoreline Erosion Since 1898. Note: * Indicates the Use of Topographic Maps. When comparing the shoreline with David Pedler’s overlay of one of Parker’s maps, the earthwork looks to be about 90% eroded. Figure SEQ Figure \* ARABIC 40: Map Depicting the Approximate Location of the Earth Ring Based on Parker’s 1907 Map and Shoreline Erosion. Note the Red Circles Represent Locations Where the Cores Were Taken In Relation to the Primary Datum. . Figure SEQ Figure \* ARABIC 41: Overlay of Parker's 1907 Excavation Notes on Current Base Map. Note: Trenches Through the Previously Documented Earth Ring Location (Google Earth 2015; Parker 1907)  Soil Core Maps/Objective 3 Below are the results after mapping and documenting the pedostratigraphy. Figure SEQ Figure \* ARABIC 42: Soil core map for AJ1-AJ6. Note: Arbitrary Levels, Munsell Colors, and Pedostratigraphy are Documented. AJ3 is Only 60 cm Deep Due to Rock Impasse. AJ 6 is 7.5 m between AJ4 and AJ5. All Munsell Designations are Wet. These Cores are from West to East. Figure SEQ Figure \* ARABIC 43: Soil Map for Cores AJ7-AJ12. Note: Arbitrary Levels, Munsell Colors and Pedostratigraphy are noted. The Void in AJ10 was Most Likely Caused by Bioturbation. All Munsell Designations are Wet. These Cores are from East to West. Figure SEQ Figure \* ARABIC 44: Soil Map for Cores AJ13-AJ18. Note Arbitrary Levels, Munsell Colors and Pedostratigraphy. All Munsell Designations are Wet. These Cores are from West to East. Characterization of Auger Samples All samples upon removal from the ground all seemed to have very similar attributes. All of them had an Ap Horizon of around 0-25 cm (some deeper than others) that exhibited more organics than deeper locations with a structure that was granular, sometimes massive and contained subangular blocky peds (Figures 45-47). The Munsell designation of the Ap horizons varied from a 10 YR 2/2 very dark brown, to a 10YR 3/3 dark brown, to 10 YR 3/2 very dark grayish brown (Figures 42-44). Some of these darker horizons spread down into the 25-40 cm range with most of the Munsell designations being a 10YR 2/2 very dark brown or a 10YR 3/2 very dark grayish brown. Between the 35-50 cm range many of the cores had spread into a Weak B or arguably a sandy 2C horizon (described more in the grain size results) with a structure of granular, sometimes massive and when present, subangular blocky peds. In this arbitrary range the average Munsell color was a 10YR 3/4 dark yellowish brown and a 10 YR 4/6 dark yellowish brown. The last 50-80 cm (50 cm depth to 130 cm depth) of each core varied but 10 YR 4/4 dark yellowish brown was generally dominant. The structure in this level was predominantly granular. Some other colors that were present were 10 YR 4/3 brown, 10 YR 5/6 yellowish brown, 10 YR 5/4 yellowish brown, 10 YR 3/6 dark yellowish brown, and 10 YR 4/6 dark yellowish brown. The soil cores nearest the shore line (AJ2-AJ8) excluding the stratigraphy of AJ1 were disturbed and the stratigraphy seemed to be upturned after the Ap horizon. For soil cores AJ9-AJ18 the pedostatigraphic colors were very similar, but there were some interesting anomalies in the cores. Core AJ10 had a large void between 55 and 90 cm. This void was probably due to bioturbation and rodent burrowing. Core AJ5 and AJ15 both had very large O/AP horizons that were from 0-50 cm thick. This thick Ap horizon may be related to the RAP excavations, previous excavations, or maybe even the previously occupied surface. Core AJ17 between 40 cm and 70 cm had very mixed stratigraphy and appeared upturned. This same phenomenon occurred in Core AJ18 between 35 and 75 cm. Mottling and disturbance was indicated in many of the deeper layers in the cores, and this occurred in cores AJ1from 25-100 cm, AJ2 from 32-72 cm, AJ3 35-60 cm, AJ4 from 25-130 cm, AJ5 from 40-55 cm, AJ6 45-112 cm, AJ7 between 25 and 100 cm, a rodent burrow from 55-90 cm in AJ10, AJ16 from 25-75 cm, AJ17 from 40-70 cm, and AJ18 from 35-75 cm. Overall, the majority of cores seemed to follow the same trend of containing darker soils at the surface and lighter soils/sediments deeper in the ground. Quantitative Texture Analysis Results The texture analysis had interesting results. The upper strata were generally made up of a loam, silt loams or a sandy loam. The deeper strata varied from a sandy loam, to a loamy sand and sometimes contained a pure sand. Core AJ18 had textures that were all over the ternary diagram, representing silty clay loam, silt loam, sandy loam, and loamy sand. This will be discussed in the discussion below. Most of the Ap horizons were made of a loam or sandy loam, and the Bw, BC and C horizons varied from a loam, silt loam, silty clay loam, sandy loam, loamy sand, sand, and sandy clay. The figures and table below represent the findings (Figure 45-50) (Table 2). Figure SEQ Figure \* ARABIC 45: Core Map Depicting Horizons and Texture Types for AJ1-AJ6. Note: AJ3 is Only 60 cm Due to Rock Impasse. AJ6 is 7.5 m between AJ4 and AJ5. These Cores are from West to East. Figure SEQ Figure \* ARABIC 46: Core Map Depicting Horizons and Texture Types for AJ7-AJ12. Note: Large Void in AJ10 is Probably Caused by Bioturbation. These Cores are from East to West. Figure SEQ Figure \* ARABIC 47: Core Map Depicting Horizons and Texture Types for AJ13-AJ18. These Cores are from West to East. Figure SEQ Figure \* ARABIC 48: Ternary Diagram for soil textures in Cores AJ1-AJ6. Red-AJ1, Green-AJ2, Blue-AJ3, Yellow-AJ4, Black-AJ5, Orange-AJ6. Figure SEQ Figure \* ARABIC 49: Ternary Diagram for Soil Textures in Cores AJ7-AJ12. Red-AJ7, Green-AJ8, Blue-AJ9, Yellow-AJ10, Black-AJ11, Orange-AJ12. Figure SEQ Figure \* ARABIC 50: Ternary Diagram for Soil Textures in Cores AJ13-AJ18. Red-AJ13, Green-AJ14, Blue-AJ15, Yellow-AJ16, Black-AJ17, Orange-AJ18. Table SEQ Table \* ARABIC 2: Horizon, Observations/Notes, and Texture Type per Core and Depth. Core Depth (cm) Horizon Observations and Notes Texture Type AJ1 0-25 Ap sandy loam AJ1 25-40 Ap disturbed sandy loam AJ1 40-52 Bw disturbed sandy loam AJ1 52-70 Bw disturbed sandy loam AJ1 70-82 2C disturbed loamy fine sand AJ1 82-100 2C disturbed loamy fine sand AJ2 0-22 Ap loam AJ2 22-32 Ap loam AJ2 32-50 Bw disturbed loam AJ2 50-62 2C disturbed sandy loam AJ2 62-72 2C disturbed sandy loam AJ3 0-25 Ap sandy loam AJ3 25-35 Ap loam AJ3 35-50 Bw disturbed sandy loam AJ3 50-60 Bw disturbed loam AJ4 0-25 Ap silt loam AJ4 25-42 Bw mottled/disturbed loam AJ4 42-60 Bw mottled/disturbed sandy loam AJ4 60-80 2C disturbed/sandy sandy loam AJ4 80-97 2C disturbed/sandy loamy fine sand AJ4 97-110 2C disturbed/sandy loamy fine sand AJ4 110-130 2C disturbed/sandy sandy loam AJ5 0-25 Ap loam AJ5 25-40 Ap sandy loam AJ5 40-55 Bw disturbed loam AJ5 55-72 Bw At 67 cm very fine grained sandy loam AJ5 72-85 2C fine grained sandy loam AJ5 85-100 2C fine grained sandy loam AJ6 0-30 Ap loam AJ6 30-45 Ap loam AJ6 45-70 Bw mottled and disturbed sandy loam AJ6 70-91 2C mottled and disturbed/sandier sandy loam AJ6 91-112 2C mottled and disturbed/sandier sandy loam AJ7 0-25 Ap loam AJ7 25-42 Ap sandy loam AJ7 42-62 Bw disturbed loamy fine sand AJ7 62-80 2C disturbed loamy fine sand AJ7 80-100 2C disturbed sandy loam AJ7 100-115 2C sandy loam AJ8 0-25 Ap loam AJ8 25-42 Bw sandy loam AJ8 42-60 2C sandy sandy loam AJ8 60-80 2C sandy loamy fine sand AJ8 80-100 2C sandy loam AJ9 0-25 Ap loam AJ9 25-40 Ap/Bw sandy at 36 cm sandy loam AJ9 40-55 2C sandy sandy loam AJ9 55-72 2C sandy sandy loam AJ9 72-102 2C sandy sandy loam AJ10 0-25 Ap loam AJ10 25-40 Ap loam AJ10 40-55 Ap/Bw loam AJ10 55-90 Bw-Void sandy/ large void probably rodent burrow sandy loam AJ10 90-115 2C sandy sandy loam AJ11 0-20 Ap loam AJ11 20-40 Ap/Bw loam AJ11 40-58 Bw sandy loam loam AJ11 58-75 Bw/2C sandy sandy clay AJ11 75-95 2C sandy sandy loam AJ12 0-25 Ap loam AJ12 25-42 Ap/Bw loam AJ12 42-60 Bw silt loam AJ12 60-76 2C sandy loam AJ12 76-90 2C finer grains of sand sandy loam AJ12 90-105 2C loam AJ13 0-25 Ap silt loam AJ13 25-40 Bw sandy loam AJ13 40-55 2C sandy loam AJ13 55-72 2C sandy loam AJ13 72-85 2C sandy loam AJ13 85-100 2C sandy sandy loam AJ14 0-25 Ap loam AJ14 25-40 Ap/Bw loam AJ14 40-55 Bw sandy loam AJ14 55-70 2C sandy loam AJ14 70-85 2C sandy fine sand AJ14 85-100 2C sandy sandy loam AJ15 0-25 Ap loam AJ15 25-40 Ap loam AJ15 40-62 Ap/Bw very thick Ap loam AJ15 62-82 2C sandy loam AJ15 82-100 2C sandy sandy loam AJ16 0-25 Ap silt loam AJ16 25-42 Ap/Bw mixed/disturbed loam AJ16 42-55 BC mixed/disturbed loam AJ16 55-65 2C mixed/disturbed loam AJ16 65-75 2C mixed/disturbed loam A16 75-90 2C sandy loam A16 90-101 2C sandy loam AJ17 0-25 Ap loam AJ17 25-40 Ap/Bw loam AJ17 40-52 Bw mixed/disturbed sandy loam AJ17 52-70 2C mixed/disturbed fine sand AJ17 70-85 2C sandy sandy loam AJ17 85-100 2C sandy silt loam AJ18 0-20 Ap silt loam AJ18 20-35 Bw mixed sandy loam AJ18 35-50 BC mixed/disturbed sandy/ large clasts loamy fine sand AJ18 50-75 BC mixed/disturbed sandy/ large clasts silty clay loam AJ18 75-92 2C sandy silt loam AJ18 92-106 2C sandy silt loam Sieve Grain Size Analysis Results The grain size analysis for the sieve analysis provided unclear results, but did provide some interesting data. In the first bucket (average 0-25 cm deep) the range of the total weight percentage of sand (-1 to 2) was from 81.11% to 99.89% with an average of 91.17% (Figure 44). The total weight percentage of sand for the last bucket in each core decreases to a range of 68.13% to 96.87% with an average grain size being 88.00% (Figure 45). Most of the cores decrease in sand percentage weight the deeper they go. For the full table on this data view the attached Appendix. Overall, the percentage of weight for each phi size varied widely amongst depths and cores. No specific pattern was indicated, but the data is available to view on the Appendix. Figure SEQ Figure \* ARABIC 51: Graph Depicting Sand Percentage in Weight for 0-25 cm in Each Core. Figure SEQ Figure \* ARABIC 52: Graph Depicting Percentage in Weight for Bottom of Each Core. Figure 53: Graph Depicting Percentage in Weight for Top and Bottom of Each Core. Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer Analysis Results The results for the particle distribution per depth varied depending on each core, but overall the general trend seems to indicate that sand, although abundant throughout the entire profile, was most prolific as the cores go deeper in the ground. In contrast, the cores had more silt towards the top near ground surface and less abundant deeper in the ground. The clay content remained consistent throughout most of the cores and remained at less than 20% with the exception of Core AJ18. Although these are the general trends, there also was some particle size anomalies found in the cores. Core AJ12 from 0-76 cm deep had more silt than sand in all of its bucket depths (0-25 cm, 25-42 cm, 42-60 cm, and 60-76 cm). Core AJ13 also had more silt in the level 72-85 cm while right next to it, AJ14 had 90% sand at the same arbitrary level (70-85 cm). Core AJ17 at the level of 85-100 cm had over 60% silt, while the level above that at 70-85 cm only had around 30%. Core AJ18 also exhibited an interesting distribution with silt being the most abundant particle size between 50-106 cm. The general range of clay size particles throughout all the cores was from 1.11% in Core AJ14, 70-85 cm to 28.86% in Core AJ18, 50-75 cm. The range of silt size particles varied from 3.35% in Core AJ14, 70-85 cm to 70.25% in Core AJ18, 0-25 cm. Finally, the range of sand particle sizes was from 28.69% in Core AJ13, 0-25 cm to 80.67% in Core AJ4 97-110 cm. A more thorough breakdown of sand percentages can be found in the attached Appendix. The graphs representing the results are below. Logic Test Results The Logic test using Microsoft Excel indicated many anomalies in average particle size per strata. The following table (Table 1) highlights each core, core depth, average particle size found in that depth measured by the Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer, the corresponding grain size as indicated by the Wentworth Scale, and the number that the logic test outputted being in 0, 1, and 2. As mentioned above, a 0 means that the average particle size in the corresponding depth was larger than the bucket above and below it, a 1 is the ideal number indicating that the depth above/below had an average particle size that was smaller or larger . A 2 indicates that the average particle size in an arbitrary level is smaller than both the levels above and below it. Table SEQ Table \* ARABIC 3: Excel Logic Test Results. Note: The Green Highlights Indicate Those Levels that have an Average Particle Size Smaller than Above and Below It, the Yellow Highlights are Levels that have a Larger Average Particle Size Above and Below It. Core Depth (cm) Average Particle Size (µm) Average particle Size Minimum? AJ1 0-25 240.0568317 Fine Sand 0 AJ1 25-40 214.0923054 Fine Sand 2 AJ1 40-52 258.8728118 Medium Sand 1 AJ1 52-70 373.0647607 Medium Sand 1 AJ1 70-82 428.1295707 Medium Sand 0 AJ1 82-100 413.8901116 Medium Sand 1 AJ2 0-22 191.7742129 Fine Sand 1 AJ2 22-32 164.6293306 Fine Sand 2 AJ2 32-50 175.7982256 Fine Sand 1 AJ2 50-62 198.7547398 Fine Sand 1 AJ2 62-72 308.8826776 Medium Sand 1 AJ3 0-25 184.630648 Fine Sand 1 AJ3 25-35 172.142783 Fine Sand 2 AJ3 35-50 467.4793049 Medium Sand 0 AJ3 50-60 180.3152707 Fine Sand 1 AJ4 0-25 122.9186825 very fine sand 1 AJ4 25-42 190.9835217 Fine Sand 1 AJ4 42-60 362.0076008 Medium Sand 1 AJ4 60-80 428.4013252 Medium Sand 0 AJ4 80-97 309.2607022 Medium Sand 1 AJ4 97-110 298.0081926 Medium Sand 2 AJ4 110-130 319.3843953 Medium Sand 1 AJ5 0-25 160.1677944 Fine Sand 1 AJ5 25-40 219.5012291 Fine Sand 0 AJ5 40-55 156.3963189 Fine Sand 2 AJ5 55-72 193.0940618 Fine Sand 1 AJ5 72-85 196.9574777 Fine Sand 1 AJ5 85-100 226.6453677 Fine Sand 1 AJ6 0-30 176.5329661 Fine Sand 1 AJ6 30-45 210.4868087 Fine Sand 0 AJ6 45-70 180.8533708 Fine Sand 2 AJ6 70-91 253.0270497 Medium Sand 1 AJ6 91-112 389.8978642 Medium Sand 1 AJ7 0-25 189.1395862 Fine Sand 1 AJ7 25-42 280.6215081 Medium Sand 1 AJ7 42-62 298.2782061 Medium Sand 0 AJ7 62-80 263.8995868 Medium Sand 1 AJ7 80-100 251.453661 Medium Sand 2 AJ7 100-115 399.6929964 Medium Sand 1 AJ8 0-25 182.6101221 Fine Sand 1 AJ8 25-42 236.4333253 Fine Sand 1 AJ8 42-60 288.1332964 Medium Sand 1 AJ8 60-80 320.0748983 Medium Sand 0 AJ8 80-100 178.544582 Fine Sand 1 AJ9 0-25 146.1828174 Fine Sand 1 AJ9 25-40 185.7374762 Fine Sand 1 AJ9 40-55 227.1502964 Fine Sand 1 AJ9 55-72 250.224173 Medium Sand 0 AJ9 72_102 219.9125813 Fine Sand 1 AJ10 0-25 141.762267 Fine Sand 1 AJ10 25-40 163.3454611 Fine Sand 1 AJ10 40-55 186.0539159 Fine Sand 1 AJ10 55-90 193.7042847 Fine Sand 1 AJ10 90-115 205.292929 Fine Sand 1 AJ11 0-20 157.1545797 Fine Sand 1 AJ11 20-40 151.7516667 Fine Sand 2 AJ11 40-58 170.1051513 Fine Sand 1 AJ11 58-75 187.3883232 Fine Sand 1 AJ11 75-95 275.3376246 Medium Sand 1 AJ12 0-25 155.0564893 Fine Sand 1 AJ12 25-42 145.2852699 Fine Sand 1 AJ12 42-60 138.2167236 Fine Sand 2 AJ12 60-76 171.7406037 Fine Sand 1 AJ12 76-90 243.1611389 Fine Sand 0 AJ12 90-105 206.9050671 Fine Sand 1 AJ13 0-25 98.32087089 Very Fine Sand 1 AJ13 25-40 150.8813554 Fine Sand 1 AJ13 40-55 153.3220902 Fine Sand 1 AJ13 55-72 185.9170764 Fine Sand 0 AJ13 72-85 172.4609555 Fine Sand 2 AJ13 85-100 291.7257505 Medium Sand 1 AJ14 0-25 166.404685 Fine Sand 1 AJ14 25-40 188.1812081 Fine Sand 1 AJ14 40-55 191.1679245 Fine Sand 1 AJ14 55-70 200.9383499 Fine Sand 1 AJ14 70-85 1350.13695 Very Coarse Sand 0 AJ14 85-100 250.6755435 Medium Sand 1 AJ15 0-25 158.3838834 Fine Sand 1 AJ15 25-40 179.8448497 Fine Sand 1 AJ15 40-62 147.1307439 Fine Sand 2 AJ15 62-82 200.2019405 Fine Sand 1 AJ15 82-100 283.1021106 Medium Sand 1 AJ16 0-25 150.1079329 Fine Sand 1 AJ16 25-42 172.9564375 Fine Sand 1 AJ16 42-55 218.3043424 Fine Sand 0 AJ16 55-65 158.9589915 Fine Sand 1 AJ16 65-75 154.4652483 Fine Sand 2 AJ16 75-90 169.374457 Fine Sand 1 AJ16 90-101 187.949551 Fine Sand 1 AJ17 0-25 183.9941359 Fine Sand 1 AJ17 25-40 156.7496149 Fine Sand 2 AJ17 40_52 245.2372048 Fine sand 1 AJ17 52_70 1207.040698 Very Coarse Sand 0 AJ17 70_85 239.1366461 Fine Sand 1 AJ17 85_100 63.81140507 Very Fine Sand 1 AJ18 0-20 28.75709357 Silt 1 AJ18 20-35 597.839833 Coarse Sand 1 AJ18 35-50 756.4745405 Coarse Sand 0 AJ18 50-75 12.37596314 Silt 2 AJ18 75-92 18.17831777 Silt 1 AJ18 92-106 27.27315 Silt 1 Results for the Organic Loss on Ignition Analysis (LOI) The results for the LOI analysis indicate that more organic matter was present in the upper strata (usually the Ap horizon) and usually had an abundance of over 1-1.5% organic matter. The deeper levels had significantly less organic matter present. The range in percentages for the upper strata varied from 0.83 % in the 0-25 cm level in AJ7 to 2.37 % OM in 0-25 cm of Core AJ4. The range in percentages for the bottom strata varied from 0.24 % OM in the 90-101 cm level of Core AJ 16 to 0.88 % OM in the 62-72 cm level in AJ2. It also appears that the levels below the Ap horizons were also fairly homogenous with organic material all being below 1% OM. The data is represented graphically below. CHAPTER 10: DISCUSSION Map Discussion The results from the ArcGIS analysis was surprising and interesting. The fact that the 1898 map is far out into Lake Erie makes sense and 60 m of erosion in 113 years is supported by the data in this and other published studies (Geir and Calkin 1983; Knuth and Lindenberg 1995). We know that the hand drawn map by Parker was made in 1907, so naturally we would overlay his map on the 1898 map because they would be the most similar. If this is the case, a little less than half of the site has eroded into the lake. In addition, the mapping also indicated that the shoreline has eroded more than 12 m since 1978, which is possible. The aerial photograph from 1951 lines up almost perfectly with the map from 1978, but one would have imagined that more erosion would have taken place between these years. This may just be the resolution of the map, which often has a 15 m error margin (USGS 1999). It is also important to take into consideration that when the USGS prints topographic maps, they often repeat them using the same or similar data they had easy access to. It may be the case that the 1978 map used data that was from an earlier time period. Topographic maps and satellite photos have different resolutions and often pose a problem when attempting to compare the two. However, when utilizing the 1951 aerial photography, the site was still about 12 m larger than it is today. When comparing topographic data from 1986 and 1994, they both seem close, indicating little erosion between these years. Regardless of which maps were used, it is clear the site does not have the same structural integrity that it did when it was excavated in the 1900s, 1960s, or 1990s. The site has changed over time, and bluff erosion is a very serious concern for this location. It is a wonder as to how far the site once extended if this much has been lost in only a 113-year time span, while the site was last occupied around 500 years ago. When comparing this data to the New York Lake Erie shoreline erosion study conducted by Geir and Calkin (1983), the average short-term recession rate in a 36-year period from 1938-1974 was about 0.10 m/yr. (0.30 ft/ yr.) or a total of 3.6 m (11.8 ft). If we use this as a proxy and extend this ratio to a 60-year time period (1951-2011), we can theoretically expect to get a total loss of about 6 m (20 ft) of shoreline erosion. Geir and Calkin’s long-term study, which took place from 1875-1974 found that the average for the Ripley location was 0.15 m/yr. (0.5 ft/ yr.) of erosion per year. When taking their 100-year study and applying it to the Ripley Site from 1898-2011, we should expect a total loss of 16.95 m (55 ft). However, after comparing this average data to the location near Ripley (Reach 10) on their figure (Figure 19), the numbers for the Ripley long term rate are probably closer to an average of 0.17-0.22 m/yr. (0.56-0.7 ft/yr.) or a total of 19.21 -24.86 m (63-79.1 ft) recession rate for a 113-year period. With this data in mind, the estimation of ~60 m of shoreline loss using the topographic map from 1898 was far too high, however, erosion rates are variable with time and there may have been catastrophic shoreline events that caused more erosion then than today. At most, the site has lost around 25 m. When comparing the Geir and Calkin figure data of shoreline lost to the 1951 aerial photograph, we should theoretically get a total recession rate of about 10.2 -13.2 m (33.5-43 ft) lost in a 60-year period. This happens to fit exactly into the ballpark of what the ArcGIS analysis had determined, which was ~ 12 m (39 ft). This is an incredible amount of erosion for a 60-year time period. This data also corresponds well with the map that was created depicting the remains of the earthwork. If 12 m of shoreline had been lost in a 50-year time period, then we could expect that 24 m has been lost since Parker excavated the site in 1907. Judging by his map and the considerable amount of erosion that has taken place since his excavation, David Pedler’s geo-referenced map depicting the earthwork extent seems accurate. When taking the Pedler map concept of the earth ring location and overlaying a Parker 1907 excavation map onto a Google Earth base map, Figure 38 shows that most of the earthwork was trenched and fallen into the Lake. Based on this analysis, only a very small portion of the earth ring in the southwestern quadrant possibly remains. Based on this information, it is important to consider that the Ripley Site is at danger of being destroyed in the future. This also helps to entertain more questions such as “How many sites have already eroded into the lake? and “how many more will in the near future?” Soil Core and Horizon Discussion After initially mapping the cores qualitatively based on color and texture, some interesting results were apparent. Cores AJ1-AJ6, or the cores along the bluff, appeared to be disturbed and there was not perfect horizon data. After looking at older maps more closely, and comparing the older hand drawn maps to the ArcGIS analysis, it is apparent that both Parker (1907) and Conklin (1962) in fact excavated this part of the site. Parker’s trenches went right through where the earthwork was present and he notes that the earth ring was found here by stating it as “having a very hard and compact soil, with the occupied soil being covered with a layer of sand and gravel” (Figure 22) (Parker 1907:518). The author found no evidence of this gravel layer, and the soil was not very hard or compact, so it may be possible that this was disturbed. This would explain why the deeper soil structure was mostly granular. Conklin also indicated that he had found the earthwork and as mentioned above, he discusses that this feature of the site was in a “yellow sand from 25-35 cm deep” (Conklin 1962). All the cores have an Ap horizon in at least the first 25 cm created by the farming since the late 1820s (as noted by Morse, pg. 84; Parker 1907:519). However, under this stratigraphic layer in cores AJ5-AJ18 a yellowish sand does exist which may be the same pedostratigraphic layer as noted by Conklin. With so many previous excavations, it is assumed that many of the cores are disturbed. For example, Core AJ4 initially appeared to be darker soil than the rest and this may be due to mixing and disturbance of the ground during these prior excavations because AJ4 was directly in the center of a previously trenched area. AJ5, AJ6 and AJ15 all turn into a sand at around 45-55 cm which may also indicate that this Ap horizon is thicker in some areas or potentially areas containing “back dirt” from excavations. The soil structure for the Ap is fairly common, being massive and granular so this would only point that this horizon is an Ap. Many of the cores have a very weak B or Bw horizon, going from a loam to a sandy loam or even a straight sand. This correlates with the fact that this site had developed on top of a till bluff with a glaciofluvial parent material. In other words, the site lies on what is believed to be an Inceptisol or more specifically a Typic Dystrocrept. Inceptisols are soils that exhibit minimal horizon development, are more developed than Entisols (commonly found as sand dunes), but still lack the features that are characteristic of other soil orders. Many of the bottom pedostratigraphic layers were found to be mottled which can also help to further indicate the soil was previously excavated or disturbed by water pooling. This occurred in cores AJ1from 25-100 cm, AJ2 from 32-72 cm, AJ3 35-60 cm, AJ4 from 25-130 cm, AJ5 from 40-55 cm, AJ6 45-112 cm, AJ7 between 25 and 100 cm, a rodent burrow from 55-90 cm in AJ10, AJ16 from 25-75 cm, AJ17 from 40-70 cm, and AJ18 from 35-75 cm. Many of these bottom layers were identified as being a 2C horizon because of the lack of silt and clays and the abundance of sediment and sand. It is also known that this site, like most northeastern soils, has an eolian cap representing a discontinuity right at the C horizon. By qualitatively examining the soil, it appears that there has been a lot of disturbance of the site and that previous excavation might have contributed to the lack of an earthwork signature. The soil structure in the deeper levels also points to it being disturbed because it was predominantly granular in all the cores. If the structures were more prismatic or columnar (the natural structure in the site) we may be able to infer that the site was un-disturbed. The earthwork could not be identified by using this method. Texture Analysis Discussion The texture of the soil at the site was typical for this location and soil type. In this type of soil known as the Chenango Series, the Ap is typically a gravelly silt loam but can be anything from sandy loam to silt loam, the Bw is also a gravelly silt loam but can include fine sandy loam, sandy loam, loam, very fine sandy loam, or silt loam, the BC is typically a gravelly loam but can include very fine sandy loam, fine sandy loam, loam, or silt loam and the 2C is typically a gravelly loamy coarse sand, but can be a loamy fine sand, sandy loam, loam, or silt loam. In the findings above, all of these horizons exhibited these textures. Two notable layers that stood out was AJ14 from 70-85 cm and AJ17 from 52-70 cm. These levels were the only levels to contain a fine sand in between layers of sandy loam. It is always an option that this could be the earthwork signature, but if the erosion data map in Figure 36 and 37 was correct, then this area would not contain remains of the earthwork. Core AJ18 also had really interesting textures types. In AJ18 50-75 cm, it had a silty clay loam, but right above this layer a loamy fine sand was present and below this layer in 75-92 cm a silt loam was present. It is possible the small samples used to conduct the study might have been either contaminated or misrepresented, but this anomaly is interesting to note. The author noted these layers as being a 2C horizon, but this may be the remains of an A2 or the old living surface of the site. Another horizon that stood out from the rest was AJ11 58-75 cm. This level’s texture was a sandy clay and was the only level to have this type. It is possible this type of clay is similar to the clay outcrops Parker (1907) describes in his report. As mentioned, AJ18 had a silty clay loam but this level’s results may have been corrupted. Sieve Analysis Discussion The sieve analysis on the soil was not very effective. While sieving did provide the author with a weight percentage, the sieve was not able to accurately separate out the silt and clay particles. Sieving the sediment and soil is often times an effective way to separate grain sizes, but it is not very effective to quantitatively determine a count of each grain size like the laser diffraction analysis can. It is important to note that the weight in percent of the whole in sand was much higher than the laser analysis. The weight in the sieve results were at the top most layer 81-over 99%, while the bottom most levels exhibited 83- over 93%. This analysis only took the weight of the whole sample as a percentage, and sand by default is heavier than silts or clays. Although this analysis did not help much, it did show that sand was most abundant in weight, which makes complete sense. Beckman Coulter LS 13 320 Laser Diffraction Particle Size Analyzer Analysis Discussion Unlike the sieve analysis, this method really helped give the author an idea of the weights and count percentages of the particle size distributions. The general trend that the soil gets sandier as the core is deeper makes logical sense and goes with the horizon and texture data. In addition, having less silt the deeper into the core is rational because the upper most horizons represent the Ap or anthropogenically altered soil. Theoretically, the living surface will contain more organics and since this has been farmed and plowed for more than a century, this plowed horizon would contain siltier and more organic particles. As indicated above, Core AJ12 from 0-76 cm deep had more silt than sand in all of its bucket depths (0-25 cm, 25-42 cm, 42-60 cm, and 60-76 cm). This is interesting because it may be that this location has been disturbed in some way. Core AJ13 also had a high percentage of silt in the level 72-85 cm but right next to it, AJ14 had 90% sand at the same arbitrary level (70-85 cm). This could be an archaeological feature or another indication of disturbance within the soil. Core AJ17 at the level of 85-100 cm had over 60% silt, while the level above that at 70-85 cm only had around 30%. Once again, this might be some indication of an anomaly in the soil, and at this depth, it may be a burial. Core AJ18 also exhibited an interesting distribution with silt being the most abundant particle size from 50-106 cm. Having more silt in the lower strata on this type of geologic formation (bluff) in an inceptisol is really interesting and important to note. The Ripley Archaeological Project (RAP) happened to conduct their excavations right near where Cores 17 and 18 were placed. This could potentially explain the un-natural distribution of organics in this location. While mentioning that these all could be anomalies or disturbances in the soil, it may also be possible that these were simply isolated autogenic events that were emplaced by some climatic episode. While the closest streams and springs are off the bluff, more research might be needed to determine if these are archaeological deposits, disturbances, or simply geologic or climatic phenomena. Overall, the diffraction analysis provided good results that was used for other analyses in this project such as the quantitative texture analysis, Excel Logic test, and the average particle size. Having an exact percentage of sand, silt and clay particles graphically represented helped to make sense of the sediment and soil stratigraphy of the site. This analysis also provided a breakdown of sand particles into very fine, fine, medium, coarse, and very coarse. This data can be found in the attached appendix, but this texture breakdown was not necessary for the results. Logic Test Discussion The Excel Logic test provided the author with a way to highlight possible anomalies in the pedostratigraphy by using average particle size. This test assumed that particle sizes got either larger or smaller with depth. While this test was thought to be good in theory, using average particle sizes is not the best way to represent particle distribution. For example, some of the levels could have had a lot of coarse sand and silt, but the average represents the level as a fine or medium sand. Many of the stratigraphic anomalies highlighted in the logic test were also not significantly larger or smaller than the levels above and below. However, some were very different. Core AJ17 52-70 cm had an average particle size of 1,207.04 µm or very coarse sand, but the levels above and below have around a 240 µm average which is more so represented by a fine sand. Core AJ18 also had a few anomalies present. In the 0-20 cm layer, it had an average particle size of 28 µm or a silt, but the layer right below this, 20-35 cm, had 597 µm or a coarse sand. The next level, 35-45 cm, had an average of 756 µm – coarse sand and the level below it, 75-92 cm, had an average of 18 µm or a silt. Core AJ18 once again, probably had corrupted data due to previous excavation. AJ14 70-85 cm also has a high average particle size of 1,350 µm or very coarse sand with the sizes above and below it being around 240 µm or a fine sand. It is interesting that a particle size can change so drastically between such small distances. This analysis made it clear that some levels were of interest, and often these levels appear to be of interest in the other analyses as well. Overall, the average grain size seems to be a fine or medium sand and this corresponds with the soil texture and particle size analysis where a sandy loam or loam are the most common texture types and sand is the most abundant out of all the levels in all the cores. This analysis was useful to highlight some anomalies in the cores, but in general, it was not as useful as expected. We can use this as one more tool in the “toolbox” to understand the site stratigraphy. Organic Loss on Ignition Analysis Discussion The LOI analysis provided the author with some interesting results. Overall, seeing that the Ap or surface level horizons were all around 0.83 % to 2.30% organic matter was fairly interesting. One would expect the plow zone or artificial anthropogenic horizon that was created by farming to contain more organics, but that does not seem to be the case. In the soil morphology section, Figure 18 indicates that the soil on the site should be no more than 4%, which is indicated in our results. As the cores go deeper, it is apparent that less organics are found which goes along with the soil horizon data, the particle size, and textures. The deeper layers usually contain more sand or sediment, which in turn, they contain less soil, silt, and clays. Every single graph had more organics at the top most layers and the organics gradually decreased with depth. This is important because it may indicate that we had not found the living surface from previous occupation that should be a buried A which is at least expected to contain material that is more organic. This analysis may also help to point out that the cores were disturbed. Excluding the recently plowed and farmed Ap horizon, all the horizons and levels below that seem to indicate very similar levels of organics. Maybe this is due to the mixing and disturbance of previous excavation? It is unknown whether this is the case, but it is a possibility.  CHAPTER 11: CONCLUSION The Ripley Site earthwork, an elusive earth ring on the eastern shoreline of Lake Erie has proved difficult to rediscover. Objective 1 was to review the current shoreline of the Ripley Site using ArcGIS and determine the nature and extent the site has been lost. Overall, based on the data obtained, around 25 m of the site has been destroyed by shoreline erosion. This conclusion alone is enough for the archaeological community to be concerned that many other sites are also at risk of being lost which can create voids to the better understanding of occupation along the Southeastern Lake Erie Basin. The area of the Ripley Site was much larger than it is today, and with this comes the realization that a lot of archaeological evidence has already gone missing into Lake Erie. However, one important conclusion is that we can use sites such as Ripley as time markers to determine long-term erosion rates of the Lake Erie bluffs. Objective 2 was to determine the site formation and stratigraphy of the site using soil coring. The site location on the bluff was formed during the last glacial retreat, but the immediate geomorphology of this location is a wave-cut bluff ~ 18.29 m (60 ft) above the shoreline that provided an excellent location for a settlement as indicated by the archaeological evidence. Through the online database by the University of California Davis, using an algorithm and predictive modeling, the site should have had an Ap horizon from 0-20 cm, Bw1 from 20-30 cm, Bw2 from 30-51 cm, BC from 51-76 cm, and 2C from 76-183 cm. Based on the auger mapping analysis, much of the site has been disturbed from previous excavations. Although it is entirely possible that there was overlooked evidence, the algorithm was more than likely correct. Any soil scientist knows that soil science and horizonation is often very subjective, but based on the soil observations, the main pedostratigraphy was Ap, Bw, BC, and 2C horizons. It is often difficult to determine the difference between a Bw1 and Bw2 horizon, but the UC Davis program was not far off from our field findings in which we found the soil to be an inceptisol. Objective 3 was to determine the current extent of erosion of the earthwork using ArcGIS, and based on the data obtained by geo-referencing Parker’s 1907 map, over 90% of the earth ring has been lost to Lake Erie bluff erosion. Once again, it is imperative that the archaeological community not only understand that there is a very high risk of site deformation along the Lake, but to understand that earlier sites have already been lost and many more important sites are at risk of being destroyed due to the tyranny of preservation. Lake Erie is such an enormous body of water that it is obvious it has been used for centuries to transport goods and people, provide fresh drinking water, and to be a major entity of subsistence for the Erie and other tribes whose communities set up villages along its shores. With this in mind, it is recommended that future work be done to try to relocate as many sites along the shoreline as possible and put this into an easy to use database. Objective 4 was to attempt to identify the earthwork signature using a variety of scientific techniques. Although no location for the earthwork was found, other valuable information was discovered. It was determined that there was a number of anomalies in the soil such as thicker A horizons and very large C horizons, and further research into these locations using excavation or shovel testing might be useful. It was also determined that the Ap horizon across the site was not uniform, re-instating the fact that a lot of the site was previously excavated. In addition, having levels that were mottled also reinforces that it is possible to re-locate previously excavated locations. The particle size analysis indicated that the soil is sandier and less silty with depth which is the original deposition of the material at the site. However, the middle horizons of the soil were often very close in particle sizes and do not necessarily indicate whether they are or are not disturbed. This same pattern of deposition was also evident in the LOI analysis, where there is fewer organics with depth indicating a sandier soil. Both of these analyses point out that the site is often anthropogenically modified at the surface and that farming with organics or manure is definitely present. This is a reminder that the site was plowed over and farmed since the early 1820s. Although Parker and Conklin discovered the earth ring in their investigations, the site has been modified numerous times because of them and since them. Between shoreline bluff erosion, trenching, excavation, plowing, farming, and general weathering of the site and downslope accumulation of material, only a very small amount of the earthwork is possibly left intact if it is there at all. Once again, this problem contributes to the much bigger picture of understanding that the Ripley Site and how it is in peril and at risk of being destroyed in the near future. Future work is recommended at the Ripley Site. A remote sensing analysis using a magnetometer might help to locate the remaining 10% of the earthwork, but with so much site disturbance this might not be useful because most of the site has already eroded away. Using chemical analyses such as a soil phosphate examination might be useful to locate the previous living surface. Finally, the last examination that is recommended which can potentially be useful is Optical Spin Luminescence or OSL dating to get a date of deposition for the Bw, BC, and C horizons. This analysis will help the principal investigator better understand when the deeper horizons were last exposed to sunlight, which will help develop a terminus ante quem for the site occupation. The Ripley Site is at risk and if we do not make an emphasis on documenting this site, we may lose important data for better understanding the Southeastern Lake Erie Basin Region in the future. ACKNOWLEDGEMENTS The author of this paper would like to thank all who assisted in the completion of this study. The author wishes to thank his committee for reviewing and assisting with edits, advice, and providing a direction to go with this project. The author would also like to thank Dr. Mary Ann Owoc and the rest of the Archaeology Graduate Class of 2014 in the thesis prep class for spending entire class periods reviewing my work and editing the technicalities. Special thanks are also due to my friends Michelle Simpson and Michael Reilly Jr. for letting me stay at their homes, often for weeks at a time, for free (I am sorry for coming in at 4 AM most nights). This thesis would honestly not have been completed without your help and hospitality. Thanks are also due to Dr. James M. Adovasio and Mercyhurst for accepting me into the program, providing a direction, providing me with help, and buying that expensive soil bucket auger for this project. Thanks to Luis L. Cabo for teaching and assisting me with the Excel Logic test. Thanks to my friends David W. Parker, Carter S. McIver, and various other friends/colleagues who constantly told me to put work on hold to go back to Mercyhurst to finish this project and provide me with more ideas for analyses. Thanks to my parents who constantly pressured me to finish this project. Lastly, I really would like to thank Lauren M. Smith; you put up with my ranting, my discussions, you edited my paper, helped me digitize the soil cores, and you always provided good company.  REFERENCES CITED Abler, Thomas S. 1970 Longhouse and Palisade: Northeastern Iroquoian Villages of the Seventeenth Century. Ontario History LXII: 17-40. Barnett, PJ. 1985 Glacial Retreat and Lake Levels, North Central Lake Erie Basin. In Quaternary Evolution of the Great Lakes, by P.F. Karrow and P.E. Calkin pp. 185-194, Ontario: Geological Association of Canada Special Paper 30. Bartley, Heather D. 2006 Geoarchaeology of an Eroded Mississippian Mound: The Belmont Neck Site (38KE6), Wateree River Valley, South Carolina. Manuscript on file, University of Georgia, Athens, Georgia. Beckerman, Ira C. 1981 Paleo-environment. In Cultural Resources in the Southern Lake Erie Basin: A Predictive Study, edited by S.A. Curtis and J.W. Hatch, Chapter 2, pp. 1-116. Report prepared for the U.S. Army Corps of Engineers by Argonne National Laboratory. Division of Environmental Impact Studies. Biggar, Henry P. (editor) 1922 The Works of Samuel de Champlain, edited by H.P. Biggar, Vol. VI. The Champlain Society, Toronto. Bolsenga, S.J. and C.E. Herdendorf 1993 Lake Erie and Lake St. Clair Handbook. Wayne State University Press, Detroit. Braun, E. Lucy 1950 Deciduous Forests of the Eastern North America. The Blakiston Company, Philadelphia. Brewer, R. 1972 Use of Macro- and Micromorphological Data in Soil Stratigraphy to Eludicate Surficial Geology and Soil Genesis. Journal of the Geological Society of Australia 19:331-344 Brose, David S. 1973 A Preliminary Report on Recent Excavations at the South Park Site, Cuyahoga County, Ohio. Pennsylvania Archaeologist 43(1): 25-43. 1976a An Initial Summary of the Late Prehistoric Period in Northeastern Ohio. In The Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium Revised, edited by D.S. Brose, pp. 25-47. The Cleveland Museum of Natural History, Cleveland. 1976b Conneaut Fort: A Prehistoric Whittlesey Focus Village in Ashtabula County, Ohio. Pennsylvania Archaeologist 46(4):29-77. 1977 Archaeological Reconnaissance and Sub-surface investigation of the Proposed United States Steel Corporation Greenfield Steel Plant, Ashtabula County, Oho, Erie County, Pennsylvania. Cleveland Museum of Natural History, Cleveland. 1978a Late Prehistory of the Upper Great Lakes Area. In Northeast, edited by B.G. Trigger, pp. 569-582. Handbook of North American Indians, vol. 15, W.C. Sturtevant, general editor. Smithsonian Institution, Washington, D.C. 1978b A Model of Changing Subsistence Technology in the Late Woodland of Northeast Ohio. In Lithics and Subsistence, edited by D.D. Davis, pp.87-115.Vanderbilt University Publication in Anthropology No. 20. Nashville. 1994 The South Park Village Site and the Late Prehistoric Whittlesey Tradition of Northeast Ohio. Monographs in world Archaeology No. 20. Prehistory Press, Madison, Wisconsin. Bursey, J.A. 1995 The Transition from the Middle to Late Woodland Periods: A Re-Evaluation. In Origins of the People of the Longhouse: Proceedings of the 21st Annual Symposium of the Ontario Archaeological Society, edited by A. Beckerman and G. Warrick. pp. 43-54. Ontario Archaeological Society, Ontario. Bush, David R. and Charles Callender 1984 Anybody But The Erie. Ohio Archaeologist. 35(1): 31-35. Bush, David R. 1984 The Erie Indians and the Whittlesey Focus: Late Aborginal Life in Northeastern Ohio. Lake County Historical Society Quarterly 26(4) pp. 17-25. Byrnes, A.M., A.G. Quinn, D.R. Pedler, and C.A. McCoy 2015 Midden or Slope Wash? Site Formation at the Eastern Edge of the Ripley Site, Chautauqua County, New York. Paper Presented at the 80th Meeting of the Society for American Archaeology, San Francisco. Byrnes, A.M., A.G. Quinn, and D.R. Pedler 2014 A Spatial Analysis of the Surface Artifacts from the Ripley Site: An Iroquoian Manifestation on Lake Erie. Paper Presented at the 79th Annual Meeting of the Society for American Archaeology, Austin. Calkin, P.E. and B.H. Feenstra 1985 Evolution of the Erie-Basin Great Lakes. In Quaternary Evolution of the Great Lakes, by P.F. Karrow and P.E. Calkin. Geological Association of Canada Special Paper 30. Carpenter, E.S., K.R. Pfirman, and H.L. Schoff 1949 The 28th Street Site. Pennsylvania Archaeologist 19(1-2): 3-16. Chapman, R. 1981 The Emergence of Formal Disposal Area and the “Problem” of Megalithic Tombs in Prehistoric Europe. In The Archaeology of Death, edited by R. Chapman, I. Kinnes, and K. Randsborg. Cambridge University Press, Cambridge. pp. 71-81. Cheyney, T.A. 1859 Ancient Monuments in Western New York. In New York State Museum 13th Annual Report, pp. 37-52. The New York State Education Department, Albany. Coakley, J.P. and C.F.M. Lewis 1985 Postglacial Lake Levels in the Erie Basin. In Quaternary Evolution of the Great Lakes, by P.F. Karrow and P.E. Calkin, pp. 195-212, St. John’s, Newfoundland: Geologic Association of Canada Special Paper 30. Conklin, Carlton S. 1962 The Dewey Knoll Site at Ripley, New York. Ms. On file, New York State Museum, Albany. 1989 The Dewey Knoll Site at Ripley, New York. The Iroquoian 16:3-29. Coyne, James H. 1895 The Country of the Neutrals: (As Far as Comprised in the County of Elgin), from Champlain to Talbot. St. Thomas, Ontario. Cross, Marlene, Mike Campbell, and Marti Martz 2007 Vegetative Best Management Practices: A Manual For Pennsylvania/Lake Erie Bluff Landowners. http://www.paseagrant.org/wp-content/uploads/2013/01/Vegetative-BMPs- Manual.pdf. Electronic document, accessed January 29, 2014. De Laubenfels, David J. 1966 Soil. In Geography of New York State, edited by J.H. Thompson, pp. 104 110. Syracuse University Press, Syracuse. Delanglez, Jean 1938 Some LaSalle Journeys. Institute of Jesuit History Publications, Chicago. 1943 Franquelin, Mapmaker. Mid-America 25(1):29-74. Doty, L.R. 1925 History of Genesee County: Comprising the Counties of Allegheny, Cattaraugus, Chautauqua, Chemung, Erie, Livingstone, Monroe, Niagara, Ontario, Orleans, Schuyler, Steuben, Wayne, Wyoming, and Yates. 4 vols. S.J. Clarke Publishing Co., Chicago. Dragoo, Donald W. 1977 Prehistoric Iroquoian Occupation in the Upper Ohio Valley. In Current Perspectives in Northeastern Archaeology Essays in Honor of William A. Ritchie, edited by R. Funk and C.F. Hayes III, pp. 41-47. Researches and Transactions of the New York State Archaeological Association 17. Rochester. Edson, O 1875 History of Chautauqua Anterior to its Pioneer Settlement. In History of Chautauqua County, edited by A.W. Young, pp. 36-49. Mathews and Warren, Buffalo. 1894 History of Chautauqua County. W.A. Fergusson, Boston. Ellis, William D. 1966 The Cuyahoga. Holt, Rinehart, and Winston Inc. New York. Engelbrecht, William 1987 Factors Maintaining Low Population Density among the Prehistoric New York Iroquois. American Antiquity 52(1): 13-27. 1991 Erie. The Bulletin Journal of the New York State Archaeological Association 102:2-12. 2003 Iroquoia: The Development of a Native World. Syracuse University Press, Syracuse. 2009 Defense in an Iroquois Village. In Iroquoian Archaeology & Analytical Scale. Edited by Laurie Miroff and Timothy Knapp. The University of Tennessee Press, Knoxville. Fenton, William N. 1940 Problems Arising from the Historic Northeastern Positions of the Iroquois. In Essays in Historical Anthropology of North America. Smithsonian Misc. Collections, 100: 159-252. Washington, D.C. Fiedel, Stuart 1999 Algonquians and Iroquoians: Taxonomy, Chronology and Archaeological Implications. In Taming the Taxonomy: Toward a New Understanding of Great Lakes Archaeology, edited by R.F. Williamson and C.M. Watts. pp. 193-204. Eastenbooks, Toronto. Fitting James E. 1964 Bifurcate-stemmed Projectile Points in the Eastern United States. American Antiquity 30:92-94. Flint, R.F. 1957 Glacial and Pleistocene Geology. New York: Wiley. pp. 553. Forsyth, J.L. 1973 Late-Glacial and Post-Glacial History of Western Lake Erie. Compass of Sigma Gamma Epsilon 51:16-26. Fowke, Gerard 1902 Archaeological History of Ohio. Ohio State Archaeological and Historical Society Publications, 44:2-64. French, Charles 2003 Geoarchaeology In Action: Studies in Soil Micromorphology and Landscape Evolution. Routledge, New York. Geier, Richard J. and Parker E. Calkin 1983 Stratigraphy and Bluff Recession along the Lake Erie coast, New York. New York Sea Grant Institute SEA 500-4-STRBR 85-72520, Albany, New York. Gendron, F. 1868 Quelques Particularites du Pays des Hurons en la Nouvelle France, Remarquees par le Sieur Dengron, Docteur en Medecine qui a Demeure dans de Pays-la fort Longtemps (1660), edited by I.B. de Rooles. J. Munsell, Albany. Goldberg, Paul, and Richard I. Macphail 2006 Practical and Theoretical Geoarchaeology. Blackwell Publishing, New York. Grant, W.L. (Editor) 1959 Voyages of Samuel de Champlain 1604-1618. Barnes and Noble Reprint Ed. Barnes and Noble, New York. pp. 283, 292. Green, William, and Lynne Sullivan 1997 Pits and Pitfalls; An Analysis of Pit Features and Site Function at the Ripley Site. Northeast Anthropology 53:1-22. Greenman, Emerson .F. 1935 Excavation of the Reeve Village Site, Lake County Ohio. Ohio Archaeological Society Publications 44:2-64. Guthe, Alfred K. 1958 The Late Prehistoric Occupation in Southwestern New York: An Interpretative Analysis. Research Records of the Rochester Museum of Arts and Sciences No. 11. Rochester. Harding, Victoria 2014 A Spatial Analysis of the Ripley Site. A Thesis Submitted to the School of Graduate Studies and Research in Partial Fulfillment of the Requirements for the Degree Master of Arts. Manuscript on File, Indiana University of Pennsylvania, Indiana, Pennsylvania. Harris, William R. 1896 A Forgotten People: The Flint Workers. Buffalo Historic Society Publications Vol. 4. Buffalo. Hart, John P., and Christina B. Reith 2002 Northeast Subsistence-Settlement Change: A.D. 700-1300. Electronic document, http://www.nysm.nysed.gov/staffpubs/docs/14644.pdf, accessed October 18, 2013. Hartley, R.P. 1958 Evidence for Low Lake Stages in the Erie Basin. Ohio Department of Natural Resources, Division of Shore Erosion. pp. 13. Hawkins, A. 2001 Genoa Frilled Pottery and the problem of the identification of the Wenro in Huronia. Ontario Archaeology 72:15-37. Herdendorf, Charles E. 2013 Research Overview: Holocene developments of Lake Erie. Ohio Journal of Science 112 (2): 24-36. Herdendorf, C.E. and M.L. Bailey 1989 Evidence for an Early Delta of the Detroit River in Western Lake Erie. Ohio Journal of Science 89:16-22. Heidenreich, C.E. 1978 Huron. In Northeast, edited by B.G. Trigger, pp. 368-388. Handbook of North American Indians, vol. 15, W.C. Sturtevant, general editor. Smithsonian Institution, Washington, D.C. 1988 An Analysis of the 17th Century Map ‘Nouvelle France.’ Cartographica 25 (3):67:111. Hodge, Frederick Webb (editor) 1907 Handbook of American Indians North of Mexico, part 1., Vol. 61, Smithsonian Institution, Washington, D.C. Hoffman, Bernard G. 1964 Observations on Certain Ancient Tribes of the Northern Appalachian Province. Anthropological Paper No. 70. Smithsonian Institution Bureau of Ethnology, Bulletin 191. Holcombe, T.L., J.S. Warren, L.A. Taylor, D.F. Reid, and C.E. Herdendorf 1997 Lakefloor Geomorphology of Western Lake Erie. Journal of Great Lakes Research 23:190-201. Holcombe, T.L., L.A. Taylor, D.F. Reid, J.S. Warren, P.A. Vincent, and C.E. Herdendorf 2003 Revised Lake Erie Postglacial Lake Level History Based on New Detailed Bathymetry. Journal of Great Lakes Research 29:681-704. 2005 Lake-floor Geomorphology of Lake Erie. World Data Center for Marine Geology and Geophysics Research Publication RP-3. pg. 26 Holliday, V.T. 1985 Archaeological Geology of the Lubbock Lake Site, Southern High Plains of Texas. Geological Society of America Bulletin 96, 1483-1492. 1987 A reexamination of Late-Pleistocene boreal forest reconstructions for the Southern High Plains. Quaternary Research 28, 238-244. 1989 Paleopedology in Archaeology. In Paleopedology: Nature and Applications in Paleosols, by A. Bronger and J. Catts, Eds., pp. 187-206. Catena Supplement 16. Catena Verlag, Cremlingen, Germany. 1990 Soils and Landscape Evolution of Eolian Plains: The Southern High Plains of Texas and New Mexico. In Soils and Landscape Evolution, by P.L.K. Knuepfer and L.D. McFadden, Eds., pp. 489-515. Elsevier, New York. 2004 Soils and Archaeological Research. Oxford University Press, Oxford. Houghton, Frederick 1909 Indian Village, Camp and Burial Sites on the Niagara Frontier. Buffalo Society for Natural Science Bulletin 9(3). Buffalo. Hunt, George T. 1940 The War of the Iroquois. University of Wisconsin Press, Madison, Wisconsin. Hyde, George E. 1962 Indians of the Woodlands: From Prehistoric Times to 1725. Norman, Oklahoma. Jenny, Hans 1941 Factors of Soil Formation. McGraw-Hill, New York. Johnson, W.C., J.R. Richardson, III, and A.S. Bohnert 1979 Archaeological Site Survey in Northwestern Pennsylvania, Region IV. Submitted by the Section of Man, Carnegie Museum of Natural History, to the Pennsylvania Historical and Museum Commission, Harrisburg. Jones, Arthur Edward, S.J. 1909 “Sendake Ehen” or Old Huronia. Fifth Rept. Bur. Archives for the Province of Ontario. Toronto. Jones, David, and Anne Jones 1980 The defenses at Indian Fort Road, Tompkins County, New York. Pennsylvania Archaeologist 50(1-2): 61-71. Jordan, Kurt 2007 An Eighteenth Century Seneca Iroquois Short Longhouse from the Towley-Read Site, ca. A.D. 1715-1754. In Archaeology of the Iroquois: Selected Readings and Research Sources edited by Jordan E. Kerber. Syracuse University Press, Syracuse. Kaszycki, C.A. 1985 History of Glacial Lake Algonquin in the Haliburton Region, South Central Ontario. In Quaternary Evolution of the Great Lakes, by P.F. Karrow and P.E. Calkin, pp. 109-123, St. John’s, Newfoundland: Geologic Association of Canada Special Paper 30. Kerber, Jordan E. 2007 Archaeology of the Iroquois: Selected Readings and Research Sources. Syracuse University Press, Syracuse. pp. 245. Knuth, Paul and Richard Lindenberg 1995 Monitoring Bluff Erosion Along Pennsylvania’s Portion of the Lake Erie Shoreline. Edinboro University Institute for Research and Community Service and Erie County Conservation District. http://www.gpo.gov/fdsys/pkg/CZIC-gb454-k68-p4-1995/html/CZIC-gb454-k68-p4-1995.htm. Electronic document, accessed January 29, 2015. Lantz, Stanley W. 1989 Late Woodland Occupation in Northwestern Pennsylvania. Paper presented at the 60th Annual Meeting of the Society for Pennsylvania Archaeology, Edinboro. Larkin, Frederick 1880 Ancient Man in America. Miles Davis, Washington D.C. Larson, Grahame and Randall Schaetzl 2001 Review: Origin and Evolution of the Great Lakes. Journal of Great Lakes Research 27(4): 518-546. Lennox, P.A. and W.R. Fitzgerald 1990 The Cultural History and Archaeology of the Neutral Iroquoians. In The Archaeology of Southern Ontario to A.D. 1650, edited by C.J. Ellis and N. Ferris. Occasional Publications of the London Chapter, Ontario Archaeological Society No. 5, pp. 405-456. Lewis, C.F.M. 1969 Late Quaternary History of Lake Levels in the Huron and Erie Basins. In Proceedings 12th Conference on Great Lakes Research. International Association for Great Lakes Research, pp. 250-270. Lewis, C.F.M., T.W. Anderson, S.M. Blasco, G.D.M. Cameron, and J.P. Coakley 1999 Did Early Holocene Lake Erie Experience Closed-Basin Conditions? In Abstracts: 42nd Conference on Great Lakes Research. International Association for Great Lakes Research. pp. A-70. Lewis, C.F.M., T.W. Anderson, S.M. C.W. Heil Jr, and P.L. Gareau 2012 Lake Levels in the Erie Basin of the Laurentian Great Lakes. Journal of Paleolimnology 47:493-511. . MacNeish. Richard S. 1952 Iroquois Pottery Types. Bulletin No. 24, National Museum of Canada, Ottawa. pp. 166. Mainfort, Robert C., Jr., and Lynne P. Sullivan 1998 Ancient Earthen Enclosures of the Eastern Woodlands. Gainesville: University Press Florida. Mayer-Oakes, W.J. 1952 Archaeological Problems in the Upper Ohio Valley, Part 2- The Northern Area. Pennsylvania Archaeologist 22(1): 37-39. McManamon, F.P. 1984 Discovering Sites Unseen. Advances in Archaeological Method and Theory 7:223- 292. Michlovic, M.G., D.G. Hopkins, and J.L. Richardson 1988 An Interdisciplinary Procedure for the Identification and Study of Archaeological Sites in Sedimentary Contexts. Soil Survey Horizons 29:3-8. Miller, Norton G. 1973 Late-glacial and Postglacial Vegetation Change in Southwestern New York State. New York State Museum and Science Service Bulletin No. 420. Albany. Morang, Andrew, and Jeffrey Melton 2001 Beach Erosion and Sediment Processes Study, North East Marina, Erie County, Pennsylvania. Coastal and Hydraulics Laboratory, U.S. Army Engineer Research and Development Center. Report prepared for U.S. Army Engineer District, Pittsburgh, PA. Morgan, Richard G. 1952 Outline of Cultures in the Ohio Region. In Archaeology of Eastern United States, edited by James B. Griffin, pp. 83-98, University of Chicago Press, Chicago. Morgan, Richard G., and H. Holmes Ellis 1943 The Fairport Village Site. Ohio Archaeological and Historical Society Publications 52:3-64 Murphy, C., and N. Ferris 1990 The Late Woodland Western Basin Tradition in Southwestern Ontario. In The Archaeology of Southern Ontario to A.D. 1650, edited by C.J. Ellis and N. Ferris. Occasional Publications of the London Chapter, Ontario Archaeological Society, No. 5, pp. 189-278. Murphy, J.L 1971 Whittlesey Ceramic Types. Ohio Archaeologist 21(1):98-308. Mueller, Earnest H. 1963 Geology of Chautauqua County New York: Part II Pleistocene Geology. New York State Museum and Science Service Bulletin No. 392. Albany. National Geophysical Data Center 1998 Bathymetry of Lake Erie and Lake St. Clair. Boulder (Colorado): World Data Center for Marine Geology and Geophysics Report MGG-13. 1 map Electronic document, http://www.ngdc.noaa.gov/mgg/image/images/erie_300.pdf, accessed July 17, 2014. National Soil Survey Center 2012 Field Book for Describing and Sampling Soils Version 3.0. National Soil Survey Center, Natural Resources Conservation Service, U.S. Department of Agriculture. O’Neal, Michael A., Matt E. O’Mansky and Joseph A. MacGregor 2005 Modeling the Natural Degradation of Earthworks. Geoarchaeology: An International Journal 20 (7): 739-748. Parker, Arthur C. 1907 An Erie Indian Village and Burial Site at Ripley, Chautauqua County, New York. New York State Museum Bulletin 117: 459-547; 518. Albany. 1916 The Constitution of the Five Nations, Or, The Iroquois Book of the Great Law. New York State Museum Bulletin 184: pp. 17. Phillips, D.W. 1989 Climate Change in the Great Lakes Region. In Report for the First U.S.-Canada Symposium on Impacts of Climate Change on the Great Lakes Basin. National Climate Program Office/NOAA, Rockville, Maryland. Pratt, G.M. 1979 Salvage Excavation at the Walnut Tree Site (33Cu40): Evidence for Agricultural Intensification in Northeastern Ohio. Ohio Archaeologist 29(1): 26-67. Prezzano, Susan C. 1992 Longhouse, Village, and Palisade: Community Patterns at the Iroquois Southern Door. State University of New York at Binghamton, Department of Anthropology. Quinn, A.G. 2003 Phase I Investigation of the North East Community Park Study Area, North East Township, Erie County, Pennsylvania. Mercyhurst Archaeological Institue, Mercyhurst College, Erie, Pennsylvania. Submitted to the Department of Planning, County of Erie, Pennsylvania. BHP ER#02-2303-049-B Quinn, A.G., J.M. Adovasio, D.R. Pedler, C.L. Pedler, D.C. Dirkmaat, D.C. Hyland, and M.R. Buyce. 1998 Late Woodland/Late Prehistoric Period in Pennsylvania, 68th Annual Meeting of the Society for Pennsylvania Archaeology. New Cumberland, Pennsylvania. 2000 The Orton Quarry Site (36ER243) and the Late Prehistory of the Lake Erie Plain, 65th Annual Meeting of the Society for American Archaeology, April 5-9. Philadelphia, Pennsylvania. Ramsden, Peter G. 1990 Death in Winter: Changing Symbolic Patterns in Southern Ontario Prehistory. Anthropologica 32(2): 167-181. Renfrew, Collin 1977 Space, Time, and Polity. In The Evolution of Social Systems, edited by J. Friedman and M.J. Rowlands. Duckworth, London. Ridley, Frank 1973 The Wenro in Huronia. Anthropological Journal of Canada 11(1):10-19. Ritchie, William A. 1965 The Archaeology of New York State. The Natural History Press, Garden City, New York. 1969 The Archaeology of New York State. Rev. ed. Natural History Press, Garden City, New York. Ritchie, William. A., and Robert E. Funk 1973 Aboriginal Settlement Patterns in the Northeast. New York State Museum Memoir 20. Albany. . Rozenwig, Mark 1983 Stage 2 Cultural Resource Investigations for Proposed Wastewater Treatment Facilities for the Town of Clarence , Erie County, New York. Ecology and Environmental, Inc., Buffalo. Sagard-Theodat, Gabriel 1939 Historie du Canada et Voyages Les Frefer Mineurs recollects y Ont Faits pour la Conversion des Infideles Depuis l’an 1615. 4 vols. Paris. Schaetzl, Randall and Sharon Anderson 2005 Soils: Genesis and Geomorphology. Cambridge University Press, Cambridge. Schock, Jack M. 1974 The Chautauqua Phase and Other Late Woodland Sites in Southwestern New York. Unpublished Ph.D. Dissertation, Department of Anthropology, State University of New York at Buffalo, Buffalo. 1976 Southwestern New York: The Chautauqua Phase and Other Late Woodland Occupations. In The Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium Revised, edited by D.S. Brose, pp. 89-109. The Cleveland Museum of Natural History, Cleveland. Schooler, E.E. 1974 Pleistocene Beach Ridges of Northwestern Pennsylvania. Pennsylvania Geologic Society, General Geology Report 64:38. Severance, Frank H. 1919 The Indian in Ohio. Ohio Archaeology and Historical Publications, Columbus 27:273-510. Shane, L.C.K., G.C. Snyder, and K.H. Anderson 2001 Holocene vegetation and climate change in the Ohio region. In Archaic Transitions in Ohio and Kentucky Prehistory, pp. 11-55. Kent University Press, Kent. Shepps, V.C, G.W. White, J.B. Droste and R.F. Sitler 1959 The Glacial Geology of Northwestern Pennsylvania. Pennsylvania Geologic Survey 4th Series Bulletin G 32. Shetrone, H.C. 1919 The Indian in Ohio. Ohio Archaeological and Historical Society Publications 27:273-510. Smith, David G., and Gary W. Crawford 1997 Recent Developments in the Archaeology of the Princess Point Complex in Southern Ontario. Canadian Journal of Archaeology/ Journal Canadien d’Archeologie 21(1):9-32. Snow, Dean 1996 More on migrations in Prehistory: Accommodating New Evidence Relating to the Size of Pre-Columbian North Iroquoian Case. American Antiquity 61(4):791-796. Squier, E.G. 1851 Antiquities of the State of New York. Buffalo. Electronic document, https://archive.org/stream/antiquitiesofsta00squirich#page/14/mode/2up, accessed March 13, 2014. Stafford, C.R. 1995 Geoarchaeological Perspectives on Paleolandscapes and Regional Subsurface Archaeology. Journal of Archaeological Method and Theory 2:69-104. Steckley, J.L. 1985 A Tale of Two Peoples. Arch Notes July/August: 9-15. Stein, J.K. 1986 Coring Archaeological Sites. American Antiquity 51:505-527 2001 A Review of Site Formation Processes and Their Relevance to Geoarchaeology. In Earth Sciences and Archaeology, edited by P. Goldberg, V.T. Holliday and C.R. Ferring, pp. 37-51. Kluwer Academic/Plenum Publishers, New York. Stein, J.K. and W.R. Farrand 2001 Sediments in Archaeological Context. University of Utah Press, Salt Lake City. pp. 10-28. Stothers, D.M. 1977 The Princess Point Complex. Archaeological Survey of Canada, Mercury Series, No. 58. National Museum of Canada, Ottawa. 1979 The Western Basin Tradition: Algonquin or Iroquois? Pennsylvania Archaeologist 49(3):13-30. Stothers and Graves 1983 Cultural Continuity and Change: The Western Basin, Ontario Iroquois and Sandusky Traditions- A 1982 Perspective. Archaeology of Eastern North America 11:109-142. Stuiver, M. and P.J. Reimer 1993 Radiocarbon Calibration Program Rev. 3.0.3. Radiocarbon 35:215-230. Sullivan, Lynne P. (editor) 1996 Reanalyzing the Ripley Site: Earthworks and Late Prehistory on the Lake Erie Plain. Bulletin Number 489. New State Museum. Albany. Sullivan, Lynne P., Sarah W. Neusius, and Phillip D. Neusius 1995 Earthworks and Mortuary Sites on Lake Erie: Believe It or Not at the Ripley Site. Midcontinental Journal of Archaeology 20(2): 115-142. Tesmer, Irving H. 1963 Geology of Chautauqua County, New York: Part I: Stratigraphy and Paleontology (Upper Devonian). New York State Museum and Science Service Bulletin No. 391. Albany. 1966 Upper Devonian Stratigraphy and Paleontology of Southwestern New York State (Erie, Chautauqua and Cattaraugus Counties). In Geology of Western New York Guidebook, edited by E.J. Buehler, New York State Geological Association, 38th Annual Meeting, Department of Geological Sciences, State University of New York at Buffalo, Buffalo, New York. pp. 47-52. Thomas, Cyrus 1889 The Circular, Square, and Octagonal Earthworks of Ohio. Government Printing Office, Washington. 1894 Report on the Mound Explorations of the Bureau of Ethnology. Twelfth Annual Report of the Bureau of American Ethnology. Smithsonian Institution, Washington, D.C. Thwaites, R.G. (editor) 1896-1901 The Jesuit Relations and the Allied Document. 73 vols. Burrows Brothers, Cleveland. http://puffin.creighton.edu/jesuit/relations/relations_01.html. Electronic document, accessed January 22, 2014. Tooker, E. 1964 An Ethnography of the Huron Indians, 1615-1649. Bureau of American Ethnology, Bulletin 190. Smithsonian Institution, Washington, D.C. Totten, S.M. 1985 Chronology and Nature of the Pleistocene Beaches and Wave-Cut Cliffs and Terraces, In Quaternary Evolution of the Great Lakes, by P.F. Karrow and P.E. Calkin North East Ohio: Geological Association of Canada Special Paper 30. Trigger, Bruce G. 1981 Prehistoric Social and Political Organization: An Iroquoian Case Study. In Foundations of Northeast Archaeology, edited by D.R. Snow, pp. 1-50. Academic Press, Inc., New York. 1985 Natives and Newcomers: Canada’s “Heroic Age” Reconsidered. McGill-Queen’s University Press, Montreal. 1990 The Huron: Farmers of the North. Holt, Rinehart, and Winston, New York. Turner, O. 1850 Pioneer History of the Holland Land Purchase. Jewett, Thomas and Company, Boston. United States Geologic Survey (USGS) 1999 Map Accuracy Standards. Electronic document, http://egsc.usgs.gov/isb/pubs/factsheets/fs17199.html, accessed February 8, 2013. University of California (UC), Davis 2013 California Soil Resource Lab. Electronic file, http://casoilresource.lawr.ucdavis.edu/drupal/node/902, accessed September 30, 2013. Vento, Frank; H.B Rollins, A. Vega, J.M. Adovasio, P. Stahlman, D.B. Madsen, and J.S. Illingworth 2008 Development of a Late Pleistocene-Holocene Genetic Stratigraphic Framework for the Mid-Atlantic Region: Implications in Archaeology. Paper Presented at the 73rd Annual Meeting of the Society for American Archaeology, Vancouver British Columbia, 26-30 March 2008. Viola, Herman J. 1976 The Forgotten People: The Woodland Erie. Journal of American History 63 (3) 686. Vogel, Gregory 2002 A Handbook of Soil Description for Archaeologists. Technical Paper no. 11. Arkansas Archaeological Survey, Fayetteville. Waters, Michael R. 1992 Principles of Geoarchaeology: A North American Perspective. University of Arizona Press, Tucson. Warrick, Gary 1988 The Iroquoian Occupation of Southern Simcoe County: Results of the Southern Simcoe County Archaeological Project 1985-1986. Report on file, Archaeology Unit, Ontario Ministry of Culture and Communications, Toronto, Ontario, Canada. 1996 Evolution of the Iroquoian Longhouse. In People Who Lived in Big Houses: Archaeological Perspectives on Large Domestic Structures, edited by G. Coupland and E.B. Banning. Prehistory Press, Madison, pp. 11-26. 2000 The Precontact Iroquoian Occupation of Southern Ontario. Journal of World Prehistory, 14 (4) pp. 422. 2008 A Population History of the Huron-Petun, A.D. 500-1650. Cambridge University Press, New York. Williamson, R.F. 1978 Preliminary Report on Human Interment Patterns at the Draper Site. Canadian Journal of Archaeology 2:117-122. Williamson, R.F. and D.A. Robertson (editors) 1998 The Archaeology of the Parsons Site: A Fifty-Year Perspective. Ontario Archaeology 65/66:1-161. Widmer, Randolph J. and Gary S. Webster 1981 Prehistory. In Cultural Resources in the Southern Lake Erie Basin: A Predictive Study, edited by Sue Ann. Curtis and James W. Hatch., Chapter 3, pp. 1-137. Report prepared for the U.S. Army Corps of Engineers by Argonne National Laboratory. White, Marian E. 1961 Iroquois Culture History in the Niagara Frontier Area of New York State. Anthropological Papers No. 16. Museum of Anthropology, University of Michigan, Ann Arbor. 1963a Settlement Pattern Change and the Development of Horticulture in the New York-Ontario Area. Pennsylvania Archaeologist 33(1-2):1-12. 1963b 1962 Excavation at the Henry Long Site. Science on the March: Magazine of the Buffalo Museum of Science 43(3): 51-56. 1972 On Delineating the Neutral Iroquois of the Eastern Niagara Peninsula of Ontario. Ontario Archaeology 17:62-74. 1976 Late Woodland Archaeology in the Niagara Frontier of New York and Ontario. In The Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium Revised, edited by D.S. Brose, pp. 110-136. The Cleveland Museum of Natural History, Cleveland. 1978a Erie. In Northeast edited by B.G. Trigger, pp. 407-411. Handbook of North American Indians, vol. 15, W.G. Sturtevant, general editor. Washington D.C., Smithsonian Institution. 1978b Neutral and Wenro. In Northeast, edited by B.G. Trigger, pp. 407-411. Handbook of North American Indians, vol. 15, W.G. Sturtevant, general editor. Washington D.C., Smithsonian Institution. Wright, James V. 1966 The Ontario Iroquois Tradition. Bulletin No. 210, National Museum of Canada, Ottawa. 1984 The Cultural Continuity of the Northern Iroquoian-speaking people. In Extending the Rafters: Interdisciplinary Approaches to Iroquoian Studies, edited by M.K. Foster, J. Campisi, and M. Methun, pp. 283-299.State University of New York Press, Albany. Wright, Roy 1974 The People of the Panther- A Long Tale. In Papers in Linguistics from the 1972 Conference on Iroquoian Research, edited by M.K. Foster, pp. 47-118. Mercury Series, Ethnology Division, Paper No. 10, National Museum of Man, Ottawa. 174 1