This research was initiated following an October of 2006 meeting at the offices of the Kane Hardwood Division of Collins Pine Corporation in Kane, Pennsylvania. At this time the author was contracted to design a predictive model for the locating of prehistoric cultural resources on Collins Pine property. This initiative was undertaken in response to the Collins Pines Holdings being certified into the prestigious Forest Stewardship Council.

Predictive models are useful tools that can be used to make assessments of various landform types encountered in the field, and the probability of such landforms to produce cultural resource sites. Information gathered from the model and/or the results of implementing the model, could potentially help steer forest management practices with regards to cultural resources.

The designing of a predictive model would be a large undertaking. Few predictive models have ever been specifically designed for the region, and those that do exist are now viewed as outdated. In designing the predictive model, the author would examine the data from numerous prehistoric sites found in a seven county area in an attempt to determine any possible trends that the data could suggest. Based on the findings of the study, certain areas of Collins Pine property, would be found to be more likely to contain significant archaeological sites than others. The results of this data analysis are presented herein.

INTRODUCTION

Location of Collins Pine Holdings

The Collins Pine Company is Pennsylvania’s largest private landowner. The property consists of approximately 127,000 acres housed in 187 separate compartments (Blaine Puller, 2007 pers. comm.) located in northwestern and north-central Pennsylvania in the counties of Cameron, Clarion, Elk, Forest, Jefferson, McKean, Potter, Venango and Warren.

This large region of the state is known regionally as the Allegheny Plateau and is included in the Appalachian Plateaus Physiographic Province. Individual tracts are located in varied topographical settings in five distinct physiographic sections of the Appalachian Plateau. The majority of the property is located in the High Plateau and Deep Valleys Sections, while a smaller number of tracts are located in the Pittsburgh Low Plateau, the Northwestern Glaciated Plateau, and Glaciated High Plateau Sections. The majority of the holdings are found in unglaciated regions, while a few small tracts, located in Potter and Warren Counties, occur within the glaciated portion of the Allegheny Plateau.

Map 13 DCNR depicting various Physiographic Provinces in Pennsylvania. Collins Pine holdings (study area) are located in Cameron, Elk, Forest, Jefferson, McKean, Potter, Venango and Warren Counties

A large portion of Collins Pine property is located in a region known locally as the “Big Level”. This large flat table like region is characterized by a broad and mildly undulating plateau that is not cut by any sizeable drainage for great distances. According to J.E. Henretta (1929):

“The Big Level extended from Wetmore and Kane on the south and west and to the north roughly following a line parallel with Hubert Run and Kinzua Creek almost to the New York Sate line, thence east to Bradford; from here the boundary runs in a southerly direction to Mount Jewett and from Mount Jewett the Big Level thrusts itself, one end east and north, including possibly forty percent of the area of Potter County and extending almost to Tioga County. The other loop or extension continues from Mount Jewett through Elk County and along the Cameron County line reaching almost to Ridgway”

From a standpoint of cultural influence or diffusion, the Collins Pine forests are located in a unique area near the headwaters of three major drainage basins. The majority of the property is found within the Allegheny River (Ohio River) basin while parcels in western Elk and northern Potter County drain into the Susquehanna and Genesee drainages. Cultural influence could spread from the west up the Ohio (Allegheny River); from the north from Lake Erie and Lake Ontario and from the east and southeast from the Susquehanna River. The Allegheny River drains into the Ohio before joining the Mississippi and flowing into the Gulf of Mexico. The Genesee drains north into Lake Ontario while the Susquehanna River drains east and south east to the Chesapeake Bay.

Definition of a Predictive Model:

What exactly is the definition of a predictive model? A predictive model is a tool that can be used by planners in determining the relative probability of finding archaeological sites on particular parcel of land. Predictive modeling is defined as a simplified set of testable hypotheses, based either on behavioral assumptions or on empirical correlation, which at a minimum attempts to predict the loci of past human activities resulting in the deposition of artifacts or alteration of the landscape (Kohler 1988:33).

It is the role of a good model to give land planners and managers an accurate idea of what to expect in an area so they can plan accordingly (Miller 1986:1). An untested model is a hypothesis, a forecast of what the researcher, on the basis of study and analysis, thinks should be the pattern of events in the future (Miller 1986:1). With this in mind the actual testing of the hypothesis is concluded in the form of the site survey typically referred to as Phase I investigation. When sites are located they can be maintained in a data base which over time can expand and improve the original model (Bona 1994). This information can subsequently be transferred onto sensitivity maps in an attempt to predict high, medium, and low probability areas for the finding archaeological sites.

There are typically two types of predictive models that various researches use. A predictive model of site locations is an extrapolation either from a set of theories, a set of data analyzed for trends, or both (Miller 1986:1). Theoretical models are known as deductive models, while models formed through data analysis data are known as inductive models.

A deductive model according to (Kohler 1988:37) begins with a theory predicting human behavior. Certain considerations are taken into account regarding deductive reasoning:

"A deductive model must: 1) consider how humans make choices concerning location ... This requires considering: (a) a mechanism for decision making; and (b) an end for decision making- what is the goal? 2) specify the variables affecting location decisions for each significant chronological or functional subset of sites; 3) be capable of operationalization; it must propose a means for measuring each of the relevant variables and must allow for a set of predictions that can be compared with the archaeological data" (Kohler and Parker 1986:432).

Somewhat opposite of the deductive model, the inductive model begins with data then builds conclusions based upon the information including biases inherent in the original data set. Bona (1994) notes that errors and/or biases include, site locations may be incorrectly recorded, environmental information may be recorded in too little detail, data may be missing from some records, or information gathered by previous researchers may differ in quality compared to the standards of present-day archaeologists. If the biases are included into a predictive model they, the results, will be perpetuated.

This particular study tends to borrow from both deductive and inductive models. This approach is rationalized by the fact that very little survey work has ever been conducted in the region under study, and, due to the fact, that the PASS files from which site data are extrapolated lack a great deal of information and contain numerous errors.

There are certain limitations to predictive models. No model, no matter how scientific the approach, can accurately determine the locations of all cultural resources within a particular study area. The full range of site types and relation to particular landforms is currently unknown.

One of the biggest problems with predictive models stems from the difficulty of assessing the probability of a particular landform by using topographic maps. Archaeological sites seem to occur in association with distinctive features of the landscape, however to define and locate these features requires field investigation (Lantz 1982:101). For example maps typically cannot predict the location of exposed bedrock on a sideslope, and seldom show the locations of springheads where sites commonly occur. Many small campsites cannot be distinguished from the hillslope as they occur on narrow benches that are less than the 20 foot contour interval on a 7.5 minute geological survey map Lantz (1982:99). Predictive models also cannot determine past environments, such as an old salt lick that attracted game to an otherwise inconspicuous area of the woods, or the location of old beech stands that once attracted passenger pigeons and Native American hunters who likely camped nearby. Models cannot determine where natural resources such as chert could occur. Chert was used by prehistoric people to fashion arrowheads and spear points.

While there are a number of caveats to formulation and use of predictive models, they also aid planners in uniformly making heritage resource management decisions across large areas in a timely and cost effective manner (Bona 1994). With a model as a guide, there can be steps taken to identify and mitigate potential threats to significant archeological sites.

Prehistoric Site Definition:

So what exactly is considered a prehistoric site a/k/a archaeological site? The following is a definition of a prehistoric site developed by the Section of Archaeology, The State Museum of Pennsylvania and the Bureau for Historic Preservation, Pennsylvania Historical and Museum Commission. This work was published in 1991 and entitled “Cultural Resource Management in Pennsylvania: Guidelines for Archaeological Investigation”.

1.) Three or more culturally modified objects, excluding fire cracked rock (FCR), found within a 15 meter diameter area while surface collecting a plowed field represent a site.

a.) Three or more flakes or any combination of flakes and tools (lithic, pottery sherds, etc.) found within 15 m of each other represent a site in plowed contexts.

b.) Single points are usually considered isolated finds, however, all Paleo-Indian point localities should be recorded as sites because of the sparseness of debris from Paleo-Indian sites.

c.) Isolated diagnostic artifacts (points, ceramics, etc.) should be fully recorded on PASS site forms. These will not be assigned official PASS site numbers, but will remain on file as important information concerning prehistoric land use.

2.) Flakes and/or other culturally modified items in adjacent 15 m spaced shovel tests represent a site regardless of the number (minimum of 2) of objects recovered.

3.) Five or more finished tools recovered within an acre indicate the presence of a site, and the consultant should further refine site boundaries within that area (unless it is a village site covering more than an acre).

4.) The presence of any subsurface culturally derived feature requires designation of the locality as a site.

It must be noted that the term prehistoric is used to differentiate between later historic period sites. Prehistory as the name implies, indicates a time before written record from a European standpoint. The definition of prehistoric site as presented above is typically written in regard to open air campsites. Throughout the Allegheny Plateau there are a number of rockshelters which are in effect closed air type sites. The rockshelter pictured below is an example of a prehistoric site contained within the protective purposes of the rock itself. Typically if one cultural item such as a chert flake is found at a rockshelter it should be considered a site. Rockshelters are oftentimes considered a site based on the presence of buried hearths even if cultural remains such as chert and pottery are absent.

Rockshelter located near Ludlow, McKean County, PA is an example of a prehistoric site

CULTURE HISTORY

The section serves as a guide as to the known cultures, complexes, and traditions found throughout the region. These are the individuals that once likely inhabited Collins Pine property and whose material culture in the form of artifacts one could expect to find when a site is encountered.

Due to the overall lack of published archaeological data regarding the Allegheny River basin, the prehistory generated for this report was largely based of the findings conducted by the author at the Indian Camp Run (36FO65) site located on Collins Pine property in Forest County, Pennsylvania. Data gathered from other sites test excavated by the author, as well as published references, were also used to create this section.

This section begins with the earliest known inhabitants to enter the region, Early Man/Paleoindians, and culminate with the Historic/Contact Indian period at about the time of the French and Indian War. This section does not include an overview of the Historic period, which in this study begins with the Colonial era and continues to 1957. This research was specifically tailored to prehistoric modeling. A Historic model would require a separate study due to the changing settlement pattern and complexity of historic events that shaped the region.

The following data shows the number of sites by county and cultural affiliations associated with each site. A generic term such as Archaic could be said to include anything from Early, Middle, Late and Terminal Archaic. The same holds true for the term Woodland. Prehistoric General could be any one of the cultural periods shown below as is typically used if the cultural affiliation is unknown.

A breakdown of cultural components found in the seven county study area is as follows. A total of 26 Paleoindian sites were identified with the following results, Cameron 0, Elk 1, Forest 2, McKean 0, Potter 1, Venango 1, and Warren 21. A total of five Late Paleoindian sites were identified with the following results, Cameron 0, Elk 0, Forest 1, McKean 0, Potter 1, Venango 0, and Warren 3. A total of 69 sites exhibited unspecified Archaic components with the following results, Cameron 3, Elk 3, Forest 2, McKean 6, Potter 2, Venango 4, and Warren 49 sites. 27 Early Archaic components were identified with the following results, Cameron 1, Elk 1, Forest 2, McKean 0, Potter 1, Venango 3, and Warren 19. A total of 38 sites were identified that exhibited Middle Archaic components with the following results, Cameron 0, Elk 1, Forest 2, McKean 0, Potter 4, Venango 4, and Warren 27. A total of 96 Late Archaic compnents were identified with the following results, Cameron 2, Elk 8, Forest 4, McKean 2, Potter 1, Venango 18, and Warren 61. A toal of 63 Terminal Archaic sites were identified with the following results, Cameron 4, Elk 3, Forest 2, McKean 1, Venango 10, and Warren 42. A total of 53 sites were identified as Woodland with the following results, Cameron 11, Elk 10, Forest 4, McKean 6, Potter 3, Venango 2, and Warren 16. A total of 53 sites were identified as Woodland with the following results, Cameron 11, Elk 10, Forest 4, McKean 6, Potter 3, Venango 2 and Warren 16. A total of 96 Early Woodland sites were identified with the following results, Cameron 1, Elk 5, Forest 4, McKean 6, Potter 1, Venango 11, and Warren 68. A total of 105 Middle Woodland sites were identified with the following results, Cameron 0, Elk 3, Forest 4, McKean 3, Potter 2, Venango 15 and Warren 78. A total of 170 Late Woodland sites were identified with the following results, Cameron 10, Elk 29, Forest 9, McKean 5, Potter 3, Venango 24, and Warren 90. A total of 32 Historic/Contact era sites were identified with the following results, Cameron 4, Elk 0, Forest 3, McKean 0, Potter 1, Venango 4, and Warren 20. A total of 486 sites were termed Prehistoric General with no cultural affiliation identified with the following results, Cameron 34, Elk 70, Forest 37, McKean 30, Potter 15, Venango 179, and Warren 121.

Early Man/Paleoindian Period

The Early Man period begins at some time around 17,000 B.C. and continues to around 9000 B.C. The Paleoindian period begins around 9,000 B.C. and extends to around 8,500 B.C. Sites of the Early Man period a/k/a Pre-Clovis (older than 9000 B.C.) are considered extremely rare with only a few known sites have been located in the Western Hemisphere. It was during the Early Man and later Paleoindian periods that the initial settlement of the America’s and the local region ensued. There is some debate as to whether these individuals or their ancestors crossed the Bering Straight into North America from Asia or entered the continent by other routes.

Diagnostic tools associated with the Early Man period are unfluted biface forms. At Cactus Hill in Virginia the Early Man a/k/a Pre-Clovis point form is a triangular biface (McAvoy 1997:111). No triangular points attributed to the Early Man period are currently recognized from Pennsylvania. The Miller Lanceolate, a possible Pre-Clovis a/k/a Early Man point form, was recovered from the Meadowcroft Rockshelter located in Washington County, PA. According to Adovasio (from Carr and Adovasio 2002:8) this projectile was located on the uppermost living floor of Statum IIa of the rockshelter. The point was bracketed between 14c dates of 11,300 B.P. +/- 700 B.P. and 12,800 +/-870 B.P. This is the oldest point currently known in Pennsylvania.

Later Paleoindians produced the distinctive lanceolate fluted point which occurs in a number of type varieties. Interestingly fluted points were not crafted in Asia including the region west of the Bering Straight where Paleoindians are thought to have migrated from. Thus, the fluted point type appears to be a New World innovation. These points, including those found in western Pennsylvania, were often manufactured from high quality lithic materials likely gathered, perhaps traded, from distant regions. Paleoindians are often thought of as nomadic big game hunters that followed migrating game species such as caribou over great distances and rarely lived in one particular location for any length of time. They likely supplemented their diets with any and every available food source including small game, waterfowl, fish, and plants and berries.

The Late Pleistocene period in which these individuals lived was characterized by rapid changes in climate. A gradual warming trend occurred throughout the end of the Pleistocene although there were fluctuations in temperature such as the Younger-Dryas cooling episode where temperatures dropped to near full glacial levels. This gradual warming would ultimately lead to the abatement of continental glaciers from western Pennsylvania. As the region began to dry out and temperatures rise the environmental conditions also improved. Both the topography and the environment would begin to more closely resemble that of the present day. As conditions improved, the peri-glacial zone which included a park and tundra (sedge-cyperaceae) environment that existed south of the terminal moraine was later replaced with fir (Abies) and spruce (Picea) and then pine (Pinus) along with certain deciduous species such as Oak (Quercus) as the Late Pleistocene came to a close. According to Carbone (1974) in all areas of the unglaciated Eastern United States there is unquestionable palynological evidence for significant shifts in the vegetation which must have been responses to the shifting climatic patterns. Carbone (1974) has also noted that within a time span of approximately 2000 years a completely modern faunal assemblage was in place by ca. 9,340 +/- 1000 years B.P. The changes in flora directly effected the changes in fauna. Large mega-fauna that once occupied a place in the Late Pleistocene were replaced by species more typical of today. Mammoths and mastodons were soon gone and replaced by caribou, elk, deer, and a number of other species.

Just when the first inhabitants entered to Allegheny, Susquehanna, and Genesee Rivers is hard to say. Individuals occupying the Meadowcroft Rockshelter, are the earliest known inhabitants in the Ohio River valley (and by further extension the Allegheny River) are some of the earliest known inhabitants in the Americas. Thirty-nine of the radiometric assays run from the Meadowcroft sample postdate 12,800 B.P (ca. 10,800 B.C.) (Adovasio 1993). And, if the deepest dates associated with cultural material are averaged, then humans were present in the Ohio drainage by ca. 15,950 B.P. (Adovasio 1993: 207). It is even possible (Adovasio and Carlisle 1986:7) that pre-Clovis producing folk were in the Ohio River valley as early as 17,650 B.C., although the date has a large range of error some 2,400 years. While there has been a great deal of controversy regarding the 14C dates from Meadowcroft over the years, the early dates cannot be ignored. In fact there are a small number of sites in the Eastern United States that are thought to be pre-Clovis in age. These include Cactus Hill (McAvoy 1997: 179) in Virginia and the the Topper site excavated by Albert C. Goodyear along the Savannah river in South Carolina. Goodyear has recently obtained dates of around 50,000 B.P. for artifacts such as chert micro tools along with comparable dates for soil samples. Although controversial, these astonishing dates would be the earliest for humankind in North America (“Clovis at Topper” Mammoth Trumpet 21-4).

The majority of the Paleoindian finds in the East and in the Great Lakes region follow a much later chronology. A suggested chronology has been proposed by Deller (1989), Deller and Ellis (1986) and Stork (1988) regarding the various fluted point traditions found there. The chronology typically begins with point types such as Gainey dating to (ca. 9000-8700 B.C.) now known as Shoop-Debert/Gainey which has also been termed Eastern Clovis. This is followed by the Parkhill Complex (ca. 8600 B.C.) of which the Cumberland-Barnes type is diagnostic. The fluted point found at Indian Camp Run on Collins Pine property is included in this typology. The Parkhill Complex is synonymous with the Middle Paleo period as defined by Gardner and Verry (1979) and Anderson (1990) who defined the period based on the presence of fishtailed/waisted bases including Cumberland, Suwanee, and Simpson types to the South. Recently Spiess, Wilson and Bradley (1998: 235) have defined a new phase for the New England-Maritimes region, the Michaud-Neponset phase, which is said to be closely related to and possibly somewhat later the Parkhill phase of the eastern Great Lakes region. The Michaud-Neponset phase also exhibits strong ties with the Mid-Atlantic region as well. The Paleochronology continues with Crowfield (ca. 8600-8400 B.C.) and the later Holcombe Beach (ca. 8400 B.C.), and Hi-Lo points (ca. 8000 B.C.). An even later non-fluted Late Paleo period is somewhat problematic with disagreements occurring over whether to call diagnostic types of this time period Paleo or Archaic (Peterson 2002:127). The non-fluted point Late Paleoindian period was dated from around 8,050 to 6,050 B.C. and possibly extended to ca. 5,050 to 4,050 B.C., a temporal assessment that was based on cross-dating with better known non fluted point types found in the Great Plains (Peterson 2002:123). Lantz (1984:218) has noted to following Late Paleo Plano types: Angostura (from Johnson per co. 1984); Miller Lanceolate, McConnell Lanceolate, Stringtown Stemmed, Sawmille Stemmed (Adovasio et al 1982:259); Scottsbluff, Dalton, considered rare, and Golondrina.

Indian Camp Run (36FO65) is one of only a handful of sites in the eastern United States where Paleoindian tools have been found in buried stratigraphic context (Myers 2007). According to Dr. Frank Vento, a nationally recognized Geomorphologist from Clarion University, Indian Camp Run is the oldest site currently known from the floodplain of the Allegheny River. This assessment was based on the examination of the age of the site soils along with the diagnostic artifacts recovered from within.

Indian Camp Run has produced a number of suspected Paleoindian age tools including one whole fluted projectile, a number of fluted knives, and a host of other tools including preforms, and broken implements likely discarded during manufacture. A photo of the whole fluted projectile along with the metrical statistics was sent to Dr. Christopher Ellis of the University of Western Ontario in London who stated that if “the projectile had been found in Ontario he would not hesitate to call it a Barnes” (Chris Ellis per com. 2005). Barnes projectile points are diagnostic of the Parkhill Complex in the Great Lakes region and are thought to date to around 10,800 to 10,500 B.P. (Ellis and Deller 2000: 253). No carbon dates have yet to be run for the type, the age has been established based on a relative chronology established based on fluctuations of glacial Lake Algonquin over time. Barnes projectiles are found sporadically throughout the Great Lakes region and were originally typed by Roosa (1965) at the type station in central Michigan (Wright and Roosa 1966). The Indian Camp run site is regionally “close” to the other Parkhill Complex sites found throughout Michigan and southern Ontario. Key sites associated with the Parkhill Complex include Parkhill, Fisher, and Thedford II in Ontario, and Barnes and Leavitt in Michigan.

Fluted Barnes like projectile from Indian Camp Run dating to around 8,600 B.C.

While the number of excavated sites producing Paleoindian artifacts found buried in situ is rare, a number of fluted points have been found in plowed fields and eroding from river banks in the Allegheny River valley. A wide variety of typologies have been noted. Lantz (1984) has observed the presence of Eastern Clovis (Gainey), Debert, Hi-Lo, Holcombe, Cumberland, Folsom, Agate Basin, and Hell Gap in his study of Paleoindian sites in western Pennsylvania. In Forest County a small number of fluted projectiles have been reported. Recently (Fogelman and Lantz 2006) have pictured two fluted points from the County including one found at Tionesta. They also note two sites with a Paleo presence including (36FO3) Squire Farm-County Home and (36FO9) Wheeler Farm, both located a few miles upriver from Indian Camp Run. In Venango County (Fogelman and Lantz 2006) have noted sites with a Paleo presence include (36VE57) Potter Village, (36VE124) Polk Farm, (36VE163) Spring Hill, (36VE229) Pinehill rockshelter and (36VE240) Rod Gold.

Within the interior reaches of the Allegheny Plateau Paleoindian sites are virtually non existent or have not been found. Elk County has one recorded site, the Mohan Run Rockshelter located near Ridgway that produced a fluted point from the surface of the shelter (James Herbstritt, pers. comm, 2007), McKean has no known Paleoindian sites, Cameron County has no currently recorded Paleoindian sites, Potter County has one Late Paleoindian site located near Shinglehouse. Warren County has some 24 sites (Fogelman and Lantz 2006) 21 of which were recorded into the PASS files.

Archaic Period

The Archaic period follows the earlier Paleoindian period and predates the Woodland period. The Archaic period has been subdivided as follows (after Fryman 1982: Table 2): Early Archaic (8500-6000B.C.), Middle Archaic (6000-4000 B.C.) and Late Archaic/Transitional (4000-1100 B.C.). Different researchers follow slightly different versions of this chronology but the author prefers Fryman’s positioning of the cultural periods for western Pennsylvania. This period is characterized by the appearance of corner and side notched projectiles, the use of the atlatl, a hunting, gathering and foraging economy, and use of many rockshelters (most notably in Late Archaic times) throughout the region.

Early Archaic

A small number of Early Archaic components have been recognized from western Pennsylvania and surrounding region. Some of these will be discussed here as they relate to the prehistory of Indian Camp Run even if not specifically found at the site.

The Early Archaic components recognized from Indian Camp Run include, Lost Lake, St. Charles, possible Charleston Corner Notched, and a possible MacCorkle bifurcate. It must be noted that recent research conducted by Carr (1998) places MacCorkle, St. Albans, and LeCroy bifurcates into the Early Middle Archaic period.

Early and Middle Archaic projectiles found at Indian Camp Run

Lost Lake point types date from around 8000 to 6000 B.C. (Justice 1987:58). Points similar to Lost Lake were found below Kirk Corner Notched at the Ice House Bottom site in Tennessee (J. Chapman 1977: 161, 166). St. Charles projectiles are said to date from around 8000 to 6000 B.C. (Luchterhand 1970:12). They too were found at Ice House Bottoms in the lower Kirk levels which were dated to around 7500 B.C. (J.Chapman 1977: 51). Charleston Corner Notched projectiles were dated to around 7900 B.C. from a hearth at the St. Albans site (Broyles 1966: 18; 1971: 56). This type predates Kirk at both Ice House Bottom and St. Albans (Justice 1987: 79). Adovasio et al. (1998: 18) report finding Lost Lake, St. Charles and Charleston like projectiles during excavation at the Meadowcroft Rockshleter located in the Cross Creek drainage near Avella, PA. A number of Early Archaic projectiles (Adovasio et al. 1974: 46-47) including St. Charles types have also been reported from the Boarts site located on the Mahoning River in Lawrence County, PA.

One heavily resharpened bifurcate was recovered from Indian Camp Run. The projectile appears to be similar to the McCorkle Stemmed variety (Broyles 1966:23; 1971:71) pictured in the St. Albans site reports. These projectile types are said to be diagnostic of the Early Archaic period MacCorkle Phase (J. Chapman 1976:6) and date to between ca. 7000-6500 B.C (Broyles (1971:71). As stated above recent research by Carr (1998) suggests a possible Early Middle Archaic placement for the MacCorkle type. At St. Albans MacCorkle type projectiles were found below the St. Albans type and above Kirk Corner-Notched projectiles.

According to Lantz(1982:37) some of the more common Early Archaic projectiles found in the Allegheny Valley include Dalton, Kirk, LeCroy and various other bifurcate types ( see Carr 1998 for new temporal designation). Kirk like projectiles have been found at the Boarts site (Adovasio et al 1974: 40) and at the Penelec (36WA152) site located near Warren, PA. Several Kirks were recorded by Mayer-Oakes from the Siggins site (36FO1) located near Tionesta in Forest County, PA. LeCroy and similar bifurcates have been recovered from Warren County near Irvine and twelve bifurcated points are recorded from Siggins (36FO1) and Schwab Run sites (Lantz 1982:38). Bifurcates have also been found at the State Road Ripple site on the Clarion River in Clarion County, PA. and have been found in rockshelters along Mix Run (36WA168) by Paul Yeagle of Warren, PA (Lantz 1982:38). Kirk, LeCroy, and MacCorkle like projectiles have been recovered by (Burkett 1999:28-30) from the Fishbasket (36AR134) located along the Redbank Creek in Armstrong County, PA. Johnson, Richardson and Bonhert (1979) have also noted MacCorkle projectiles from the Pymatuning and French Creek areas of western Pennsylvania.

Middle Archaic

Proposed Middle Archaic projectiles identified from the Indian Camp Run sample include the following: Stanly/Neville like, Big Sandy II/Otter Creek like, and possibly MacCorkle. According to (Carr 1998: 77) in the Upper Ohio drainage, Middle Archaic sites are characterized by Stanly Stemmed, Big Sandy, Otter Creek, and Brewerton Side Notched types with heavy basal and notched grinding. As noted in the Prehistory found in Indian Camp Run Miscellany this author prefers to keep the Brewerton Series in the Late Archaic period, which also corresponds with Fryman's (1982) placement, due to changes in population density, changes in projectile morphology due to extensive use of localized lithic sources and, among other things, extensive utilization of rockshelter sites, which prior to Brewerton was only sporadic. This author believe the various Brewerton types represent an important time marker different from earlier components

The Stanly/Neville like projectiles found at Indian Camp Run both exhibit strait bases. Justice (1987: 99) considers Stanly and Neville to be regional variants and are considered to be morphological correlates. These types have been dated to around 6000 B.C. (Broyles 1969:35) to around 5000 B.C. (Coe 1964: 35-36). Dincauze (1976: 26-29) obtained dates for the Neville type ranging from circa. 5800-5000 B.C. at the Neville site in New Hampshire. Similar to the Stanly/Neville typology both the Otter Creek and Big Sandy II types may be regional morphological correlates. Justice (1987: 68-69) equates the Otter Creek with the Big Sandy II and others. Dates for the Otter Creek/Big Sandy II include the following: In New York at the Sylvan Lake Rockshelter dates as early as 4610 B.C. have been obtained for Otter Creek (Funk 1976); a date of 4340 B.C. was obtained at the Shafer site in the Schoharie Valley (Wellman and Hartgen 1975). There is also evidence from the Modoc Rockshelter in Illinois that the type or a related type may have Early Archaic origins with dates as early as 8000 B.C. (Fowler 1959a:19).

According to Lantz (1982: 39) Middle Archaic Big Sandy, Otter Creek and a sundry of other varieties are numerous in the upper Allegheny Region of western New York and northwestern Pennsylvania (from Dragoo and Lantz 1969; Lantz 1971). Johnson et al 1979:62 has suggested that Big Sandy-Otter Creek like projectiles are numerous in the upper Beaver drainage and the Lake Erie Plain (see Mayer-Oakes 1953; Prufer and Sofsky 1965; Brose et al. 1978) but thin out near the Allegheny River. Otter Creek like projectiles were reported from the Boarts site (Adovasio et al 1974: 47) in Lawrence County, PA; from the Wadding Rockshelter (George and Bassinger 1975) located along the Mahoning Creek in Armstrong County, PA and from at the Split Rockshelter (Herbstritt and Love 1975) located in the Clarion River drainage of Elk County, PA.

Late Archaic/Terminal Archaic Period

Late Archaic/Terminal Archaic forms identified from Indian Camp Run include the following: Lamoka, Brewerton Corner and Side Notched, Steubenville, Genesee, Susquehanna Broad and Forest Notched projectiles. A small number of Brewerton Side and Corner Notched projectiles have been recovered at Indian Camp Run.

Late Archaic Brewerton projectiles from Indian Camp Run

Brewerton projectiles typify the Late Archaic period in the upper Ohio valley circa. 3500 to around 1700 B.C. In Armstrong County at the Brown Towing Company site (36AR188) located along the banks of the Allegheny River near Kittanning, George and Davis (1986) have radiometrically dated a Brewerton component to 3330 B.C. +/- 170 years. Both Brewerton Side and Corner Notched types were recovered from similar strata at the Sheep Rockshelter in Huntingdon County, Pennsylvania (Bebrich 1968: 310, 326). According to Ritchie (1961) the type is distributed throughout all of New York state. Dragoo (1959) noted the type was a dominate type found throughout the upper Ohio valley on sites attributed to the Laurentian tradition. In Johnson, Richardson, and Bonhert’s (1979) regional study of northwestern Pennsylvania Brewerton types were easily the most common projectile type identified. Of the 1346 projectiles examined 355 or 26.37% were considered Brewerton. In the Clarion River valley, at the Split Rockshelter (36EL4) excavated by Herbstritt and Love (1975), Brewerton side notched projectiles were the most common projectile recovered accounting for (24/67) 35.82% of all projectile types. All of the Brewerton types combined accounted for (37/67) or 55.22%. Dutch Hill (36JE132) rockshelter located along the main stem of the Clarion River has also produced both Brewerton Corner and Side Notched projectiles (Myers n.d.). Brewerton projectiles were recovered by Burkett (1999) from both Fishbasket and Fishbasket North sites in Armstrong and Clarion Counties, PA. Brewerton Side Notched projectiles were recovered by (George and Bassinger 1975) from the Wadding Rockshelter in Armstrong County, PA and both Brewerton Side Notched and Corner Notched point were found at the Boarts site (Adovasio et al. 1974) in Lawrence County, PA.

Late Archaic Lamoka like projectiles from Indian Camp Run

A total of two Lamoka like projectiles have been recovered from Indian Camp Run to date. In New York State this projectile type dates from around 3500-2500 B.C. (Ritchie 1961: 29-30, 1969a:31-32). In the upper Susquehanna Valley a later Lamoka Phase occupation was dated to around 2570 to 1800 B.C. (Funk and Rippeteau 1977: 30). Recent excavations along the West Branch of the Susquehanna at Allenwood, PA have produced radiocarbon dates for a Lamoka occupation from 3700 to 4000 B.P. (Wall 2000: 10). Lamoka projectiles have been found by the author at Ham’s Rockshelter (36EL66) in the Clarion River valley. Lantz (1982:41) notes that the Lamoka Phase was well established in the upper Allegheny region at sites such as Cold Spring (30CA10), Lyn Beach (30CA11) and Quaker Bridge (30CA6). Lamoka projectiles have been recovered from the Buckaloons site (36WA95) (Lantz 1975: 6) and in many rockshelters within the Allegheny National Forest (Lantz 1982:41). Lamoka projectiles have also been recovered from Fishbasket (36AR134) on the Redbank Creek in Armstrong County and from Wadding (36AR21) (George and Bassinger 1975) and Split Rockshelter (36EL4) (Herbstritt and Love 1975) in Armstrong and Elk County respectively.

Two Genesee like projectiles were recovered at Indian Camp Run. These types date to the Frontenac Phase (Ritchie 1965) of the Terminal Archaic period and have been dated from 2980 B.C. +/- 260 (Arnold and Libby 1951: 114) to 1723 B.C. +/- 250 (Ritchie 1961:24, 1969a:108) in the Northeast. According to (Lantz 1982:41) Genesee points have been recovered from many sites above Warren, PA including Bone Run (30CA13), the Trailer Site (30CA23), and Quaker Bridge (30CA6). Large stemmed points were also recorded from the Siggins site (36FO1) in Forest County, PA. Genesee like projectiles have been reported (Adovasio 1974: 43) from the Boarts site in Lawrence County, PA.

Terminal Archaic Genesee like projectiles from Indian Camp Run

A number of Steubenville like projectiles have been recovered from Indian Camp Run. The Steubenville Stemmed and Lanceolate projectiles are included in the so-called “Panhandle Archaic” coined by Mayer-Oakes (1955b:36) and found at sites such as 46BR31 (East Steubenville), 46HK1 (New Cumberland Heights) and 46BR29 (Half Moon). Steubenville Stemmed and Lanceolate were found to be associated with the McKees Rocks Mound (36AL6) but are thought to have been brought in from a nearby site to be used as the mound fill (Mayer-Oakes 1955). A recently compiled table of dates for Late Paleo and other lanceolate point assemblages from the region (Lepper 2005:379) includes the following dates for various Steubenville types: 4220 +/-500 B.P. from the East Steubenville site (Crane and Griffin 1958; Dragoo 1959); 4120 +/-200 B.P. from the Globe Hill Shell Heap (Murphy 1977); 3970 +/- 85 years B.P. from Meadowcroft Rockshelter (Adovasio et al. 1988; Boldurian 1985:144); and from the Byler Mound 3115 +/-80 years B.P. (Zakucia 1956, 1974). Fox Creek like projectiles reported by Herbstritt and Love 1975: 34) may in fact be Steubenville typologies.

Steubenville like projectiles found at Indian Camp Run

One of the larger cultural components identified at Indian Camp Run includes those projectiles prescribing to the Late Archaic/Terminal Archaic era. More specifically several projectiles similar to Susquehanna Broad were recovered along with numerous associated steatite fragments.

Transitional/Terminal Archaic projectiles from Indian Camp Run. Projectiles conform to the Susquehanna Broadspear type and were manufactured from high grade Ohio flints. The large blade in the top row, center, was manufactured from South Mountain rhyolite found possibly in Adams County, PA

Susquehanna Broadpoints are said to date from circa. 1700 to 700 B.C. (see Justice 1987: 169). This time frame includes portions of the Terminal Archaic and Early Woodland periods . Susquehanna Broad projectiles are diagnostic of Ritchie’s (1961) Frost Island phase and have been radiocarbon dated from 1670 B.C. +/- 110 to 1520 B.C. +/- 1520 in Massachusetts (Dincauze 1968: 72-77) and (Kinsey 1972: 354) obtained a date of 1650 B.C. from the Zimmerman site in the Delaware Valley. At Zawatski terrace located on the Allegheny River in western New York a date of 1180 B.C. was obtained by (Calkin and Miller, 1977: 311). Justice (1987:167) considers the related Forest Notched (Prufer and Sofsky (1965:33-34) consider Ashtabula a Susquehanna correlate) to be morphological correlates of the Susquehanna Broadspear which likely accounts for entry of the Susquehanna Broadspear into Early Woodland times. Forest Notched projectiles are considered Early Woodland in age in the upper Ohio Valley and while they may be ancestral to Susquehanna Broadspears will be addressed below as an Early Woodland and separate typology.

While only Susquehanna Broadspears were found at Indian Camp Run other Late Archaic/Terminal Archaic forms have been identified throughout western Pennsylvania including Ashtabula, Perkiomen, Koens Crispin/Snook Kill, and Lehigh Broad.

Lantz (1982) has noted that Late Archaic/Terminal Archaic groups were extensively occupying rockshelters within the Allegheny National Forest region of western Pennsylvania including sites in McKean and Warren Counties. One rockshelter in Forest County was said to produce Transitional material. While employed by the Forest Service the author located a Perkiomen projectile from the surface of a rockshelter in Warren County, PA and during a “Rainbow Family”gathering on Queen Creek in Warren County fragments of a steatite bowl were uncovered (per com. John C. McLaughlin 1991). Both sites are located in the East Hickory Creek drainage a tributary of the Allegheny River.

Broad spear points (Johnson et al. 1979: 66) have been reported from the flood plain of the upper Beaver drainage (Mayer-Oakes 1953, 1955; Prufer and Sofsky 1965; Adovasio et al. 1974) and in the middle and upper Allegheny by (Mayer-Oakes 1955, Dragoo and Lantz 1967a, 1969; Lantz 1967a, 1971; Herbstritt and Love 1975 and George and Bassinger 1975). A Lehigh Broad spearpoint (Myers n.d.) was recovered during excavations at the Dutch Hill (36JE132) in Jefferson County, PA. Burkett (1999) has reported various broaspear types from the Fishbasket sites located along the Redbank Creek in Armstrong and Clarion Counties, PA. Elseware, in Clearfield County (Matlack 1990) has reported Susquehanna Broadpoints, Perkiomen, Lehigh, Fishtail points and steatite as having been found at a number of sites in the Susquehanna drainage. Mayer-Oakes (1955: 62) pictured a number of Ashtabula projectiles from the Lavant site in Chautaqua County, NY.

The soapstone fragments found at Indian Camp Run traveled a long way into the Allegheny River valley where the material does not occur naturally. In Pennsylvania massive metamorphic steatite (soapstone) is found in the eastern portion of the state in the Great Valley and Piedmont Sections of the Ridge and Valley and Piedmont Provinces respectively (Gordon 1959). It is worth noting that examples of South Mountain Rhyolite (vicinity of South Mountain, Adams County, PA) were recovered from Indian Camp run and identified by Jack Holland of the Buffalo Museum of Science. Also all of the Susquehanna Broadspears found at Indian Camp Run were manufactured from Flint Ridge and Upper Mercer lithics. Soapstone was being brought in from the east while high grade lithic material was apparently coming from the west.

Early Woodland Period

The Early Woodland period in western Pennsylvania spans the years1100-100 B.C. (After Fryman 1982: Table 1). This one thousand year time period is characterized by the appearance of the earliest clay ceramic containers, finely crafted stemmed and notched projectiles, burial mounds in some regions, and the increased dependency on cultivated foods. The period is typically referred to as Adena throughout the Ohio River valley as many Early Woodland sites exhibit examples of material culture related to the Adena culture including Indian Camp Run.

Early Woodland projectiles from Indian Camp Run; Top row: Kramer (Cresap), Meadowood (largest point), Adena (beaver tail), Middle row: Adena, Meadowood; Bottom row: Forest Notched, Meadowood

The major concentration of the Adena culture was located further south down the Ohio River. According to (Webb and Snow 1945:132) two major centers of Adena occupation were recognized one on the Scioto River in southern Ohio and another on the Kanawha River near Charleston, West Virginia. Lesser centers were said to have been located in Indiana, Kentucky, Ohio, and West Virginia. Some of the best known mounds include the Grave Creek mound (Norona 1957), Beech Bottom (Bache and Satterthwaite 1930), Natrium (Solecki 1953), Half-Moon (Fetzer and Mayer-Oakes 1951) and Cresap (Dragoo 1963), Murad (McMichael and Mairs 1969), Willow Island (Hemmings 1978), along with a host of other lesser known mounds.

While there are a number of known mounds in northwestern Pennsylvania these are usually attributed to the later Middle Woodland period. There are only a handful of Early Woodland mounds in the upper Allegheny valley. Mounds of Early Woodland age include, the Glacial Kame Mound (36WA17), Z2 (36WA140), Mound Site (36WA206), possibly the Nelse Run Mound (36MC29). The Willow Bay site (36WA305) found and recorded by the author while employed with the Forest Service was located while water levels dropped at the Allegheny Reservior. An Adena assemblage manufactured from Flint Ridge Ohio lithics was located in a tightly circumscribed area. Without further research it would be hard to say for certain whether this was a burial or other site type.

According to (Dragoo 1971: 203) the number of earthen mounds of supposed Adena origin decreases as we approach Pittsburgh. No earthen mounds of definite Adena origin are known for the Allegheny Valley except in the immediate area of Pittsburgh with the McKees Rocks mound representing the most eastern Adena structure of sizable proportion (Dragoo 1971).

At Indian Camp Run various items associated with the Early Woodland period include not only Adena, but items from other culture/traditions including Kramer, Meadowood and the Transitional/Early Woodland Forest Notched like projectiles. While not found at Indian Camp Run No. 1 (36FO65) Half Moon ceramics have been found at neighboring Indian Camp run No. 2 (36FO66). A mixed Early Woodland assemblage of this nature would lead one to believe that several Early Woodland groups were occupying the Allegheny River valley. The following is description of the Early Woodland components recovered at Indian Camp Run.

One classic Adena Stemmed “Beavertail” projectile and another yet unclassified stemmed (possibly Robbins like ) Adena projectile were recovered from Indian Camp Run. Adena points are diagnostic of the Early Woodland period Adena Culture dating to around 800 to 300 B.C. (Hemmings 1978: 5, 38). Dragoo (1963: 294) has proposed that the date range for Adena in the Ohio valley was circa. 700 to 70 B.C. Adena Stemmed forms were found between Cresap types and later Robbins forms at the Cresap Mound (Dragoo 1963: 110-114). The main area of the Adena Culture is in the central Ohio River valley and major tributaries (Webb and Baby 1957; Dragoo 1976b: 1). In the region of Forest County, PA (Carpenter 1942) described an unclassified Early Woodland aspect related to the Adena-Middlesex complexes. Ritchie 1965 coined Middlesex as an Adena infusion of cultural traits into the Northeast (see Ritchie 1938: 100-103; 1944: 112-115; 1951: 131-133).

Other Adena remains have been found at a number of sites throughout the region. According to (Dragoo and Lantz 1967) the bottom of a Half Moon cordmarked vessel and several Adena points of exotic flint were recovered from the Onoville Bridge (30CA5) site in Cattaraugus County, NY. Adena points were also recorded at the Onoville Store (30CA34) site and at Cold Spring (Lantz, 1971) in southwestern New York.. Cresap type projectiles were recovered by John Zavinski of Warren, PA from a possible burial mound near Garland, PA (Lantz 1982: 45). According to Johnson et al. 1979:72 a large number of Adena Stemmed and were identified in Crawford and Erie Counties, PA. particularly from collections around Pymatuning Marsh and the French Creek floodplain. Adena projectiles have been reported by (George and Bassinger 1975:16) from the Wadding Rockshelter located along the Mahoning Creek in Armstrong County, PA., and an Adena Robbins like projectile was recovered by (Herbstritt and Love 1975: 32) from the Split Rockshelter in the Clarion River drainage. Adena like projectiles were also recovered from the Boarts Site in Lawrence County, PA excavated by (Adovasio et al. 1974). Burkett (1999) has reported Adena like projectiles from the Fishbasket sites located along the Redbank Creek in Armstrong and Clarion Counties, PA. In Clearfield County, PA Matlack (1990) has reported a number of Adena projectiles from the Susquehanna Valley.

Another projectile was recovered that fits the classic description of Kramer Stemmed. The projectile type was originally contrived by (Munson 1966a; 1971: 6-7, 49, Fig.4) as Early Woodland in age and often associated with thick grit tempered pottery known as Marion Thick in the mid west. The Marion phase was a latter phase of the Red Ocher complex (Munson 1966a). Red Ocher material culture has been identified in the Allegheny River valley. Lantz (1966: 7-10) has described finding a cache of three turkey tail blades in a burial pit at Cornplanter Grant (36WA82). According to Dragoo (1963: 238) Turkey Tail points are associated with the Red Ocher culture and are thought to date to Early Woodland times. Dates associated with the Marion Phase typically average around 500 B.C. (Munson 1966b:117; 1982:3). The distribution of the Kramer type is generally confined to the lower Great Lakes Region of Illinois, Indiana, Michigan, Wisconsin and into Missouri (Justice 1987), now Pennsylvania as well.

Two projectiles similar to Mayer-Oakes (1955) Forest Notched point were recovered. The Forest Notched projectile was first identified by Mayer-Oakes (1955: 58-61) while examining examples from the Siggins (36FO1) (type site) site located a few miles upriver from Indian Camp Run in Forest County, PA. The projectile type is thought to be genetically related to the Susquehanna Broad type (Johnson et al 1978:43; George 1998: 22) and has origins in the preceding Transitional Archaic period and was still manufactured into the Early Woodland period. Dates associated with Forest Notched projectiles include the following: Feature 3 excavated by Stanley Lantz from the Ohioview site (36BV9) produced a radiocarbon date of 2690 +/- 70 B.P. With a 2 sigma calibrated range of B.C. 980-780 and an intercept of the radiocarbon age with calibrated curve of 825 B.C. Another feature contained a likely Forest Notched point associated with Half Moon ware and a date 2740 +/- 120 B.P. A feature associated with two Ashtabula projectiles was dated to 1080 B.C. +/- 90 and 930 B.C. +/-90 that according to (George 1998:32) demonstrates the possibility that Forest Notched could be descended from the Susquehanna Broadspear. Justice (1987: 169) believes the Ashtabula, Converse and Forest Notched are all morphological correlates of the Susquehanna Broadspear.

A small collection of Forest Notched projectiles was said to have been found in several rockshelter within the Allegheny National Forest (Lantz 1982: 43). Forest Notched projectiles have been reported by Burkett from the Fishbasket villages in Armstrong and Clarion Counties. Forest Notched projectiles have been found in association with Turkey Tail points in the Red Ocher burial at Cornplanter Grant (Dragoo and Lantz 1967). Johnson et al. (1979:70) report that Forest Notched projectiles were commonly found in the French Creek drainage.

Three Meadowood projectiles have been recovered to date at Indian Camp Run. Meadowood types as described by Ritchie (1961) are diagnostic of the Early Woodland Meadowood Phase which has been dated from 1300 B.C. to 500 B.C. (Ritchie and Funk 1973:116; Granger 1981:63). In the upper Susquehanna drainage they have been dated to as early as 1230 B.C. (Funk and Rippeteau 1977). Ritchie (1965: 180) has also obtained dates of 988 B.C. from Oberlander No. 2 and 841 B.C. from Hunter. At Morrow Cemetery they have been dated to 563 and 630 B.C. (Ritchie and Funk 1973).

They have been found throughout the lower Great Lakes and into the Northeast. They constitute a small percentage of the projectiles identified in (Johnson, Richardson and Bonhert’s 1979) study of Northwestern Pennsylvania (21/1346= 01.6%). They have been recovered at the Wadding rockshelter (36AR21) in the Mahoning Creek drainage in Armstrong County (George 1975); at the Split Rockshelter (36EL4) (Herbstritt and Love 1975) and Middle School (36EL94) site (Myers n.d.) in the Clarion River drainage in Elk County. Lantz (1982) has noted Meadowood points in the upper Allegheny at Quaker Bridge (30CA6), Bone Run (30CA7), and Cold Spring (30CA10). They have been recovered from likely burials along the Conewango Creek near Akeley (Lantz, 1982) and at Buckaloons (Lantz, 1975) in Warren County. They have also been recovered from excavations at the Fishbasket sites on the Redbank Creek in Armstrong and Clarion Counties (Burkett, 1999). Matlack (1990) has noted Meadowood projectiles from a number of sites along the West Branch of the Susquehanna drainage.

Many sites associated with the aforementioned projectile types also exhibit some form of thick grit tempered ceramic. In the upper Ohio River valley Half Moon Cordmarked is the most common Early Woodland ceramic and has been found by the author at a number of sites in the Allegheny River, Clarion River and Redbank Creek drainages. No Half Moon cordmarked ceramics have been positively identified from Indian Camp Run (36FO65) which is somewhat perplexing due to the preservation quality and large amount of ceramics recovered at the site. Interior cordmarked ceramics (Half Moon Cordmarked) however have been recovered a few meters away at the Indian Camp Run No. 2 (36FO66) site. These ceramics were found in association with a fireclay pipe blank (See: This website: Testing and Indian Camp Run No. 2 for details of the Early Woodland finds).

Middle Woodland Period

The Middle Woodland period ranges from circa. 100 B.C. to 1000 A.D. (After Fryman 1982: Table 1). This period is characterized by numerous burial mounds and associated , sometimes widely publicized grave offerings most of which are synonymous with the Hopewell interaction sphere.

Some of the common projectile types found in the region include Snyders and Gibsons and expanding base points such as Steubens, Chessers, Jack’s Reef/Raccoon Notched and later triangles such as Levanna and Waterford. Many of these especially in the early portion of Middle Woodland were manufactured from high quality Ohio flints.

Middle Woodland Raccoon Notched and Chesser Notched points

There are numerous Middle Woodland mounds found in the Allegheny River valley. Areas such as the Buckaloons (bean fields) and the terraces along the upper Allegheny River valley near the Pennsylvania/New York border appear to have been large regional centers. There were at least seven mounds (possibly 9) known to have existed at Irvine (Carpenter 1971:267) and many important mounds such as Sugar Run in the upper Allegheny region. And, while these groups were extensively utilizing floodplain locals for their villages and ceremonial centers they were also extensively utilizing rockshelters in the surrounding hills and adjacent watercourses including such areas as the Allegheny National Forest (Lantz 1982).

There appears to have been more than one Middle Woodland group occupying the region (Johnson et al. 1979:76; Lantz 1982:46) one possibly evolving from an Early Woodland base and the other group a possible intrusion of Hopewell people from Ohio. The first localized group built stone mounds, manufactured limestone tempered ceramic vessels, and placed very little “high church” grave offerings in their burials (Johnson et al. 1979: 76). The second group manufactured earthen mounds, produced grit tempered pottery with fabric impressions and utilized an abundance of exotic Hopewellian material (Johnson et al. 1979: 76). All of the Middle Woodland sites exhibiting Hopewellian material are associated with mounds including Cold Spring, Sugar Run and Buckaloons (Lantz 1982:51). This second group was also responsible for constructing many of the local mounds that have been assigned to Ritchie’s (1965) Squawkie Hill phase which was based on a single date of A.D. 160 +/- 100 years. Ritchie (1994:216-217) included the following mounds into Squawkie Hill: In western New York, the Squawkie Hill and Geneseo mounds (Ritchie 1938a; 1944: 202-207) and the Wheatland, Killbuck and Vandalia mounds (Carpenter 1950:307; 1950a) in western Pennsylvania, the Sugar Run mounds (Bliss 1942), Irvine, Cornplanter and Corydon mounds (Thomas 1894: 499-502; Carpenter 1956), the Nelson mound (Carpenter and Schoff 1951) and the Danner mound (Clark, Lantz and Robinson 1960).

Dates associated with some of the early Middle Woodland sites from the region include the following: a date of A.D. 35 was obtained from the Byler Mound (33TR5) (Zakucia 1974) which according to (Johnson et al. 1979: 75) is the only acceptable carbon 14 date available for mounds in this portion of the Glaciated Allegheny Plateau. Mound 3 at Buckaloons was dated by (Lantz 1982:49) to A.D. 400. The Middle Woodland occupation of the Reider site was dated between A.D. 420 to A.D. 590 (Lantz 1982: 47). Recent dates obtained by (McConaughy and Johnson 2003:109) for the Sugar Run Mound include 1700 B.P. +/-70 or A.D. 250 uncorrected and calibrated to A.D. 380.

While the Hopewell influence of the upper Ohio valley was extensive (Mayer-Oakes 1955: 216; Ritchie 1965: 228), at some time around A.D. 500 the occupations were either terminated or culturally altered (Dragoo 1963: 293). Lantz (1989) has suggested at around A.D. 500 the local Middle Woodland and Intrusive Hopewell cultures appear to have merged resulting in the Allegheny River Phase of A.D. 500-950 of which the distinctive Raccoon Notched point assemblage is diagnostic. Raccoon notched projectiles are found in association with triangular points. Some of the later occupations of the Irvine Mounds include Mound 3 which according to (Lantz 1989) contained Middle Woodland material that is later than Hopewell and more closely related to Raccoon Notched assemblage including triangular points. Mound 5 contained triangular points (Lantz 1989:50). According to (Justice 1987:220) both Jack’s Reef and Raccoon Notched projectiles attained popularity in the Late Point Peninsula culture Kipp Island Phase around A.D. 500 before being replaced by Levanna triangles in the Hunter’s Home phase (Ritchie 1969a: 254, 233, Plate 81). Jack’s Reef points are said to be diagnostic of the Intrusive Mound Culture in Ohio (Mills 1922; Ritchie 1937). The Raccoon Notched projectile is very similar to Jack’s Reef with the only difference being Raccoon Notched projectiles are side notched (Justice 1987:219).

Excavation of Irvine Mound #2 at Buckaloons, Warren County, Pa in the 1930's. Photographs taken by Harry Schoff (from Carpenter 1956)

To date very little information was gathered at Indian Camp Run that could be said to concern the Middle Woodland period. The early dates associated with Mead Island technically place that group into the late late Middle Woodland period but they will be addressed in the section detailing Late Woodland. One Middle Woodland Raccoon Notched projectile was recovered from Indian Camp Run No. 2 (36FO66) located a few meters to the east of 36FO65. This projectile was found to have been manufactured from high grade Upper Mercer chert and was found above and within and earlier Early Woodland component replete with interior cordmarked pottery and a fireclay tube which was likely a pipe blank. The projectile was also found in association with some exotic lithics (Pennsylvania jasper) from the Susquehanna drainage which is not commonly found in the Allegheny drainage, and Upper Mercer chert.

One projectile was located at Indian Camp run that could be said to be Middle Woodland in age. The projectiles appears to be a Chesser Notched form. This point type appeared during the terminal Middle Woodland period at around 300 A.D. and attained popularity during the Late Woodland period (Prufer and Shane 1970:84). The projectile was said to fit between Middle Woodland Synders and Late Woodland Jacks Reef (Justice 1987: 214). Radiocarbon dates associted with this type from sites such as McGraw, Raven Rocks, and other sites in Ohio range as late as A.D. 500 to A.D. 700 (Shane 1975b; Prufer 1981:78). No Middle Woodland ceramics are known from the Indian Camp run sample.

Late Woodland-Late Prehistoric-Protohistoric Period

Late Woodland (circa A.D. 1100 calendar years to contact with European groups) (Late Prehistoric circa A.D. 1100 to 1580/1590; Protohistoric to circa A.D. 1635).

Many of the following terms, Late Woodland, Late Prehistoric, Protohistoric are used by researchers to denote certain time periods at various locations throughout the Ohio River valley. Certain groups found in western Pennsylvania are closely related to groups further downstream in the Ohio River valley necessitating the use of the differing chronologies when discussing these groups. This was an era when new ideas and technologies were being passed into the upper Ohio drainage via diffusion from the larger groups located down river. Shell tempered pottery was one of these new innovations that first reached western Pennsylvania at around A.D. 1100 and typically denotes the so called Late Prehistoric period. Shell tempering was likely borrowed from Fort Ancient groups and spread into the West Branch of the Susquehanna River. At sites where both grit and shell tempered pottery occur, the term Late Woodland/Late Prehistoric is often used.

In a chronology devised for southwestern New York, northern Ohio and northwestern Pennsylvania (see Johnson and Myers 2004) the Late Woodland begins around A.D. 1100 in calendar years and continues until the appearance of European trade goods. The Middle Ohio River valley chronology is earlier with Late Woodland beginning at around A.D. 550 and continues to around A.D. 1150 in calendar years. The period of A.D. 1150-1580/1590 is the Late Prehistoric period. Following A.D. 1580/1590 the Protohistoric period begins and continues until around A.D. 1635 (Johnson and Myers 2004: 90).

The Late Woodland/Late Prehistoric period is generally characterized by the use of small triangular projectiles used as arrowpoints, a wide variety of clay ceramic forms, a settlement pattern consisting of hamlets, small camps, and large stockaded villages often surrounded by many acres of fields for the growing of corn and other cultigens. Located far away from the main villages/habitation areas were a number of related satellite campsites including small stockades, open air basecamps, and rockshelters which were located on or near trail systems that continued to distant locations.

Late Woodland Madison projectiles recovered at the Dutch Hill Rockshelter, Jefferson County, PA

The Collins properties cover a vast region of north-west and north-central Pennsylvania. A number of Late Woodland/Late Prehistoric groups could be found occupying sites within the property. Major potential groups that could be found on the property include: Allegheny Valley Iroquois; Mead Island, McFate-Chautauqua culture and variants such as the Bell-Kalgren phase groups, Owasco, and Stewart Phase Shenks Ferry. Other lesser known and/or unidentified groups, phase, traditions, and complexes could also be present.

Late Woodland-Late Prehistoric-Protohistoric assemblages found at Indian Camp Run

At the Indian Camp Run site, two Late Woodland/Late Prehistoric groups have been identified including Chautauqua (French Creek Phase) and Mead Island cultures. Both groups fall within a series of Phases representing various time periods, some coeval. The last Prehistoric group (prior to European Contact) to occupy the Indian Camp Run site is represented by Chautauqua Cordmarked pottery. This group falls into Johnson’s (1994a, 1999a) Glaciated Allegheny Plateau (GAP) tradition. This tradition is comprised of three phases including from earliest to latest, Mahoning phase (ca. A.D. 1100-1300); French Creek phase (ca. A.D. 1275/1300 to 1400); and the McFate Phase (circa. A.D. 1400 to 1575). The Chautauqua Cordmarked pottery found at Indian Camp Run can likely be attributed to the so called French Creek phase. Although well represented in upland rockshelters and stockades in western Pennsylvania where it is often found in association with Chautauqua Cordmarked pottery, no McFate phase ceramics have been recovered from Indian Camp Run. In terms of the amount of material cultural and subsurface features, the most prolific group to occupy Indian Camp Run corresponds to Lantz’s (1982) Mead Island tradition. The Mead Island tradition lasted from approximately A.D. 970-1300.

The following are Late Woodland groups identified from Indian Camp Run (36FO65):

The Chautauqua Cordmarked Component

Chautauaqua Cordmarked is a distinctive shell tempered ceramic ware which has been found in some abundance at Indian Camp Run. According to Johnson (from Johnson and Myers 2004) toward the end of the Mahoning phase, approximately A.D. 1250-1275 in calendar years, igneous rock as a tempering agent in the preceding Mahoning ware, was gradually supplanted by pulverized mussel shell. This new shell tempered ware is essentially the Chautauqua Cordmarked type.

The Chautauqua ware found at Indian Camp Run likely falls into Johnson’s (from Johnson and Myers 2004) French Creek phase. One 14C sample was extracted from Feature 22, a large hearth, which was found to have been associated with portions of a shell tempered Chautauqua Cordmarked vessel. Dates associated with the particular assay are as follows: The conventional (un calibrated ) radiocarbon age was dated at 500+/ 50 years B.P. (Before Present); the 2 sigma calibrated results were A.D. 1325 to 1345 (Cal BP 625 to 605) and A.D. 1395 to 1460 (Cal BP 555 to 490; the 1 Sigma calibrated result was A.D. 1410 to 1440 (Cal BP 540 to 510); and the intercept of radiocarbon age with a calibration curve was A.D. 1425 (Cal BP 525). Within those ranges the feature and ceramics can be attributed to the French Creek phase.

Feature 48 Indian Camp Run-structure as evidenced by row of post molds (left center) of picture. Hearth feature at top of photo contained quantities of Chautauqua Cordmarked pottery and carbonized nut hulls

Chautuaqua Cordmarked was first described by (Guthe 1958) in his monograph entitled, the Late Prehistoric Occupation of Southwestern New York. The pottery is found at a number of sites including the type station Westfield and Burning Spring located in Chautauqua and Cattaraugus Counties in southwestern New York (Guthe 1958). Chautauqua ware was described as a shell tempered variety which was often accompanied by a grit tempered counter-part typed as Westfield Cordmarked. Schock (1974) reported on a number of sites in southwestern New York in which Chautauqua Cordmarked was found to be the dominant ceramic type. The results of this study were reported in his Phd. dissertation entitled, the Chautauqua Phase and other Late Woodland Sites in Southwestern New York. In this work, Schock (1974), coined the term Chautauqua Phase which was said to represent the latest prehistoric occupation in southwestern New York. In the early stages of the Chautauqua Phase, cordmarked pottery predominates before being gradually replaced by simple stamped wares. Johnson (1999:3) believes this change occurred at some time during the late sixteenth century. Simple stamped Chautauqua phase ceramics are currently unknown in the upper reaches of the Clarion River drainage.

According to (Johnson 1993:3) Chautauqua ware first appears in the French Creek and upper Beaver drainages ca. A.D. 1200 and would later become the dominant ceramic variety produced on the glaciated Allegheny Plateau during the middle of the Late Woodland period before being rapidly replaced by McFate Incised.

According to Johnson (1999), the distribution of Chautauqua Phase ceramics is centered on the glaciated portion of the Allegheny Plateau in northwestern Pennsylvania and adjacent southwestern New York. The ceramic type is commonly found throughout the Allegheny River, Clarion River and Tionesta Creek basins and has been identified from numerous rock shelters including Dutch Hill Rockshelter (36JE132)(Myers 2001),upland open air campsites, and stockades such as the McKinley Earthwork (36EL17) (Smith and Herbstritt 1976: 29). Chautauqua Cordmarked ceramics would continue to be produced up until the mid to late sixteenth century when the glaciated Allegheny Plateau became abandoned (Johnson 1999:3).

Chautauqua Cordmarked is often found on proto-Erie sites on the Lake Erie Plain and in the Upper Allegheny River valley (Johnson 1999:2) and on sites that produce Neutral-Wenro pottery types Guthe (1958). In many early reports the type is often described as being related to Monongahela Cordmarked and typically confused as one and the same. The grit tempered variety, Westfield Cordmarked, is thought to be somewhat related to Mahoning Cordmarked and Ripley Cordmarked (Guthe 1958).

An example of an upland stockade used by the McFate-Chautauqua tradition. Remnants of the McKinley Earthwork (circa. 1992) can be seen in the center of the photograph

Further afield, Chautauqua Cordmarked ceramics are associated with moderate sized villages in the French Creek Valley including Wilson-Shutes (36CW5) and from three at the McFate site (36CW1) and a number of smaller hamlets, farmsteads, or specialized extraction camps in the Shenango River drainage and Pymatuning Marsh area (Johnson from Johnson and Myers 2004). Shell tempered Chautauqua ceramics also appear on proto-Iroquoian sites in the upper Allegheny River Valley, along the Lake Erie Plain from Conneaut Creek in Ashtabula County, Ohio, to Cattaraugus Creek in Erie County, New York. (see Brose et al 1978; Dragoo 1966, 1976, 1977). Johnson (from Johnson and Myers 2004) has also noted that Chautauqua ware has been located as far north as the Portageville site in the Genesee Valley of western New York (Barber 1965, Johnson 1975, MacNeish 1952) and from the Whittlesey tradition Lyman (33LA2) site on the lower Grand River in Lake County, Ohio (Murphy 1971).

The Mead Island Component

The Mead Island tradition was originally recognized and defined by Dr. Stanley Lantz formerly of the Carnegie Museum of Natural History and was initially known from six major sites recorded in Warren County including the type station on Mead Island, an island in the Allegheny River near Warren, PA.

Map of the Mead Island site (36WA111) showing house patterns inside of a stockade (from S.W. Lantz 2004)

Lantz (1989) has suggested that the heart of the Mead Island cultural sphere once included large portions of the central Allegheny River valley from Warren southward to below Tionesta and perhaps as far as Kittanning. The full extent of their territorial range is however still unknown.

Map detailing the locations of important Mead Island tradition sites in the Allegheny River basin. Note: Indian Camp Run (36FO65) is one of the sites listed (from Lantz 2004:15)

Temporally the Mead Island folk date to the Late Woodland/Late Prehistoric period dating between circa. A.D. 970 and A.D. 1300 or A.D. 1025 to A.D. 1300 in calendar years (See: Lantz 2004: 1, 3). Seven of the nine dates run from the Penelec site located near Starbrick (Warren, PA) fall within a calendar range of A.D. 1025 through A.D. 1300. Recently Lantz (2004) has suggested changes be made to the existing radiocarbon dates associated with Mead Island to reflect the above mentioned range. The old model had Mead Island groups occupying the region from circa A.D. 850 1200. Dates included from the Penelec Site (36WA152) which has assays ranging from A.D. 840 +/ 45 to A.D. 1130 +/ 60; Rieder (36WA103) A.D. 1020 +/ 70 to A.D. 1200 +/ 70 and the type site Mead Island A.D. 970 +/ 75 to A.D. 1110 +/ 60. According to Lantz (2004) there have been recent challenges to the accuracy of the Gakushuin Laboratory dates (Blakeslee 1994).

Mead Island ceramics are fairly distinctive and occur in a number of type varieties. According to Lantz (1989) these ceramics share a number of affinities with Ft. Ancient Baum and Whittlesey Foci ceramics as well as with Monongahela and Iroquois type varieties. It is common for Mead Island ceramics to exhibit a thickened rim strip, add on collars; decoration applied by the edge of a cord wrapped paddle is common as well as a series of undecorated wares. Grit tempered pottery is the most common however shell tempered pottery is commonly found. The examples of pottery shown below are most likely attributed to the Mead Island tradition.

Mead Island ceramic vessel under excavation at Indian Camp Run

Mead Island projectile types include both Levanna and Madison like. Diagnostic of their projectile forms is the very high degree of flint knapping skills.

According to Lantz (1989) their settlement pattern is characterized by the construction of stockaded villages in the river bottoms. Inside of these villages round houses were constructed as evidenced by post mold patterns. These houses averaged approximately 8m in diameter and included a rock covered storage pit. The round houses also typically exhibited a centrally located hearth around which other activities occurred. Rectangular subterranean “sweat lodges” are another type of structure that occurs within the stockaded village.

Mead Island folk were not just confined to the river and stream valleys. According to Johnson et al. (1979: 84) there distinctive ceramics can be found in at least a small number of rock shelter type sites found in the uplands. In Venango County there ceramics have been identified from the Boyer’s Run rock shelter (36VE6).

Other important Late Woodland/Late Prehistoric components potentially found on Collins Pine property or in nearby locations.

McFate Phase

The Late Woodland/Late Prehistoric McFate people likely represent an Iroquoian speaking population that inhabited western Pennsylvania, northeastern Ohio and southwestern New York. They can be identified by their distinctive pottery known as McFate Incised which is typically found in upland locales on the Allegheny Pleateu including many rockshelters. This type will almost certainly be found on Collins Pine property. It is often found in association with, and most likely related to, Chautauqua Cordmarked pottery hence the term McFate-Chautauqua culture, which is actually a tradition within the broader Eastern Woodlands Culture.

This group is the latest group in Johnson’s (1994a, 1999) Glaciated Allegheny Plateau (GAP) tradition. The McFate Phase dates to (circa. A.D. 1400 to 1550/1575).

McFate Incised ceramics found at the Dutch Hill rockshelter (36JE132)located in Jefferson County, PA

McFate ceramics share affinities with a number of related types that share a common temporal position and a nearby proximity. Types such as Monongahela Cordmarked and Incised, Reeve Opposed, Funk Incised, Shultz Incised, Lancaster Incised, Richmond Incised and the McFate-Kalgren-Quiggle-Bell ceramics. According to Johnson (1999:8) McFate Incised ceramics were modeled after medium to medium high collared Pound Blank prototypes which were described by MacNeish (1952) as Neutral-Wenro pottery types that appeared on proto-Erie sites along the Lake Erie plain in the latter half of the fourteenth century.

The pottery is generally characterized by the addition of shell tempering to the paste, although limestone, creek gravel, and crushed chert (Richmond Incised) are known. The vessels are manufactured by the modeling method where a base is formed and thinned. Additional clay slabs are added to thicken the walls which are then thinned by using a paddle and anvil. McFate vessels typically exhibit high bulbous collars attached to a vessel jar. These collars are typically decorated with rectilinear incising with a common form being zones of right and left oblique lines separated by a zone of horizontal lines. Some of the later vessels begin to exhibit punctates on the lips and along the base of the collar. Some vessel exhibit dentate incising. Decoration is typically applied over cordmarked surfaces and necks are often wiped smooth (for a complete description see Johnson 1994b).

McFate Incised was originally described by Mayer-Oakes (1955) in his regional study entitled “Prehistory of the Upper Ohio Valley; An Introductory Archaeological Study”. The ceramic type was named for the type station, the McFate site located in Crawford County. Mayer-Oakes (1955) considered McFate Incised to be a regional variant of Monyock ( short for Monongahela/Youghiogheny) ware and described the type as being found primarily in the upper Allegheny drainage at such sites as McFate and Westfield in Chautauqua County, New York. The major difference distinguishing between McFate Incised and Monongahela Incised was that McFate Incised exhibited incising over cordmarking while Monongahela Incised exhibited incising placed over a smooth surface (Mayer-Oakes 1955:204).

Map of the stockaded villages know as the McFate (36CW1) Site in Crawford County, PA, prepared by Harry Schoff (from Burkett and Cunningham 1997)

McFate Incised is commonly found at site locations throughout the many tributaries of the central and upper Allegheny River, along the Lake Erie plain, and eastward into the upper West Branch of the Susquehanna drainage. According to Johnson (1999:7) the distribution of McFate Incised is centered on the glaciated Allegheny Plateau and particularly in the French Creek Valley of Western Pennsylvania. It was in the French Creek valley and other locations throughout the glaciated portion of the Allegheny Plateau that the McFate people constructed year round permanent horticultural (Johnson et al 1979:88) villages. From these villages individuals would travel abroad conducting their daily and seasonal activity including hunting and gathering, and trade or commerce.

Allegheny Valley Iroquois

The Late Woodland Allegheny Valley Iroquois (Erie) occupied the upper Allegheny River region and are known largely from salvage excavation conducted by the Carnegie Museum of Natural History prior to the construction of the Kinzua Dam (Lantz 1989:1). The Allegheny Valley Iroquois occupied a region that is now unfortunately largely covered by the reservoir. They lived in the region for some 700 years beginning around A.D. 940 to around A.D. 1525 and likely evolved from a Middle Woodland base (Lantz 1989:2). Their village plan consisted of small circular villages which after A.D. 1100 were all surrounded by stockades (Lantz 1989:4). Houses were rectangular in shape and increased in size as time progressed (Lantz 1989:4). Allegheny Valley Iroquois ceramics in the early period are reminiscent of preceding Middle Woodland wares. Various ceramic types associated with the group showing strong relationships with the Ontario Iroquois Tradition include Ontario Oblique, Glen Meyer Necked, Levanna Cord on Cord, Jacks Reef Corded Collar, Owasco Corded varieties including Wickham Corded Punctate, Owasco Corded Horizontal and Princess Point (Lantz 1989:5). Chautauqua Cordmarked and Westfield types are found on all sites (Lantz 1989:5). Lantz (1989:5) also noted that by A.D. 1100 the Allegheny Valley Iroquois were extensively interacting with the Lake Plain Erie documented by the presence of Sceiford Plain.

There is a good potential to find site attributed to this group in Collins Pine holdings in and around the Kinzua Dam and on some of their more northerly holdings. A recent site located by Kelly Lounsberry of Bolivar, NY and located near Bullis Mills may be related to Allegheny Valley Iroquois. Another site recorded as Warrant Rockshelter (36FO106) (formerly known as Parrish Rockshelter Complex 36FO42) located near the village of Parrish in Forest County, PA produced ceramic remains from one or possibly two vessels attributed (Johnson n.d.) to the Allegheny Valley Iroquois.

Stewart Phase Shenks Ferry

Material culture attributed to the Late Woodland Stewart Complex of the Shenk’s Ferry culture will likely be found on Collins Pine holdings in the Susquehanna River basin. In the Ohio River basin their remains are found on occasion.

Currently only two sites are known by this author where Stewart Complex remains have been recovered west of the West Branch of the Susquehanna Valley. Both are temporary rock shelter type sites located in extreme rugged upland locations overlooking the Clarion River valley in Elk County, PA. These sites were most likely located near trail systems that linked the upper Ohio Valley and the West Branch of the Susquehanna. Since portions of Collins Pine property are found in areas located near the West Branch of the Susquehanna or importantly near trails that link adjacent drainages such as those linking the upper Allegheny and the West Branch of the Susquehanna, a section regarding the Stewart Complex of Shenks Ferry was thought to be viable as almost certainly other sites belonging to this group will be discovered.

Stewart Complex pottery found near Johnsonburg, PA

The term Shenks Ferry as related to a particular prehistoric group or groups of people was coined after excavations were conducted in 1930 by Donald Cadzow near the deserted village of Shenks Ferry in Lancaster County, Pennsylvania (Graybill, Heisey and Kinsey 1971). This site would become the type station for material culture that would be recognized as belonging to a Shenks Ferry culture or tradition. A related but more northerly and westerly group would be identified as the Stewart Complex. The ceramics found in Elk County, Pennsylvania are related to the westerly Stewart Complex.

As a group, the Stewart people existed during the later portion of the Late Woodland/Late Prehistoric Period, from around A. D. 1300 to perhaps as late as A.D. 1520 in certain parts of the Susquehanna valley. Hatch (1981: 10) has proposed, based on the excavation of the stratified ceramic deposits found at the Fisher Farm site in Centre County, that the Stewart Phase existed at least in that area from earlier than A.D. 1300 to as late and possible later than A.D. 1520. Herbstritt and Kent (1990) have offered a different viewpoint noting that around A.D. 1425 to 1450 the Stewart Complex people began to closely interact with the McFate people of western Pennsylvania to form the McFate-Kalgren Phase. This development would mark an end to the classic Stewart Phase.

The ceramics were described by (Witthoft 1971:468) as “being round-based with channeled collars, crushed rock tempered, cordmarked, and of somewhat gritty paste”. The major differences between the ceramics found in the lower Susquehanna Valley and those of the West Branch include design, in which he noted that all of the West Branch rim shapes and motifs could be duplicated in the Shenk’s Ferry series. Hatch (1983) perhaps summed up the situation best by describing the Stewart Complex as a set of attributes that metrically distinguish these ceramics from those of the lower Susquehanna valley. Also, Kent (1984: 126) has noted that Stewart Complex pottery typically exhibits smaller collars than the southern counterpart. Generally speaking Stewart Complex ceramics were somewhat more conservative in nature than other Shenk’s Ferry pottery types. Temper preferences were also noted from region to region including differences between the Stewart site sample and Shenk’s Ferry ceramics of the lower valley.

Later, Heisey (1971) would apparently place the Stewart Complex into the Blue Rock Phase, the earliest of three phases proposed regarding Shenk’s Ferry ceramics. The Lancaster and Funk Phases were said to post date the Blue Rock phase. Graybill (1989) would place the Stewart Complex into what he tentatively described as the West Branch Tradition which included the Clemson Island complex (Jones 1931) and the McFate-Quiggle horizon of (Smith 1984).

Recently Herbstritt and Kent (1986) have outlined developments of the Shenk’s Ferry culture. In the case of the Stewart Complex they suggest that at around A.D. 1300 Stewart Complex pottery began to move into the West Branch Valley. They note that currently there are two theories regarding this development:

1). The first theory finds the people of the West Branch valley borrowing ideas from the Blue Rock phase folk located in the lower Susquehanna Valley.

2). The second theory proposes an in situ development in which the Stewart complex “evolved out of a local antecedent culture identified as Clemson Island/ Owasco.

Eventually according to (Herbstritt and Kent 1986) Stewart Complex pottery, begins to show up in association with shell tempered pottery whose origins lie in the McFate-Chautauqua culture of northwestern Pennsylvania. They state that the result of this blending of Stewart Complex pottery with McFate-Chautauqua is yet another new ceramic phase known as the McFate-Kalgren Phase (also known as the Bell Phase) which began to move east by around A.D. 1425-1450. According to (Herbstritt and Kent 1990: 14) this period marked the end of the Stewart Complex as the newer ceramics began to be tempered with shell, exhibit high collars, and were decorated with oblique and horizontal incised lines.

Within close proximity of Collins Pine tracts are sites identified with having McFate-Quiggle like pottery which may be ancestral to Stewart Phase pottery having undergone a gradual transformation from McFate to McFate-Kalren-Quiggle-Bell Phase. (36CM21) Emporium, (36CM46) Wyside 1, (36CM47) Wyside 2, and (36CM48) Rammage Flats all are said to have McFate-Quiggle like pottery.

Owasco

The Late Woodland Owasco culture dates from around A.D. 1100-1300 and has for some time been thought to represent the early culture (MacNeish 1952) from which Iroquoian culture rose . This group was originally recognized by Dr. William Ritchie from excavations at the Sackett site at Canadaigua (Ritchie 1936a), Ontario County, and from excavations at Levanna, Cayuga County (Ritchie 1928), and Castle Creek, Broome County (Ritchie 1934), all located in south central New York state.

Ritchie assigned Owasco to three phases based on ceramic traits including the Carpenter Brook Phase (A.D. 1100); Middle Owasco (A.D. 1130-1190); and Castle Creek (A.D. 1300). Recent research has challenged the dates and validity of the concept of an Owasco culture (see: “The Death of Owasco” by Hart and Brumbach 2003). Regardless of what Owasco is, or is not, we still have diagnostic Owasco pottery to account for.

The major characteristics of the pottery was (Ritchie and Funk 1973) the use of grit tempering agents such as quartz, vessels with paddle and anvil construction (as opposed to coiling), corded exterior surfaces, decoration on the shoulder, rim, lip, and inner rim areas, using cord wrapped stick or paddle edge. Motifs consist largely of plats, herringbones, and lines arranged vertically, obliquely, or horizontally. Vessel are large with broad flat lips.

A growing number of sites attributed to this group have been found in the region. According to Matlack (1990) there is a significant Owasco occupation along the Bennetts Branch of the Sinnemahoning Creek near Penfield, PA. Matlack (1990) has suggested that Owasco groups, possibly originated in New York State and migrating south although they did not appear to venture south of the West Branch of the Susquehanna River valley.

This author is aware of a handful of sites where Owasco like ceramics have been recovered. In Elk County Owasco like ceramics have been recovered from rockshelter sites and on floodplains of the creek valleys.

Owasco like vessel recovered from a rockshelter burial found south of Ridgway, PA

One site is known as the Pinnacle Rock rockshelter (36EL49) and is located just south of Ridgway, PA in an area known locally as Island Run, a small west flowing tributary of the Clarion River. The site was excavated during the Spring of 1975 and reported on by Ken Burkett (see: K. Burkett: PA Arch. :1977, Vol. 47, No. 4). During the excavation one particular feature on interest was investigated. Excavators exposed a dark circular feature on the rock shelter floor that was surrounded by several small post molds. The feature was found to contain the burial of a flexed female skeleton that was subsequently examined by Dr. Don Dragoo of the Carnegie Museum. He concluded that the female was around 45-50 years in age and measured approximately 5 feet 2 inches tall. Her burial contained a number of internments including, a turtle shell cup, mussel shells and a deer scapula awl. Clutching in her hands positioned near her face was a nearly completely intact ceramic vessel.

A drawing of the vessel by Cheryl Burkett as seen on page 51 appears to be of a classic Castle Creek Phase Owasco vessel. The vessel according to Burkett (1977) was tempered with finely crushed quartz and exhibited a low collar. The motif etched into the neck consists of horizontal oblique lines in a herringbone like pattern. The body of the vessel was cordmarked and exhibited a globular form similar to classic Stewart Phase pottery. It is apparent that a great deal of care was taken to bury this person in what is even by today’s standard a very remote location. Since no known village sites are located in even close proximity to this site it seems likely she died while traveling the various trail systems linking the drainages.

Another site where Owasco remains may have been recovered was at the Split Rockshelter (36EL4) located a short distance from Pinnacle Rockshelter. A collection owned by Mr. George Miller of Brockway, PA was viewed by the author on July 4, 2003 at the Brockway Independence festival. One of the collected samples was from the Split Rockshelter (36EL4). This rockshelter site was excavated by James T. Herbstritt and Dave Love in the early 1970's and reported on in Pennsylvania Archaeologist Vol. 45: No. 4 (pages 22-44). The George Miller collection contained several rim examples of what could possibly be Owasco pottery. According to James T. Herbstritt (per. comm. 2006) of the Pennsylvania Historical and Museum Commission there are other sites attributed to the Owasco culture in the Toby Creek valley near Brockway, PA.

One vessel recovered from Indian Camp Run appears to be an Owasco Corded Oblique vessel of the Carpenter Brook phase. Owasco is temporally similar in age to Mead Island so the possibility exists that the vessel from Indian Camp run is a copy of an Iroquoian vessel. The vessel is also limestone tempered which may represent a regional manifestation. In a recent conversation with Dr. Frank Vento of Clarion University the school has conducted test excavations at an Owasco site on the Allegheny River. In the upper Allegheny Valley (Dragoo 1976:85) Owasco ceramics were found at the Kinzua and Gould sites. Recently excavations (Burkett 1999) at Fishbasket (36AR134) located in the Redbank Creek drainage near New Bethlehem, PA have produced evidence of a longhouse which contained quantities of Owasco Corded Horizontal.

Contact Period (circa. A.D. 1635-1770)

Historic Contact period sites are rarely found on the interior drainages and mountainous “back water” regions of the Allegheny Plateau. Little archaeological field work has been done in these areas so what limited available data that is known concerns the larger river valleys such as the Allegheny, Genesee and Susquehanna Rivers. Some 32 sites were mentioned in the CRGIS files for the seven county study area. Many of these locations appeared to have been recorded based on “hear say” including historical narrative, rather than from hard evidence in the form of diagnostic artifacts recovered through archaeological testing. Indian Camp run is currently the only known site located on Collins Pine property to produce items from this enigmatic era. A discussion of the Contact Era was generated based largely on the interpretations of the artifacts recovered from Indian Camp Run.

Sometime at the end of the Protohistoric period circa. 1635 the Allegheny Plateau appears to become largely abandoned by native inhabitants (Johnson and Myers 2004). This marks the end of the prehistoric period. Material culture attributed to this period is virtually non existent with only a handful of sites being known. The Crowe-O’Conner site (36CW39) located in the French Creek drainage represents the final prehistoric Woodland site currently known in the Allegheny valley (Johnson and Myers 2004:104). This site is thought to date between A.D. 1585 to A.D. 1655 based on the presence of Acorn Ring Bowl pipes, a pipe type that has been documented on Proto-Seneca sites of that period (Johnson 2001c; Wray et al. 1991).

A number of events associated with the ensuing Historic period, influenced groups living in western and central Pennsylvania. According to Wallace (1991) It has been calculated that before the arrival of the Mayflower in 1620 not less than five hundred and possibly more than six hundred voyages had been made to New England alone.

The French, Dutch and English would establish settlements on the St. Lawrence and Hudson Rivers and on the coast of Massachusetts. Due to the demand for furs they were soon descending into the Great Lakes (Lantz 1982:102). Champlain would explore the Great Lakes as far west as Lake Huron in the early portion of the seventeenth century. In 1622 Father de la Roche Daillon traveled in New York and Pennsylvania (Schenk 1887:22). In 1668-1670 Rene Robert Cavalier de LaSalle reached the Ohio River region (Parkman 1926). In 1739 Baron de Longueuil by way of Lake Chautauqua and the Conewango Creek, entered the Allegheny River on his journey down the Ohio and Mississippi Rivers to join other forces from Michilimackinac, Detroit, the Illinois and Louisiana, for a campaign against the Chickasaw Indians in northern Mississippi (Kent 1991:3).

The English had established trade centers on the Ohio River. According to the Pennsylvania Provincial papers, XIII, 9-13, Croghan and others had stores on Lake Erie…up and down all that fine country watered by the branches of the Miamis, Scioto, and Muskingum Rivers, and upon the Ohio from Buckaloons an Indian town near its head to below the mouth of the Miami River, an extent of 500 miles on one of the most beautiful rivers in the world. The French did not want to lose control of their colonies in Illinois and Louisiana and saw the Ohio River as a natural road the reach these areas. Similarly the English colonies of Pennsylvania and Virginia saw the Ohio country as a logical area for expansion (Kent 1991).

This tension between groups would eventually lead to war. In 1749 Céloron de Blainville began an expedition to claim the Ohio River for France. He began at the rapids of the St. Lawrence above Montreal and eventually advanced to Lake Chautauqua and continued down the Conewango Creek to the Allegheny River at present day Warren. Opposite the mouth of Conewango Creek Céloron buried a lead plate claiming the Ohio River for France (Kent 1991:7). This even would mark the beginning of the dispute over the Ohio Country which would culminate with the Seven Years War a/k/a the French and Indian War.

In 1753 the French brought armed forces to Presque Isle and prepared to occupy the Allegheny River Valley (Wallace 1991:147). In 1754 General Washington surrendered Fort Necessity and Braddock’s defeat in 1755 caused concern the French army would soon invade settlements in the Susquehanna Valley (Wallace 1991:147). Following these events a number of Indian raids against frontier settlers began as the French and Indian policy was to force farmers from the land (Wallace 1991: 147-148). The war would drag on with a number of battles. On November 25, 1758 General John Forbes reached the Forks of the Ohio to find that Fort Duquesne had been burned the day before (Wallace 1991:150-151). The French would hold forts at Venango, LeBoeuf, and Presque Isle for another year until Sir William Johnson captured Fort Niagra which severed communication with the Ohio Valley at which time the French abandoned their remaining forts (Wallace: 1991:148).

With the English now in control of the Ohio Valley and French possessions in the Great Lakes the Indians again were concerned (Wallace 1991: 151). The English were less popular than the French with the various Indian groups in the region. The Indians would join forces and go to war in an event known as Pontiac’s War. Numerous battles would continue until in 1764 Bouquet took an army west into Indian country and in a treaty held at the Forks of the Muskingum, brought hostilities to an end (Wallace 1991: 153).

The end of Pontiac’s War would result in a number of refugee groups being forced west into the Allegheny River valley, being displaced by war and European expansion. Some of these events directly effected sites such as Indian Camp Run.

The Six Nations, by the Treaty of Stanwix in 1768 sold their Susquehanna lands to the English which meant that groups inhabiting the Susquehanna valley would be forced to leave, including those living at Wyalusing and Shamokin. The Moravian missionary David Zeisberger would follow and was able to convince the church fathers that the future of the missionary effort lay in the Ohio country where the main body of the “heathen” Delaware had been pushed following Pontiac’s War (Weslanger 1972: 284).

In 1767 Zeisberger established a small mission near the mouth of the Tionesta Creek in the midst of a number of mixed Munsie and Mingo villages known as Goshgoshing. Kent (1981) has identified the occupants as being Munsie Delaware and Fox Indians. These refugee villages were settled a few years before Zeisberger arrived in 1765. The Goshgoshing name was derived from Gosch-gosch “hog” with the locative signifying “place of hogs”. In all there were three Goschgoshing villages, one 2 miles above and the other 4 miles below the central village the mouth of the Tionesta Creek (Donehoo 1928).

Zeisberger was not well received by the many of the Indians when he first reached the Allegheny River. He was in conflict with the shaman Wangoman who continually attempted to control the populous do to their belief that he possessed witchcraft and that their lives were in his power (Wallace 1991:40). He was also not popular with the landowners and was eventually ordered to leave by the Seneca chiefs in Geneseo.

On April 7, 1769 the station at the mouth of the Tionesta was abandoned and along with the Christian Indians he moved 6 miles upriver to a mission in which he named Lawunahhannek or “middle stream place" (Donehoo 1928). Lawunahhannek was located near two other villages known both as Hickory Town. Lawunahhannek would exist for only a year and was eventually abandoned on April 17, 1770. Zeisberger became disenchanted with Allegheny valley when he learned that the land was not fertile and eventually packed up the Indians in 15 canoes and moved to the Beaver Creek.

The nearest of the villages to Indian Camp Run was the lower Goshgoshing village situated on an area known as Holeman’s flats. The name Indian Camp Run according to a local informant was said to be derived not from the village directly across the river but from a camp located somewhere in the headwaters of the run. Whether this information is true or folklore remains to be seen. Located a few miles down river from Indian Camp run in Venango County, PA was a village described by Kent (1981) as ?ussunnadohtaw (First Letter Missing) Old Town. The dates and the identity of the occupants are unknown.

It is possible that the French were responsible for trading the first contact trade items into the region as they had early outposts at the mouth of French Creek (present day Franklin) and at Buckaloons (Irvine). The British also had constructed an outpost at the Buckaloons which for a number of years was an important Seneca village as well as a refugee center for Delaware, Shawnee and other displaced groups (Lantz 1975). There were also a number of “refugee” camps located along the Allegheny River in the vicinity of the site. Therefore, it seems most likely that the beads were left on site by a Delaware, Mingo, Seneca, or other displaced inhabitant to the region during the French and Indian War era, circa. A.D. 1750. According to Lantz (per comm 2002), who has examined numerous beads from the Penelec (36WA152) site in Warren, PA, suggested that the smaller "sand" beads could have a mid eighteenth century origin. This would account for them not being mentioned by Kidd and Kidd (1970).

An interpretation of Historic-Contact era trade items found at Indian Camp Run:

Approximately one hundred items have been recovered from Indian Camp Run that can be positively identified to the Contact period. And, while it is hard to say when the first European trade items arrived at the site, and to which group they can be attributed, based on the characteristics inherent in the sample a relative age of between A.D. 1690-1750 has been suggested.

Examples of European glass trade beads

A number of small glass beads were recovered. A total of 56 blue, white, and black glass beads have been recovered to date. Artifacts of this nature have their origins in Europe and were traded largely with the various Iroquoian groups including the Seneca who laid claim to the Allegheny River. Just where the beads were manufactured is unknown, however some possibilities include the Netherlands, France, and possibly Italy. European manufactured beads were often produced by one country and then sold to another so the possibility exists that while the beads may have been manufactured in the Netherlands for example they may have reached the Iroquois via the French.

The Beads were classified according to Kidd and Kidd’s (1970) method of classification where each type is assigned a specific roman numeral and letter depending on size, shape, style and method of manufacture. Tubular beads of the IA, IIA and IIIA varieties Kidd and Kidd (1970) were identified in the Indian Camp Run assemblage. According to Kidd and Kidd (1970) tube beads were drawn beginning when a gather of molten glass was placed on the end of a blowpipe. A pocket of air was then blown into in the center of the molten glass. Another rod was placed at the opposite end and upon which both ends were drawn apart to form a long tube. They were then cut into various sizes and are usually very small. They were in effect mass produced.

The IIA variety white beads were the most common seemed to closely match the IIA11 through IIA15 varieties of small and very small monochrome, opaque and translucent “sand” beads. Due to the age of the beads it was not always possible to determine the exact color such as the difference between white and oyster white, hence the need to place a range on the beads i.e. IIA12-15. It must be noted that no exact match of Kidds bead descriptions could be found on their chart regarding the very small "sand" beads. They clearly do not describe circular shaped white “sand” beads that are as small as the smallest white beads found at Indian Camp Run. They mention circular beads that are small in size that range in diameter from 2-4 mm, but no very small circular white beads. The Indian Camp Run “sand” beads range in size from 0.82 mm to 1.88mm. The larger of the white beads in the Indian Camp Run sample conforming to the "small" variety ranged in size from 2.43mm to 2.59mm.

One small tubular monochrome blue bead possibly an opaque variety was recovered. This type may be a IIA37 or IIA41. This bead was found within a few cm of lead .56 caliber musket ball. The bead is of small size, circular in shape 3mm in diameter. Another variety recovered was one type IIIA “Straw” or tubular polychrome bead. It appeared to be a IIIA11 variety which is described as navy blue in color. This bead type exhibited a gray layer and a blue core and was ground on both ends. It measured 3.15 mm in width and was 8.11 mm in length. It was found in association with a small white IIA12 variety bead.

One small tubular monochrome black bead was recovered that seemed to match Kidd’s (1970) IA2 variety. This bead was small in size and translucent to light producing a somewhat dark red or “root beer” like color when held up to a light. It was found in the same context as a .45 caliber musket ball.

Catlinite bead (hairpipe) being recovered in situ

Another type of bead recovered during excavation is a long tubular form known as a hairpipe. Two fragments of a catlinite hairpipe were recovered during the 2002 field season. Hairpipes are long tubes placed in one’s hair for ornamental purposes. Catlinite is not native to the Eastern United States and is found in quarries in Minnesota and South Dakota (Kinsey and Custer 1982 PA Arch Vol. 52 No. 4-3). The hairpipe measures slightly more than 4 inches in length (109mm) and is rectangular in shape measuring 5.59 mm in width and 7.05 mm in height. The drilled orifice measures 2.25 mm in diameter. According to Turnbaugh (1979) items produced by Native Americans such as catlinite became fashionable during the 18th century as the period was marked by a revival of native made goods.

Musket balls and French and English gunflints

A number of items associated with early black powder guns have been recovered. Eight gunflints manufactured from lithic material originating in Europe of both French and English design were recovered at Indian Camp Run. Six are described below. These gunflints are prismatic in shape and vary in color from amber to a lighter grayish color. The amber colored flints are smaller in size measuring: 1). 20.76 by 18.55 by 6.32 mm in thickness; 2). 17.59mm by 19 mm by 6.01 mm in thickness; 3). 19.69 mm by 17.18 mm by 5.65 mm in thickness. The gray gunflints measure as follows: 1). 26.59 mm by 24.03 mm by 7.3 mm in thickness; 2). 25.88 mm by 23.29 mm by 8.59 mm in thickness; 3). 20.09mm by 17.89mm by 5.59mm in thickness. Lantz (1980) has noted that typically the amber colored flints are thought to be of French manufacture while the larger grey flints are British in origin.

Ten lead musket balls have been recovered. The majority measure to nearly .45 caliber (.4375) in diameter while others are larger, one measuring approximately .57 caliber (.5625) and another (.6125) or potentially a .62 caliber gun. Three lead balls are “mushroomed” indicating it has been fired. Typically a deer or other animal when shot was brought back to camp and skinned resulting in the pulling of the spent balls from the dead carcass. A firm date has not been established for the musket balls other than to suffice they are historic and are likely temporal with the other historic items. This first guns to be traded into the Ohio valley region were during the fur trade that began in the late sixteenth century. According to (Shumway 1985:11) various forms of flintlocks were quite common among Indian groups that occupied frontier regions in Pennsylvania, Maryland, Virginia and perhaps New York at the time of the French and Indian War.

Kaolin pipe fragments

Seven fragments of a Kaolin pipe have been recovered. Five of the fragments were pipe stem components while two fragments were of the bowl. The bowl of the pipe exhibits a heel a/k/a spur. No maker marks are visible on the pipe that could potentially aid in the dating of such a piece although likely manufactured in Europe. Pipes of this nature Kent (1984) are commonly found on sites dating to the 17th and 18 th centuries in eastern Pennsylvania and have been reported by Zakucia (1960) from Kuskuskies (36LR11) in Lawrence County, PA. As stated above Dr. Stanley Lantz has suggested a possible late 17th century date for the pipe.

Historic pipe with T.D. makers mark attributed to England circa 1750 A.D.

Four fragments including a large portion of the bowl and various portions of the stem were recovered from a clay pipe. On the back of the bowl (stem side) were the initials T.D. At the base of the bowl was a spur. The bowl diameter measures around 1 inch in size. One measurement taken of the bore of the pipe stem was 2.4 mm. Only a small portion of the stem has been recovered and this segment measures 71.6mm. T.D. makers marks are attributed to Thomas Dormer (1748-70) or Thomas Dormer & Son (1754-1756) of London (Oswald 1975: 67-68, 135). Mr. Dormer was the only individual with T.D. initials that exported his pipes to America and the West Indies (see Oswald (1975, Appendix C, page 135). The likelihood that this is one of his pipes seems pretty good.

Two brass buttons were recovered. These were found in direct association with three gunflints and the hair pipe fragments. These buttons are flat and round with an eye molded to the center of the back. There appears to be a greenish metal alloy that has eroded from the surface of the button which is reddish in color. Buttons of this type fall into the Type 7 category detailed by the North Carolina Department of Archives and History from their chart entitled “Analysis of the Buttons From the Ruins at Brunswick Town and Fort Fisher, North Carolina 1726-1865". The buttons fit into the first half of the time period between 1726 and 1776 and correspond to time frame established by the other historic artifacts found at Indian Camp Run.

Brass bead

One brass bead was recovered from Indian Camp Run. The example pictured above is nearly identical to those pictured by (Wray et al. 1987: 49, 51, 59, 195) from the Adams and Culbertson sites and from the Tram and Cameron sites (Wray et al. 1991: 72, 246) all located in Livingston County, NY. These sites fall into the eastern and western Seneca site sequence between A.D. 1560-1600 (see Wray et al. 1991:4).

Several of the items from Indian Camp Run where viewed by Steven Warfel, Assistant Curator of the Pennsylvania Historical and Museum Commission in Harrisburg, PA.

Mr. Warfel has made the following assessment of the site sample:

"The "hair pipe" is nearly identical to ones found at the early 18th century Conestoga Indian Town site (36La52) in Lancaster County. Most of ours measure 4.5-4.75 inches long [see Figure 38 in Barry Kent's book, Susquehanna's Indians (1984), item at top] Likewise, the illustrated glass bead types, including the so-called "sand beads," are prevalent at Conestoga Indian Town. Barry Kent dates the Conestoga Town site, excavated by the Museum Commission, to the period 1690-1740".

Based on the findings of the research it would seem plausible that the Indian Camp Run Historic/Contact era sample fits nicely into the relative age of 1690-1740.

Other Contact period sites could possibly be found on Collins Pine property. The Forbidden Path passes very near to Collins Pine property in Potter and McKean Counties. This trail was possibly used by Champlain’s emissary, Etienne Brulé in 1615 (Wallace 1987:47). The trail was definitely used by David Zeisberger in 1767 on his way to the Delaware villages on the Allegheny River in defiance of a Seneca chief (Wallace 1987: 47). In 1779 portion of the trail was used by General John Sullivan when he burned Seneca villages (Wallace 1987:48) during the Clinton-Sullivan campaign.

MODEL FOR THE LOCATING OF VARIOUS SITE TYPES

The Database

A database of 873 sites located in 7 counties including, Cameron, Elk, Forest, McKean, Potter, Venango and Warren, was analyzed to produce the model. Other existing models produced for the region were also used for comparative purposes.

Site locational information was extracted primarily from three sources including published site reports found in various texts, a data base of archaeological site locations maintained by the author, and those found in the Pennsylvania Cultural Resources Graphic Information System (CRGIS). The CRGIS data base is a computerized reproduction of the PASS (Pennsylvania Archeological Site Survey) form which has been used for many years to inventory sites in the State. There are limitations when using the CRGIS system. The system is still in its infancy and needs a little fine tuning. The author began by examining all of the sites that he has recorded into the PASS files and subsequently the CRGIS. Many inaccuracies were found. Some were placed for example in the wrong township or municipality and some were placed on the wrong topographic map. Others were said to exhibit cultural components that were not located by the author, although these may be updates added by other researchers. Some sites described as open air campsites were said to be rockshelters and vise versa. Certain sites were not found in the system at all and are presumed to have been placed in the wrong county. If there were discrepancies found with those sites that I am familiar with, then there must be numerous errors in the data base as a whole.

One of the biggest problems with the CRGIS is that the PASS forms are seldom filled out in there entirety and subsequently contain biases. In many instances these forms contain what could be best described as bits and pieces of information and in some cases no information at all. Many sites were recorded with no description of topographic setting(s) while others were recorded with no mention of the artifact types found or cultural affiliation. Since many PASS forms were filled out in the 1960’s and 1970’s, they contain outdated information especially regarding cultural components. Also, when filling out a site form, what one person feels to be important, may not be to another. For example the most sensitive site location in the world is a floodplain or terrace located at the confluence of drainage. Many of the PASS forms make note of a major confluence (i.e. Allegheny River and Tionesta Creek) but do not make mention of a small run and the major river or a spring that empties into the major river. With that in mind the distance to the nearest confluence may be inaccurately described on the PASS form, the site in fact may be located near the spring and the major river (still a confluence). This type of information could lead a person studying site data to assume that the site was much further from a confluence than is actually the case. Also, many individuals note the location of the nearest water from a map instead of searching for the nearest water location on the ground. Many recorded rockshleters only note the nearest stream or river and not the nearest springhead or water source. It is these types of biases that are difficult to sort through when determining landform probability. In fact the category describing the nearest water source was dropped from the study due to extreme biases in the sample.

The results of findings:

Found within the seven county data base was a total of 873 sites. The majority of the sites were found in the Allegheny River basin which totaled 782 sites. 89 sites were located in the West Branch of the Susquehanna drainage while only 2 sites were found in the Genesee drainage.

These totals reflect biases as the majority of the archaeological field work has been conducted in the Allegheny River basin. In areas such as Potter County little archaeological field work has ever been conducted resulting in only a handful of sites ever being recorded. No rockshelters have ever been recorded in Potter County.

The county with the most recorded sites was Warren (325), followed by Venango (242), Elk (121), Cameron (62), Forest with (55), McKean (28), and Potter with (26) recorded sites.

Sites were compared by type, such as open air campsites vs closed or protected rockshelter type sites. A total of 859 sites produced data in this category. There were 537 (63% of all sites) open air campsites and 322 rockshelters (37% of all site types) in the site sample. Specific site types such as springhead campsites and burials were included as open air sites. The large number of rockshelters sites found in the data base reflects the fact that many cultural resource studies have been conducted in the uplands in areas conducive to the formation of such site types. It must be noted that in many instances the PASS files did not indicate the type of site, so the total in this category is somewhat less than the total site database examined.

A comparison of rockshelters to open air campsites was calculated by county. Open air site types were somewhat prevalent accounting for some 63% (no.= 537) while rockshelters produced some 37% of the site totals with 322 sites of that nature.

Cameron County has 4 rockshelters and 58 open air type sites; Elk County has 54 rockshelter and 67 open air type sites; Forest County has 28 rockshelter and 27 open air type sites; McKean County has 17 rockshelters and 11 open air type sites; Potter has 0 recorded rockshelters and 26 open air type sites; Venango has 105 rockshelters and 137 open air site types; Warren County has 114 rockshelter and 211 open air site types.

General Statements Regarding Site Locations:

There are a number of universal criteria that effect site locating. One important consideration is the slope of a particular land form. According to Stewart and Kratzer (1989) a low relief landform of 15% or less slope is a basic characteristic of site location. Hughes and Weissman’s (1982:86) reanalysis of several upland studies indicates that slopes of 8% or less are frequently the rule. Many sites have multiple aspects and some are even flat but as a general rule of thumb slope less than 15% will produce sites. This basic statistic is in regard to open air type sites as rockshelter type sites will occur in areas of steeper slope although many of these occur on benches and the floors are typically fairly flat.

Sites are also commonly found on well drained landforms. Cowin (1981, 1991) found during research conducted in west central Pennsylvania that 91% of the sites she examined were found on well-drained to moderately well-drained soils. Some terraces located along river bottoms that look good on a map, turn out to have poorly drained soils with swampy conditions. In some instances, seemingly good terraces and/or floodplains when field inspected, will be found to be completely scoured by the river resulting in deflation of the site and the soils. Occasionally a landform that was considered a prime camping spot in prehistoric times may no longer be desirable and conversely a spot deemed as good by today’s standards may have been wet during ancient times.

The aspect (direction a landform faces) of a particular landform in prehistoric times does not seem to be as important a criteria as the location of landform that holds the site. Prehistoric people did not seek out southeast facing landforms to construct sites. This is in direct contrast to later Historic period sites which tend to be found with southern facing aspects (Miller 1986). Many sites exhibit multiple aspects while many sites located on the floodplains and terraces of the various drainages exhibit no aspect as they are flat.

Sites are occasionally located with a windbreak that shelters the site. Similar to the placement of historic farmsteads in a depression with hills surrounding the house, prehistoric sites also took advantage of natural windbreaks. The Crane Run open site which is possibly a burial location was located by the author during a cultural resource survey while employed with the Ridgway Ranger District of the U.S.D.A. Forest Service. The site was positioned on a narrow plateau rim at the 1780’ elevation with a slight 4% east southeast aspect. A rise in the plateau directly behind the site blocked the westerly wind from campsite. This observation was noted by the author while conducting test excavations at the site.

Water is typically an important variable when looking for the location of an archaeological site. The waterways were the roads and provided sources of drinking and bathing water. Food resources whether they were big game or gamebirds such as ducks and geese, as well as plant resources always thrived in the fertile river and stream valleys. Stream junctions, springheads, single and converging drainage heads (active and inactive) and upland bogs, swamps, or poorly drained habitats show a high correlation with archaeological sites (Gardner 1978: 20-21, 31-32; 1983: 21-22; Gardner and Boyer 1978; Wall 1981; Lantz et al. 1982:97; Lantz 1986:11; Hughes and Weissman 1982:90a; Stewart 1983:53, 68-69; Tolley 1983:109, 113). Water will most always be located a short distance from a site of any importance. As stated above, the information gathered in the CRGIS files was examined but not used in this study because it was simply inaccurate. Many people filling out the PASS site forms were placing the nearest river or stream as the closest water source and that is simply not factual information and it would have tremendously biased statistics. In a recent study of all of the rockshelters recorded from the PASS files in western Pennsylvania (Tippins 2007) noted a similar problem in that “the distance to water can be a highly subjective measurement, as some recorders considered seasonal springs or seeps to be the nearest water source, while others appeared to discount these water sources and reported the distance to the nearest stream”. Funk (1989:87) has noted that “it is likely many…sites are near springs either not observed by archaeologists or extinct since prehistoric times. Rockshelter sites are typically further from water than sites in other settings. Tippins (2007) noted that the average reported distance of water to a rockshelter was some 143 meters. This author believes that in most instances a source will exist within a short walk from the site. In the case of the Dutch Hill rockshelter a perennial spring was located some 30’ from the site area which could easily fill pots for drinking and boiling purposes. The Elk County Earthworks appear to be further from water than most sites although there are springs near the Kane and Russell City earthworks within 500’ to 800’ and an extinct drainage corridor has been observed by the author some 300’ from the McKinley Earthwork.

In higher order drainages, within the valley and floodplain, sites can occur on the general course of a stream with no other apparent features deciding the location (Stewart and Kratzer 1989). Stewart (1983:68-69) also notes that in many of these situations, springheads, tributary streams or floodplain marshes are usually within 1500 feet of a site. Lantz (1982) has also noted that fishing stations occur on narrow terraces…adjacent to rapids or eddies. In low order stream and run valleys with relatively recent floodplains, prehistoric sites can occur near the toe of slope at the valleys edge on colluvial landforms (Wall 1981).

Sites in upland setting occur on saddles between drainages, on hilltops with well drained soils and highly defensive positions, near springheads, and at locations with prominent views. Many of these types of sites will be located along a trail. If one follows a low order drainage to the headwaters eventually a spring will be reached and at the same time the traveler will have used the path of least resistance to reach the crest of a hill or ridge. In northwestern and north-central Pennsylvania portage routes will likely have sites on the saddles such as those found near Keating Summit located on the portage between the upper Allegheny River and the West Branch of the Susquehanna. Also, the hilltops surrounding the higher order drainages have a good chance to produce site locations due to the close proximity to areas of intensive use. This may or may not be the case in the back water regions.

Hilltop in background was used as a temporary campsite

An examination of sites by particular landform or setting was initiated as they were recorded from the PASS form. A total of 236 sites were found on floodplains including 46 in Cameron, 20 in Elk, 14 in Forest, 7 in McKean, 17 in Potter, 40 in Venango, and 92 in Warren County. There were 150 sites recorded on terraces including 3 in Cameron, 8 in Elk, 10 in Forest, 3 in McKean, 4 in Potter, 31 in Venango, and 91 in Warren County. There were 115 recorded sites on hillslopes with Cameron reporting 6, Elk 27, Forest 6, McKean 14, 0 in Potter, 19 in Venango and 43 in Warren County. There were 6 sites located on saddles including 1 in Cameron, 3 in Elk, 0 in Forest, 1 in McKean, 0 in Potter, 1 in Venango, and 0 in Warren County. There were 60 sites located on stream benches including 0 in Cameron, 12 in Elk, 1 in Forest, 2 in McKean, 1 in Potter, 31 in Venango, and 13 in Warren County.

122 sites were recorded from the upper slopes including 0 in Cameron, 13 in Elk, 17 in Forest, 2 in McKean, 2 in Potter, 51 in Venango, and 37 in Warren County. The middle slopes produced 68 sites including 0 in Cameron, 13 in Elk, 3 in Forest, 9 in McKean, 0 in Potter, 22 in Venango, and 21 in Warren County. The lower slopes produced 42 sites including 0 in Cameron, 5 in Elk, 1 in Forest, 0 in McKean, 0 in Potter, 33 in Venango, and 3 in Warren County. Upland flats produced 9 sites including 0 in Cameron, 2 in Elk, 0 in Forest, 0 in McKean, 1 in Pooter, 6 in Venango, and 0 in Warren County. Hilltops/ridgetops produced 42 sites including 0 in Cameron 16 in Elk, 3 in Forest, 1 in McKean, 1 in Potter, 4 in Venango, and 17 in Warren.

The data set presented above indicates that the majority of the sites some 52% were located on a terrace, floodplain or stream bench located along a drainage course. Many of the sites found on the upper slopes, middle slopes, and lower slopes are rockshelter type sites. Sites found on saddles, upland flats and hilltop/ridgetopes are often small open air campsites and occasionally will be stockaded forts.

Types of sites:

Villages

Prehistoric villages are found on the floodplains and terraces of the major drainages. These types of sites are typically large in size, requiring many acres of space, and were occupied on a somewhat permanent basis by a large population. These “main” site locations were the epi-centers where all other activities originated as many functions occurred in and around these sites. Village sites were often the location of ceremonial activity. Often times people were buried inside the limits of the village or in an ossuary nearby. Many villages in Late Woodland/Late Prehistoric times were nucleated, surrounded by stockades in which numerous houses were placed. This is direct contrast to a settlement pattern consisting of small hamlets or dispersed cabins/farmsteads that may be located for miles along the floodplain of a river.

The majority of the village sites will be located in areas where a wide variety of resources can be exploited such as fish, fowl, large and small game, and a sundry of plant and animal resources. The larger river and stream valleys are best suited to this description. The large terraces in and around the Penelec Site (36WA152) and the Buckaloons sites for example could have supported many acres of corn which could serve to feed large villages and attract other food resources such deer and smaller animals.

One important factor in the locating of a major horticultural village was growing season. As many as 130 frost free days are required to support a large horticultural village. In our study area the majority of the land area has much less than 130 frost free days. It was through the domestication of certain plant types such as corn, beans, and squash (the three sisters) that led to population increase and overall stability of the group. The largest population centers had an agricultural based society. People tended to the crops rather than spending time constantly hunting and gathering to supply the village. Hunting and gathering was always an important function but was used more as a supplement to the main crop rather than the focus of the group. The entire Allegheny Plateau is located in an area that is only marginally suited for the growing of 8 and 10 row flint maize. As many as 120 days between killing frosts are required to support a viable agricultural economy (Yarnel 1964) although Johnson (per. comm. 1994) believes that actual number is closer to 130 frost free days.

Because village sites were few and far between, this has led early white European settlers to believe that the Indians did not live in the region. Village sites tend to be located around the periphery of the Allegheny Plateau especially in areas such as the Glaciated portion of the Allegheny Plateau and in particular the French Creek Valley, Lake Erie Plain and upper Allegheny River valley in southwestern New York. River valleys such as the Allegheny River or West Branch of the Susquehanna have longer growing season due to the ameliorating effects of morning fogs which occur in late summer and into the fall. Based of data compiled by Dr. William C. Johnson which recorded the mean frost free day growing season at certain location between 1951-1980 the following data was generated. Clarion 130.3 days; East Brady 154.7; Ridgway 126.6; Tionesta 147; Brookville 130.6; Kane 95.1; Bradford averages to 107; Franklin 156; Warren 144.7; Coudersport 102.1. Erie on the Lake Plain had 174.4. These statistic don’t take into account the fluctuation in temperature for example between 1961-1970 Ridgway experienced 6 of the 9 years with less than 130 frost free days which generally means that aboriginal crops would have failed six times in nine years. A group could not manage a village with this type of failure rate for the staple crop.

In summation no viable horticultural villages occur where the growing season is less than 130 days. Major village sites will be located along the larger river valleys with ample space and an adequate growing season.

Campsites

Associated with the large villages were a number of satellite type sites including open air camps and rockshelters. These were typically occupied for short terms and were seasonal in use. These camps were typically associated with hunting, gathering, and fishing activity and trade and commerce. The Indian Camp Run site is a typical open air campsite where small groups occupied the terrace conducting the activities of processing game, gathering and storing nuts, berries, and corn. Fishing and fowling were most likely important tasks to the individuals that occupied the site as well. In Late Woodland times both Mead Island and later French Creek phase individuals constructed small houses on the site while earlier individuals camped out under the stars. Indian Camp Run was located on a major trail with traffic generally flowing along the river. A ford known historically as Holeman’s Ferry crossed the river at or near the site.

Many open air camps are associated with trails and typically occur near a river crossing local or on a hilltop near a saddle. Campsites such as fishing stations are often associated with natural features such as rapids or near eddies (Lantz 1982).

If we examine the attributes of the Indian Camp Run site we find that the site is located in a prime strategic location for exploiting many resources. Indian Camp Run enters the Allegheny River directly across from the upper end of Holman’s Flats, a large stretch of bottom lands that extends some 2.5 miles along the north bank of the river. Holman’s Flats undoubtedly offered abundant supplies of floral and faunal resources that could have been exploited by numerous groups over time. If conditions were unfavorable for a river crossing to Holman’s Flats, the terrace at Indian Camp Run would have been a logical place for mobile bands to camp until conditions improved. Just upriver from Indian Camp Run and below the confluence of the Tionesta Creek, the Allegheny River passes through a constricted corridor where a number of islands have formed clogging the river channel. Steep cliffs occur on both sides of the river and very little floodplain/terrace development exists in this area, making traversing of the valley floor difficult by foot. Foot travelers, both man and beast, would have been forced to ascend into the uplands above the river or hop from island to island to continue along the river. This portion of the Allegheny River Valley was known in historic times to have posed potential problems for travelers. The Venango-Conewango trail which led from present day Franklin to Warren passed near the site. General William Irvine traveled the path while exploring the Donation Lands and describes the journey as follows: “From Oil Creek to Cuskakushing (Tionesta), an old Indian town, is about seventeen miles, the whole of the way is barren, high mountains, not fit for cultivation; the mountain presses so close to the River that it is almost impassable, and by no means impracticable [practicable] when the river is high, then travelers either on foot or horseback are obliged to ascend the mountain and proceed along the summit” (Wallace 1987:175). For game species crossing the constricted river, hopping from island to island (when possible), the arrival on Holeman’s Flats might be their first opportunity for a well need rest and grazing. Interestingly Holeman’s ferry was located at or very near to the Indian Camp Run site location, so the river must have been easy to cross at this location.

Important landforms associated with camps will typically have a source of potable water. Many small campsites will be found on flats around spring heads. These are the sources for the best drinking water and a typically considered high probability locales. The author has recorded a number of sites located at a springhead including Little Arnot (36WA312), Laurel Mill (36EL89), and Tionesta Spring (36EL94). In every instance these sites were positioned on flats near a perennial or ancient spring head. Little Arnot and Tionesta Spring were located in stream valleys at the base of hillslope a short distance from a larger stream. Laurel Mill was located on a hillslope at the head of small, now extinct run. The Little Arnot site was unique in that the wooded area had never been cultivated. A circular ring of rocks was located near the spring which produced chipping debris inside. This fire ring is likely a prehistoric feature that had survived for hundreds of years in the remote wooded location.

Little Arnot Run open air campsite located on flats around spring. The spring was likely dug out and ponded in Historic times

In summation numerous campsites will be located in the river and stream valleys on well drained landforms such as terraces, floodplains, and stream benches. These typically occur at confluences of streams, rivers and runs. They will be found to be in association with trail systems. They will be located near springheads and on saddles which were occupied while crossing ridges (trails).

Rockshelters

Rockshelters will be found in abundance on Collins Pine property. These sites were used as temporary and seasonal domiciles and offered the wayfarer a certain amount of protection from the elements. Utilized rockshelters occur as large multi-overhang complexes to lone single exposures. Many rockshelters were located along trail networks and in some instances (e.g. Minister Valley Trail) the location of the rockshelter(s) likely dictated the location of the trail itself. There is no good way to predict where a rock outcrop will occur. Some common locations of exposed bedrock occur at the point of a ridge where erosional processes for millions of years have exposed the bedrock. Rock outcrops also occur on the sideslopes of most drainages throughout the region. As a general rule of thumb the best rockshelters with intact living floors will occur in areas of moderate slope as opposed to steep hillslopes. Crown crests occur on hilltops in some areas such as in the Clarion River drainage. The fact that rockshelters are found in headwaters of various drainages makes them ideal domiciles for travelers.

Rockshelter on hill above the East Hickory Creek in Forest County, PA

It has been demonstrated that trails (Lantz 1969; Myers 1997) often used drainage corridors as the paths of least resistance to climb up and out of the river and stream valleys. The streams and runs themselves were followed which could gradually elevate the traveler from valley floor to ridgetop. Here in many instances bedrock is exposed and rockshelters occur. In the case of the trail through Minister Valley rock walls are found formed along most tributaries of the run and adjacent drainage ways. Numerous stopping points were available for the traveler.

Examples of trail shelters:

Bogus Run Rockshelter (36FO56) complex, located near Sheffield Junction a few miles west of Russell City, PA., is an example of a rockshelter ideally situated to receive traffic via a long system of trail corridors (i.e. drainage corridors). The site is positioned in the headwaters of the Tionesta Creek on a saddle that separates the Clarion River watershed by some 2200’. The alignment of the drainage corridors in this location provides the traveler with many miles of straight line walking before reaching the drainage divide that houses the site. Bogus Run, a small tributary of the South Branch of Tionesta Creek, could then be followed which would serve as a gradual incline out of the valley floor to the rockshelter complex positioned at the 1860’ elevation. Following a nights rest or perhaps a few days stop for hunting and gathering, the saddle of the ridge could be then crossed and the journey could continue into the Clarion River drainage.

Another site located in a similar situation is the Watson Branch Rockshelter (36FO50). This site is located a few miles west of the Bogus Run Rockshelter near Pigeon, PA. This site is also positioned in a direct alignment with north to south flowing drainages. In this case the Bluejay Creek located in the Tionesta drainage to the north and the Spring Creek drainage located in the Clarion River system could be traveled. Once again the site could be reached by following a trail out of the valley floor to the rockshelter waiting on the saddle. The journey could then continue into the adjacent drainage with very little effort.

Dutch Hill rockshelter located on a hillside bench overlooking the Clarion River

Dutch Hill Rockshelter (36JE132) is located in a slightly different situation. This site is positioned above a river road, in this case the Clarion River. The site is located on a hillside bench at the 1640’ elevation some 400 feet above the valley floor of the Clarion River (Myers 2001). The Clarion River gently meaders in it’s course some 100 miles from headwaters to confluence with the Allegheny river. Numerous sites such as Dutch Hill Rockshelter would have become known to the regions inhabitants and visited while passing through the region, likely canoeing or traversing the valley floor below.

Elevational Data

Rockshelter sites found throughout the study area were found to range in elevation from as low as 920’ in Venango county to as high as 2360’ in McKean County. The average was found to be 1640’ while the mode or most common elevation was at the 1800’ elevation. These elevations can used as a guide in determining possible elevations of exposed bedrock by county.

Cameron
Range=1600’ to 2260’
Median=1930’
Mode=Multimodal 1600’, 1890’, 2170’, 2260’

Elk
Range=1280’ to 2200’
Median=1740’
Mode=2100’

Forest
Range=1500’ to 1860’
Median=1680’
Mode=1700’

McKean
Range=1680’ to 2360’
Median=2020’
Mode=2020’

Potter
No Rockshelters

Venango
Range=920’ to 1600’
Median=1260’
Mode=1240’

Warren
Range=1200’ to 2020’
Median=1740’
Mode=1800’

Rockshelter Orientation (Aspect)

Theoretical archaeology takes into account variables in human activity that should produce the best results when selecting a site or rockshelter. The so called best rockshelter would be said to be oriented southeast to receive early morning sunlight and warmth. And, while any nice rockshelter would be used regardless of orientation, data gathered from 138 rockshelters by the author suggest the following regarding aspect.

Direction
North 4
Northeast 13
Northwest 9
South 18
Southeast 28
Southwest 24
South-southeast 1
East 19
East-northeast 1
East-southeast 5
West 12
West-southwest 3
West-northwest 1
Total 138

Sizes of rockshelter type sites varies greatly and nearly anything that could offer shelter has the possibility of being utilized. According to Tippins (2007) the average size of a rockshelter in western Pennsylvania is 9.8 meters in length by 3.6 meters in width or 35.3 m². Tippins (2007) notes that the largest rockshelter in the region is (36VE46) Little Scrubgrass III rockshelter at 460.6m² while the smallest is (36EL45) 1.8m² the Heigle rockshelter. While the roof elevations vary greatly from side to side and front to back at a rockshelter, Tippins (2007) noted that the average roof height was 3.0 meters while the highest ceiling was recorded as 22.7 meters and the minimum 0.9 meters.

Based on an examination of the topographic maps detailing the various parcels owned by the Collins Pine Corporation there will be a number of prehistoric rockshelter sites located within those bounds.

Upland Stockaded Forts:

There are a small number of stockaded forts located throughout the region. These sites are in effect small upland villages most likely used for hunting and gathering purposes and later for staging areas for groups as they abandoned the Allegheny Plateau. Interestingly vestiges of the stockades that surrounded these sites can still be seen in areas that have never been plowed. The stockades were likely constructed for protection from the weather and possibly from hostile groups. The sites are located specifically on drainage divides on long lobate hilltops high above the valley floor. These lobate hilltops are defensible on three sides and have shallow but well drained soils.

ISmith site located near Bolivar, New York. Note: earthen stockade wall can be seen running from left to center of the photo

The stockades occur in close proximity to major trail networks. In the case of the Elk County Earthworks they are located approximately two miles from the Iroquois Main Road. The Hickory Kingdom forts are most likely located just off of the Kersey Road and another possible trail that followed close to present day Route 255 into the headwaters of the Bennett’s Branch of the Sinnemahoning. The Smith site located near Bolivar, New York is located in close proximity to the Oswayo and Forbidden trails and possibly another trail that followed along the Little Genesee Creek. They appear to have been positioned close enough to the trails to reach in matter of minutes yet far enough to prevent surprises by unfriendly groups.

These sites typically occur in clusters such as the Elk County Earthworks which has four known stockades and a number of auxiliary sites close by. At Hickory Kingdom located near DuBois, two stockades are known. Why they would occur in multiple numbers remains a mystery. It is possible that the sites were used during a period of migration. During the early 16th century the Allegheny Plateau was abandoned by native inhabitants and these sites would have been logical stopover locales as the gradual migration occurred.

Burials

In the seven county study area thirty-four burial sites have been identified. Twenty two of these are located in Warren County in the upper Allegheny River basin from Buckaloons northward to the New York border. Nine of the mounds alone are found at Buckaloons which is a wide floodplain terrace located around the confluence of the Brokenstraw Creek and Allegheny River. This area appears to have been a large ceremonial center attributed to the Middle Woodland period and associated with the internment of the dead. These burial mounds often contained multiple individuals some of whom were cremated. It was not uncommon for later groups such as those from the Contact and Historic periods to bury individuals in the mounds. In the case of the Willow Island mound in West Virginia a large farmhorse of possible 19th century age was buried in the mound (Hemmings 1978). Some of the mound groups may be associated with a village such as the Irvine Mounds. Lantz (1989) has noted that Middle Woodland Raccoon Notched projectiles have been recovered from nine sites and that (36WA99) located on the floodplain of the Allegheny River is the most extensive and prolific. In the case of the Sugar Run mound, a recent reevaluation (McConaughy and Johnson 2003) of Carpenter’s original assessment of village and mound, suggests that evidence is lacking that the mound is in fact associated with a Middle Woodland village. There are a few stonepile burials located in upland locations. One stonepile located by the author near Highland Corners in Elk County is of unknown age. A stonepile located near Wetmore in McKean County produced Flint Ridge chalcedony from the surrounds of the pile (Dean n.d). Chalcedony at a site is often suggestive of a Middle Woodland age.

Tionesta Mound located near Ludlow, PA-upland stonepile and possible burial. Flint Ridge lithics were reported by Robert Dean as having been found near the mound

Burials have been found in rockshelters. At the Pinnicle Rockshelter (see Burkett 1977) located just south of Ridgway in Elk County the body of a Late Woodland age woman was recovered. She died while traveling abroad and was buried at the rockshelter far away from a home village.

Burial location

Upland burials (unspecified stone piles or earthen mounds??)
Total=2

Upland stonepiles
Total= 2

Floodplains, terraces, stream benches
Total=28

Rockshelter burials
Total= 2

Grand total= 34 sites

The affiliations of the burial sites include: 11 unknown, 1 Terminal Archaic, 5 Early Woodland, 15 Middle Woodland, 3 Late Woodland, and 1 Woodland unspecified. One burial mound exhibited a count for Early, Middle and Late Woodland periods. No Paleoindian or Archaic burials are known for the study area.

Quarries

These are sites associated with the gathering of raw materials typically used in toolmaking endeavors. Very few of these sites are known from the study area. One such site Campbell Quarry (36JE47) located south of Brookville in Jefferson County has produced workable chert that can be found eroding from the banks of the small runs. South of Warren along the Allegheny River is a hill known as "paint hill". At this location Native Americans were said to gather red iron ore pigment for use as face paint.

Rock Art Sites

Sites containing petroglyphs, a/k/a rock art or carvings are rarely found in the study area. Rock art sites typically occur from Venango County southward and are common in southwestern Pennsylvania. One petroglyph in Clarion County located near the Cooksburg bridge is controversial and may in fact be of historic origin.

TRAILS

A study of the various trails systems of western and central Pennsylvania was initiated. These trails occur both in and around Collins Pines Holdings. These routes were important in prehistoric times to journey to far away lands and to conduct seasonal and cyclical hunting and gathering forays nearer to home. Many were later used by early settlers as they entered the region. Like the highways of today, these routes would have been associated with a number of stopping points or temporary campsites. Where you have trails you will have sites.

The majority of what is known about trails in Pennsylvania was researched and written by the late Paul A.W. Wallace. His book entitled “Indian Paths of Pennsylvania” first published in 1965 is still one of the best selling books at the State Museum in Harrisburg. In this work he describes the locations of the trail often with some historic narrative regarding its use. Mr. Wallace also offered routes for driver wishing to travel as close to the actual route the Native Americans used as possible.

Locally many trails have been researched by the author. Results of this study were originally presented at a symposium sponsored by the Carnegie Museum of Natural History in Pittsburgh in 1994. This symposium organized by Richard L. George, Staff Archaeologist with the Carnegie Museum, focused on current research into the McFate Culture of northwestern Pennsylvania. The paper presented by this author entitled “An Examination of Late Prehistoric McFate Trail Locations” was subsequently published in the journal Pennsylvania Archaeologist (see: Volume 67(1): 45-53).

In this study proposed trails once used by the McFate culture were defined based on the location of temporary site types such as rockshelters and small open air campsites. It should be noted that the majority of these trails were used well before the Late Prehistoric period McFate (circa. A.D. 1450-1550) and some likely served as entrance ways for some of the regions earliest inhabitants.

When examining the trails discussed by Wallace or by the author is must be noted that the exact location of the trails will never be known unless specific locations are mentioned in historic text. There are no longer any clues to indicate there presence, such as wagon ruts, or blazes between trees. The routes of the various trails were located based on the location of known sites, a common sense approach to examining the data, and at times conjecture. It must be noted that the location of the Catawba Path as placed on the master set of Collins maps is purely conjecture as Wallace offers very little clues as to the exact location. Due to the fact that the location of this path is poorly known, the values of high, medium, and low probability were not adjusted to reflect this trail passing through particular parcels. The Catawba trail was said to enter Pennsylvania some 6 miles east of Bradford and follow a course southwest to Tallyho. It would have seemed more logical that instead of climbing the steep hills east of Bradford that a course along the valley floor along the the Tunungwant Creek and its tributaries (see Trail 5 Tunungwant-Clarion River portage detailed below) could be followed. This would be a path of little resistance. At Tallyho the trail left the Kinzua Creek drainage by climbing the ridge and then continued on its southwesterly course past Kane and into the Clarion River drainage. Some have speculated it then followed closely to SR 66 before turning south. Unfortunately we will never know the exact course.

While the locations of most prehistoric trails can only be surmised, it must be noted that portions of some historic trails are still visible. One example would be Forbes Road built on the earlier Raystown Path as a military road in 1758 (Wallace 1987: 198). Visible portions of this road/trail have been mapped and photographed near the crossing point along the Allegheny Front near Rhor’s Gap in Allegheny Township, Somerset County (see: Waddell and Bomberger 1996:46, 49). This trail was also used by early white settlers entering the Ohio country. The possibility does exist that a remnant of an old trail or road could be found in an old growth forest such as Hearts Content where some hemlock are said to be 400 and beech 350 years old www.dcnr.state.pa.us/forestry/oldgrowth/heartscontent.aspx. Similarly in some of the more remote locations of Collins Pine lands, perhaps clues as to early road could be found.

Methodology to Examine Potential Trails

In order to observe potential travel routes used by Native Americans a data base consisting of rock shelters, open air campsites, and upland stockaded forts was compiled by examining the site files found at the Allegheny National Forest and the Carnegie Museum of Natural History. A sample of sites, located in portions of Cameron, Clearfield, Elk, Forest, Jefferson, McKean and Warren Counties, were plotted on 1/24,000, 1/50,000 and 1/250,000 topographic maps. This permitted local and regional analysis of site locations and their relationship to nearby topography. Trail systems were then formulated by examining the following criteria: the density of sites found in a particular drainage corridor and the direction in which the drainage aligns, the location of a particular site and its relationship to an important topographic feature, and the occurrence of long parallel drainage corridors linked by sites located usually in the headwaters. Often times the newly proposed trails directly linked with previously recorded late prehistoric and historic trails.

Trail Observations

1) The trail segments were often linked by a series of key site locations which occurred near important topographic features such as saddles and stream and river confluences. Upon examining the data base, temporary site types such as rock shelters and open air campsites, were often found either in the heads of the low order drainages or near river and stream confluences. Lantz (1969) noticed that rock shelter sites were commonly located in the headwaters of low order drainages. The low order drainage corridors were used as paths of least resistance leading up out of the valleys to the plateau tops and from there these trails crossed the many ridges located throughout the Allegheny Plateau. Many long trail segments were linked by sites located on saddles between adjacent corridors. It was proposed that if a rock shelter or other site type occurred on a drainage divide at the head of two twenty-mile drainages then the trail system could extend at least forty miles. River and stream confluences were also popular locations for temporary site types. At these locations the trails would split often requiring a river crossing. Both the headwaters locales and the river and stream confluences may have marked the beginning and end points of a daily journey.

2) Many major trails followed navigable water routes such as the Clarion River and West Branch of the Susquehanna. In the region separating the upper Ohio River and the Susquehanna River many long travel corridors exist that are separated only by narrow ridges. The navigable drainages flow in opposite directions; the Ohio River drains to the southwest and the Susquehanna River drains to the southeast. To travel through this region one of the river systems would have to be walked while the other could be traveled by some form of canoe. Hasentab (1987) has proposed that many sites were located on portage routes between adjacent navigable waterways. A traveler may have walked into the headwaters of one drainage and then crossed a divide into another. After traveling downstream a certain distance, a port of entry may have been reached and the journey could then proceed by water.

3) The appeared to be an increase of traffic using the trails in the Late Prehistoric period. This can be viewed by looking at various rock shelter sites containing McFate Phase cultural material. In many instances a seemingly poor rock shelter with limited space and a lack of protection from wind and rain had been occupied. These site types were often chosen in spite of being located within a short walk of a large rock shelter or rock shelter complex (i.e., >3 overhangs). It would be logical to assume that the protective and spacious rock shelters would be chosen first over all others if found to be vacant. Groups frequenting the trails during the late prehistoric period must have been large enough to necessitate use of all available rock shelters in a given vicinity. Similarly, size and/or quality of a particular rock shelter did not appear to be indicative of a particular individual’s status. Identical cultural material can be found at both the large and small rock shelters as if site selection was made only on a first come first occupy basis.

4) Several of the stockaded forts that were occupied by McFate groups were located within a short distance of historically recorded paths. It would seem likely that the locations of these particular McFate sites were selected due to their close proximity to a particular trail system. In order to further examine routes outside of the data base sample, previously recorded historic trails located throughout the McFate sphere were investigated. It was surmised that these routes could have also been used by the McFate. George (1974) noted the strong correlation between Historic period Indian paths and the location of Monongahela culture sites in southwestern Pennsylvania. The logical assumption was that the historic trails presumably had prehistoric antecedents (Johnson et al. 1989:3). For example, the McFate site in Crawford County is located near a cross roads of several historically known paths including the Venango Path which linked Erie with Pittsburgh and the Kuskusky Path which ran from Meadville to New Castle. The Quiggle site in Clinton County is located near the Great Shamokin Path which ran east to west linking Sunbury to Kittanning and the Pine Creek Path which linked the Genesee River to the West Branch of the Susquehanna near Jersey Shore. Similarly, the Catawba trail or Iroquois Main Road that linked southern New York with southwestern Pennsylvania and points further south is located only two miles east of the Elk County Earthworks.

Proposed Trail Systems

Modern trail leading to remote Elk county, PA rockshelter. This trail follows the path of least resistance to the site. Could this modern trail have prehistoric antecedents?

Trail System Linking Buckaloons to the West Branch of the Susquehanna (Trail 1)

Trail system one originated at Buckaloons on the Allegheny River west of Warren, Pennsylvania and traveled into the Tionesta Creek and Clarion River drainages and may have ultimately reached the West Branch of the Susquehanna River (see fig. 2). This trail may have tied in with the Presque Isle Portage trail (Wallace 1987:140) that journeyed from Erie south to Fort Le Boeuf (Waterford) located on French Creek and the Brokenstraw Trail that followed along the Brokenstraw Creek to Buckaloons (Wallace 1987:25-26). Another possible route leading to Buckaloons began around the western side of Chautauqua Lake and likely journeyed south along the Little Brokenstraw Creek to Pittsfield before reaching the Brokenstraw Trail. Regardless of the route chosen to reach Buckaloons a rather prolific trail corridor seems to begin just to the south across the Allegheny River and follows the ridge toward the Tionesta Creek. This system seems to directly align with traffic using the Brokenstraw drainage. After following the ridge toward the Tionesta Creek a crossroads of sorts was reached that separated the Tionesta drainage from the Allegheny River. The density of sites including McFate occupied sites became high in that area. This particular trail continued south toward the Tionesta Valley and seemed to expand in width as five north to south corridors received usage. Traffic followed the Minister Creek, which is an identified trail (see: Minister Creek Trail-Lantz 1969), Fools Creek, Upper Sheriff Run, Lower Sheriff Run and Bobbs Creek south into main stem of the Tionesta Creek. The trail then turned east and journeyed to Lynch where the Bluejay Creek enters the Tionesta Creek. The Bluejay Creek was then followed to its headwaters near Pigeon located on a saddle associated with numerous sites that linked the Tionesta Creek and the Clarion River. To enter the Clarion River, Spring Creek was followed to the main stem of the Clarion River at Hallton. From Hallton, the trail followed the Clarion River to the vicinity of Portland Mills, a location containing several route options which ultimately reach the Susquehanna drainage. The Little Toby Creek, Belmouth Run, Laurel Run and Dog Hollow Run all received heavy McFate movement. By following the Little Toby Creek south, the Chinklacamoose Trail or Great Shamokin Path (Wallace 1987:61,174) could be reached. By following Belmouth, Laurel and Dog Hollow Runs, located in the vicinity of numerous McFate sites including the Split (36EL4) Rock shelter (see: Herbstritt and Love 1975), the trail could reach the base of Boones Mountain. This separates the Clarion drainage from the West Branch of the Susquehanna. From there it joined with the Kersey Road. From the History of Elk County, “ in the year 1810, William Kersey, land agent for the Fox and Norris Company, Massachusetts traveled the Indian Trail across Boones Mountain to erect a gristmill” (Wessman, 1981:303). According to (Wallace 1987:78) this route linked Luthersburg to the head of Elk Creek near Kersey in Elk County. It was apparently an offshoot of the Great Shamokin Path which at approximately 15 miles west of Clearfield journeyed north, crossed over Boones Mountain, and Helen Mills before it turned northeast.

Chautauqua Portage Variant Linking the Upper Allegheny to the West Branch of the Susquehanna (Trail 2)

Trail system two may have been an extension of the Chautauqua Portage trail (Wallace 1987:136) which was a part of a well known travel route that existed between the St. Lawrence and the Ohio River systems. The Chautauqua Portage trail ran nine miles from Barcelona Harbor to Mayville on Chautauqua Lake. From Mayville the trail continued around the lake shore reaching the town of Celeron. It was near this location that the Conewango Creek was then followed south to the Allegheny River (Ohio River) at Warren. This route was used by the French including, De Longueuil in 1739 and Celoron de Blainville in 1749 (Wallace 1987:136,137). At Warren instead of following the Allegheny River into the Ohio country another route may have crossed the Allegheny River near the mouth of Dutchman Run and entered the ancestral Tionesta Creek corridor. Southward, the trail continued through Sheffield and on to Brookston where a trail split occurred. One route followed the south branch of the Tionesta Creek to the Elk County Earthworks, a cluster of four known stockades, constructed by the McFate. The other route followed Bogus Run to its headwaters and reached the Bogus Run (36FO56) Rock shelter, located on a narrow saddle between the Tionesta Creek and the Clarion River. Both routes continued south towards the Clarion River. From the Elk County earthworks the Clarion River was reached via Bear Creek placing the traveler near Portland Mills. From Bogus Run, Spring Creek was likely followed once again reaching the Clarion River at Hallton. The Clarion River was then followed to its mouth via the Little Toby Creek. Following a similar scenario as trail one, this route crossed Boone's Mountain divide into the Susquehanna drainage (see fig. 2).

Kinzua Creek-Clarion River Portage (Trail 3)

This trail began in the upper Allegheny River and entered the Clarion River near the vicinity of Kane. It began along the Allegheny River which it followed south to the mouth of the Kinzua Creek. It continued south along the Kinzua Creek following a course described by (Wallace 1987:129) as Pigeon Path. This particular trail followed the south branch of the Kinzua Creek toward its headwaters at Kane. From there it crossed the divide and entered the Clarion River drainage near the Seneca spring in East Kane (see fig. 3).

Tionesta Creek (Trail 4)

This trail may have been a short cut from the upper Allegheny River to Franklin. Portions of the trail were described by (Wallace 1987:129) as pigeon path used by Seneca hunting parties while gathering pigeons. It was also popular with the McFate and began in the upper Allegheny River and journeyed south until reaching the Kinzua Creek. Instead of following the Kinzua Creek drainage to the divide at Kane as did trail 3 this route turned west at Dunkles Corners and climbed over Gibbs Hill before entering the Tionesta drainage near Ludlow. At Ludlow the trail continued southwest following along the East Branch of the Tionesta Creek toward Donaldson. From there the main stem of the Tionesta Creek was followed to Tionesta where it joined with the Allegheny River (see fig. 4).

Tunungwant-Clarion River Portage (Trail 5)

This trail may have linked the Niagra Frontier and points in Cattaraugus County, New York with the Clarion River and West Branch of the Susquehanna River. The northern most portion of this trail was known as the Cattaraugus Path which ran from the Buffalo Creek through Cattaraugus to Salamanca where it made connections with several trails including one that journeyed west to Warren (Wallace 1987:33). The proposed trail begins near Riverside Junction south of Salamanca at the mouth of Tunungwant Creek. It continued south to Bradford where a trail split occurred. One route followed the Kendall Creek into its headwaters and passed over the ridge into the Potato Creek of the upper Allegheny River. It continued south eventually climbing over a ridge and into the Driftwood Branch of the Sinnemahoning Creek en route to Emporium. The other route continued its southerly course along the Tunungwant Creek eventually reaching the headwaters. From there the divide was crossed separating the Tunungwant from the Kinzua Creeks near Tally Ho. The path then entered the Clarion River drainage near the vicinity of Lantz’s Corners and Mt. Jewett. From there the Clarion River could be traveled south (see fig. 3).

Possible Variant of the Goschgoschink Path (Trail 6)

This trail may have been a variant of the Goschgoschink Path which according to (Wallace 1987:61) ran from West Hickory, through Luthersburg to Clearfield. The proposed route may have been a short cut from the Allegheny River at West Hickory to Brookville and points beyond. This route left West Hickory and crossed the Allegheny River near Endeavor. From there Beaver Run was traveled into the headwaters. The trail climbed over the divide at Whig Hill, entered the Tionesta drainage, and proceeded south toward Kelletville. It then followed the Salmon Creek south to its headwaters located near Marienville and entered the Clarion River drainage. The trail proceeded south following such drainages as the West Branch of the Millstone Creek and Maple Run to reach the Clarion River in the vicinity of Clarington. After crossing the Clarion River the path then proceeded over a divide and into the Redbank Creek drainage. It could then continue south to Sigel and join the Catawba Path or rejoin the Goschgoschink Path at Brookville. Another version of this trail may have began at Tionesta and followed the Coon Creek southeast to Vowinkel and crossed the divide into the Maple Creek before reaching the Clarion River near Clarington (see fig.3).

The East Hickory Trail (Trail 7)

This trail linked the Allegheny River at East Hickory in Forest County with the Allegheny River east of Warren and may have been a shortcut or bypass of Buckaloons and Conewangotown. The trail followed the East Hickory Creek into its headwaters and crossed a divide before entering into the West Branch of the Tionesta Creek (see fig 4). The trail then likely joined the main stem of the Allegheny River near Warren. This trail could also have joined a path called the Browns Run/Kinzua Trail which followed Browns Run east to its headwaters before it traversed the ridge to Jakes Rock (Lantz 1969:1-5).

The Minister Trail (Trail 8)

The Minister Run trail was first reported to Stan Lantz by Paul Yeagle of Warren, Pennsylvania. Portions of this trail are intertwined with the proposed Trail System 1 and Trail System 7. According to Lantz (1982), the Minister trail begins at the Tionesta Creek and journeyed north along the Minister Creek into the headwaters. From there it climbed north over the drainage divide and entered the Farnsworth Branch and West Branch of the Tionesta Creek.

Cornplanter-Venango Trail Variant (Trail 9

At the town of Tidioute in Warren County a trail system linked the Allegheny River with Titusville on Oil Creek and may have been a variant of the Cornplanter/Venango Trail that linked the upper Allegheny River with the mouth of French Creek near Franklin in Venango County (see Wallace 1987:41). The proposed route left Tidioute and followed the Tidioute Creek and other nearby drainages west and entered into tributaries of Oil Creek. The route extended to Titusville and points to the south (see fig. 3).

The Clarion River Road (Trail 10)

The Clarion River road was well known to McFate groups. Numerous sites occur sporadically throughout the valley from its headwaters in McKean County to its mouth located near Foxburg in Clarion County. A long direct east to west running route existed that could link the West Branch of the Susquehanna River with the Allegheny River via the Clarion River corridor. The trail may have started in Emporium and followed West Creek into its headwaters located near St. Marys. From there it may have entered Elk Creek where the trail became navigable. Elk Creek enters the Clarion River at Ridgway. From here the journey west could continue to the Allegheny River and points along the main stem of the Ohio River (see fig. 4).

Sinnemahoning Path (Trail 11)

This trail system linked the Upper Allegheny River, Genesee River and points in the Susquehanna drainage and is actually part of three known trails. The most northerly route had possible connections north of Olean in Cattaraugus and Allegany Counties, New York. At Olean the Ichsua Path journeyed south to Port Allegheny, Pennsylvania. At Port Allegheny the trail was then called the Sinnemahoning Path which entered the Susquehanna drainage after crossing the ridge at Keating Summit. It proceeded east along the West Branch of the Susquehanna River before ultimately reaching the Great Island at Lock Haven (Wallace 1987:76,155). Early settlers used portions of this route to reach the Driftwood Branch and Bennett’s Branch of the Sinnemahoning Creek (see fig. 3). They followed a route similar to the Forbidden Path (Wallace 1987:46) which journeyed west from Tioga in Bradford County, Pennsylvania, continued across the Genesse River and reached the Allegheny River south of Portville, New York. A description of this route appeared on a handbill printed in 1811 in Burlington, New Jersey advertising 140,000 acres for sale in M’Kean and Clearfield Counties (Hughes 1981:7-9).

Trail Leading up Prather Run (Trail 12)

Both Beers History of the Counties of McKean, Elk and Forest, PA (1890 reprint), and the History of Forest County 1867-1967, by Childs (1989), allude to a trail that followed along Prather Run near East Hickory. According to Beers History, “On the path leading up Prather’s run from the Allegheny River….” This trail apparently exited the valley by following Prather Run into the headwaters and then proceeded in an easterly course toward the upper forks of Ross Run. This trail may have lead near an Indian Camp located on the hill above Ross Run (Childs 1989: 28). A small drainage known as Camp Run that enters Beaver Run to the north may signify the general location of the Indian Camp. The trail appears to have crossed Fork Run which is located just east of Whig Hill. From Fork Run the path appears to have entered the Tionesta Valley. It is possible that this route is a variant of the Goschgoschink Path. Another alternative route of the Goschgoschink path was proposed by (Myers 1997) and described above.

Trail leading from East Hickory to Tionesta Creek via Jug Handle Run (Trail 13)

According to Beers History (1890: 831) from a paper written by Daniel Harrington an “Indian trail or footpath was used by the first white settlers in crossing the river, at the mouth of East Hickory to the Tionesta Creek striking the latter stream (Tionesta Creek) about ten miles from its mouth”. He further states that “the path was used more by Indians than by whites” This trail could possibly have used Prather Run or another stream to access the hilltop, before turning south towards Jug Handle Run. Harrington continues…“A little over two miles from Tionesta and directly on this path, there is a, large strong spring of pure water, known as Belisle spring. It took it’s name from a man who had a hunter’s camp at this spot. The spring is the source of Jug Handle Run” The mouth of Jug Handle Run is located approximately ten river miles east of Tionesta and a possible location of where the trail entered the Tionesta Creek. It would seem that the Belisle spring is located more than two miles from Tionesta however.

The Oswayo Path (Trail 14)

This trail linked Genesee to Shinglehouse (Wallace 1987:121). According to W.W. Thompson (Legend of White Lilly (Rose) Lake” Historical Sketches of Potter County, Pennsylvania (Coudersport 1927) an Indian path ran from Genesee, PA up Rose Lake Run to Rose Lake and down the Oswayo Creek past the towns of Oswayo, Coneville, and Millport to Shinglehouse (Wallace 1987: 121).

The Goschgoschink Path (Trail 15)

This trail linked West Hickory on the Allegheny River with Clearfield (Chinklacamoose) on the West Branch of the Susquehanna. According to (Wallace 1987: 61) at the Delaware Town of Goschgoschink (West Hickory) the Goschgoschink Path forded the Allegheny River. It ran south crossing the Tionesta Creek near Nebraska. Continuing south by way of Tylersburg and Helen Furnace, it joined the Venango-Chinklacamoose Path before continuing east to Luthersburg. At Luthersburg it joined the Great Shamokin Path and followed it to Chinklacamoose (Wallace 1987: 61).

Cornplanter’s Path (Trail 16)

Cornplanter’s path ran from Cornplanter’s Town (Jenuchshadego or Burnt House) on the Upper Allegheny to the mouth of the Conewango Creek at present day Warren, PA (Wallace 1987:39). The path was said to have used multiple routs. Various drainages on the Allegheny River were used such as Cornplanters’ Run and Hodge Run to reach Quaker Ridge. A fter crossing Quaker Ridge, the trail entered the Conewango drainage by following Glade Run, Hatch Run, Akeley Run and via Indian Hollow (Wallace 1987: 39).

The Cornplanter-Venango Path (Trail 17)

This trail bypassed the Allegheny River and proceeded from Venango (Franklin) and journeyed to Conewango (Warren) (Wallace 1987:41). The trail passed through Titusville and entered the Brokenstraw drainage near Pittsfield. One variant is shown turning east at Titusville and crossing the ridge to Tidioute. From there it followed the Venango-Conewango Path (Wallace 1987: 41).

Venango-Conewango Path (Trail 18)

The Venango-Conewango trail led from present day Franklin to Warren and passed near Indian Camp Run. General William Irvine traveled the path while exploring the Donation Lands and describes the journey as follows: “From Oil Creek to Cuskakushing (Tionesta), an old Indian town, is about seventeen miles, the whole of the way is barren, high mountains, not fit for cultivation; the mountain presses so close to the River that it is almost impassable, and by no means impracticable [practicable] when the river is high, then travelers either on foot or horseback are obliged to ascend the mountain and proceed along the summit” (Wallace 1987:175).

Catawba Path (Trail 19)

The Catawba Path a/k/a the Iroquois Main Road linked Ichsua (Olean, NY) with the Carolina’s (Wallace 1987: 27). Locally the trail entered Pennsylvania a few miles east of Bradford and continued south crossing the Kinzua Creek at Tallyho. The trail then passed very near present day Kane, PA and Highland Corners before crossing the Clarion River near the mouth of the Millstone Creek. The trail continued along it’s southerly course passing through Sigel and Corsica before eventually leaving Pennsylvania south of Gans in Fayette County, PA.

Allegheny Portage-Big and Little Portage Paths (Trail 20)

These trails linked the upper Allegheny with the West branch of the Susquehanna River. The Big Portgage Path led from Emporium to Port Allegheny and points north. The trail followed Portgage Creek crossing the dived at Keating Summit and continued into the Allegheny Portage Creek (Wallace 1987: 135). The Little Portage Path Linked Costello with Port Alegheny and was said to be some five miles shorter than the Big Portage Path. The Little Portage Path ran along Freeman Run and climbed the hills to Keating Summit. At Keating Summit the trail joined the Big Portgage Path (Wallace 1987: 135).

Forbidden Path (Trail 21)

This path led from Athens on the North Branch of the Susquehanna River through Elmira on the Chemung River to Canisteo and continued to Shnglehouse and Olean (Wallace 1987: 46-47). The trail was said to be forbidden to white men due to security purposes of the Seneca (Wallace 1987: 46). David Zeisberger was the first white man to record passage over the trail (Wallace 1987: 47) which greatly angered the Seneca and would later cause Zeisberger to move his mission from the Allegheny River to Muskingum country in Ohio.

CONCLUSION

Predictive models aid land managers in determining areas that are more likely to produce cultural resources than less sensitive areas. Throughout the region there are various landform types that based on certain conditions would have been popular site locating areas for prehistoric people. While a predictive model cannot portend where each and every site will be located, certain basic principles regarding settlement pattern, when followed increase the chances for the locating of archaeological sites.

If a landform exhibits positive attributes and/or negative attributes a grade can be assigned such high, medium, or low sensitivity. These areas can then be marked onto maps showing proposed sensitivity levels. Areas could then be tested accordingly based on these sensitivity levels to prove or disprove the model. Certain areas including those labeled high probability could then be avoided during management activity.

Some of the high probability areas found on Collins Pine property include dry river and stream terraces at or near a confluence of two drainages. Areas such as saddles specifically located at the heads of drainage corridors, hillside benches, and well drained hilltops are common site locals. Due to the amount of property owned by Collins Pine that exhibits moderate sideslope, rockshelter sites will be a common site type found on the property. Since a large portion of Collins Pine property is located away from the major drainages such as the Allegheny River, the significance of the trails that once passed through the property will be extremely important to the locating of prehistoric sites, including rockshelters and other small temporary campsites. The majority of the village sites found in the region are located near the higher order drainages in areas with ample growing season for maize and other crops. In the uplands stockaded forts could be located especially on tracts of land located on drainage divides and near trails.

The Collins Pine property is large and numerous topographic situations exist along with multitudes of undiscovered sites. Based on the finding at Indian Camp Run (36FO65) alone every cultural period from Paleoindian to recent Historic is found on Collins Pine land. Currently Indian Camp Run is considered the oldest site on the entire Allegheny River. Other significant sites will be located.

This model like any other can only be validated through testing and inventorying of sites over time. It is strongly recommended that a program of implementing and testing of the model be introduced in an effort to determine what cultural resources exist on the property. If these non- renewable resources are lost the record of early human occupation of the region will be lost as well.

BIBLIOGRAPHY

Adovasio, J. M.
1993 The Ones that Will Not Go Away: A Biased View of Pre-Clovis Populations in the New World. In From Kostenki to Clovis, Upper Paleolithic-Paleo-Indian Adaptations, edited by Olga Soffer and N. D. Praslov. pp. 199-216. Plenium Press, New York.

Adovasio, James M., J. Donahue, J. Gunn, and R. Stuckenrath with assistance of J. Herbstritt and W.C. Johnson
1982 The Meadowcroft Rockshelter/Cross Creek archaeological project: Retrospect 1982. In Collected Papers on the Archaeology of Meadowcroft Rockshelter and the Cross Creek Drainage, edited by R.C. Carlisle and J.M. Adovasio. Department of Anthropology, University of Pittsburgh.

Adovasio, J. M. and R. C. Carlisle
1986 Pennsylvania Pioneers. Natural History 12:20-27.

Adovasio, James M., A.T. Boldurian and R.C. Carlisle
1988 Who Are Those Guys?: Some Biased Thoughts on the Initial Peopling of the New World. In Americans Before Columbus: Ice Age Origins, edited by R.C. Carlisle, pp. 45- 61. Ethnology Monographs No. 12, Department of Anthropology, University of Pittsburgh. Pittsburgh.

Adovasio, James M., R. Fryman, A.G. Quinn, D.C. Dirkmaat, and D.R. Pedler
1998 The Archaic West of the Allegheny Mountains: A View from the Cross Creek Drainage, Washington County, Pennsylvania , pp 1-28. In The Archaic Period in Pennsylvania. Hunters-Gatherers of the Early and Middle Holocene Period, edited by Paul A. Raber, Patricia E. Miller and Sarah M. Neusis, Recent Research in Pennsylvania Archaeology No. 1.. Pennsylvania Historical and Museum Commission. Harrisburg.

Anderson, David G.
1990 The Paleoindian Colonization of Eastern North America. A View from the Southeastern United States. In Early Paleoindian Economies of Eastern North America, edited by Kenneth B. Tankersley and Barry L. Isaac, pp. 163-216. Research in Economic Anthropology Supplement 5, JAI Press, Greenwich, Connecticut.

Anonymous
1991 Cultural Resource Management in Pennsylvania: Guidelines for Archaeological Investigation. Pennsylvania and Historical Museum Commission, Bureau for Historic Preservation, Harrisburg.

Anonymous
2006 Clovis at Topper. Mammoth Trumpet 21(4):1

Arnold, J.R. and W.F. Libby
1951 Radiocarbon Dates. Science 113(2927):111-120.

Bache, Charles and Linton Satterthwaite
1930 Excavation of an Indian Mound at Beech Bottom, West Virginia. The Museum Journal, Vol. 21, Nos. 3-4, pp. 133-188. University Museum, University of Pennsylvania, Philadelphia.

Barber, Daniel M.
1965 Portageville Site: Cultural and Temporal Significance. Museum Service, Bulletin of the Rochester Museum 38(7-8):65-71.

Bebrich, Carl A.
1968 A Supplementary Report on the Lithic Artifacts from the Sheep Rock Shelter. In Preliminary Report of Archaeological Investigations of the Sheep Rock Shelter Site, Huntingdon County, Pennsylvania. Edited by Joseph W. Michaels and James S. Dutt, pp. 313-452. Department of Anthropology, Pennsylvania State University, Occasional Papers in Anthropology No. 5.

Beers, J.H. (pubs.)
History of the Counties of McKean, Elk, and Forest, Pennsylvania. Chicago.

Blakeslee, Donald J.
1994 Reassessment of Some Radiocarbon Dates from the Central Plains. Plains Anthropologist, Vol. 39, No. 148.

Bliss, Wesley
1942 Archaeological Field Activity of the Pennsylvania Historical and Museum Commission in 1941. Pennsylvania Archaeologist 12 (2): 35-38.

Boldurian, A.T.
1985 Variability in Flintworking Technology and the Krajacic Site: Possible Relationships to the Pre-Clovis Paleoindian Occupation of the Cross Creek Drainage in southwestern Pennsylvania. PhD. dissertation, University of Pittsburgh, Pittsburgh.

Dalla Bona, Luke
1994 “Volume 3: Methodological Considereations A Report Prepared for the Ontario Ministry of Natural Resources. Lakehead University: Center for Archaeological Resource Prediction, Thunder Bay, Ontario.

Barber, Daniel B
1965 Portageville Site: Cultural and Temporal Significance. Museum Service, Bulletin of the Rochester Museum 38(7-8):65-71.

Blakeslee, Donald J.
1994 Reassessment of Some Radiocarbon Dates from the Central Plains. Plains Anthropologist, Vol. 39, No. 148.

Brose, D.S.
1994 The South Park Village Site and the Late Prehistoric Whittlesey Tradition of Northeast Ohio. Monographs in World Archaeology 20. Prehistory Press, Madison, Wisconsin.

Brose, David S., William C. Johnson and Michael J. Hambacher
1978 Excavations at the East Wall Site (33AB41), Conneaut, Ohio. Report prepared by the Cleveland Museum of Natural History, Cleveland, Ohio, for United States Steel Corporation, Pittsburgh, Pennsylvania.

Broyles, Bettye J.
1966 Preliminary Report: The St. Albans Site (46KA27), Kanawha County, West Virginia. West Virginia Archaeologist 19:1-43.

1969 Distribution of Southeastern Archaic Projectile Points in the Ohio Valley. Southeastern Archaeological Conference Bulletin No. 11: 31-36.

1971 Second Preliminary Report: The St. Albans Site, Kanawha County, West Virginia. West Virginia Geological and Economic Survey Report Archaeological Investigation 3.

Burkett, Carl K., Jr., and Richard K. Cunningham
1997 The McFate Site and Late Woodland Settlement and Subsistence in French Creek Valley, Northwestern Pennsylvania. Pennsylvania Archaeologist 67(1):4-20.

Burkett, Kenneth
1977 A Rockshelter Burial in Northwestern Pennsylvania. Pennsylvania Archaeologist 47(4): 8-26.

1999 Prehistoric Occupation at Fishbasket. Pennsylvania Archaeologist 69(1): 1-100.

Campbell C., and I.D. Campbell
1989 The Little Ice Age and Neutral Faunal Assemblages. Ontario Archaeologist 49:13-33.

Calkin, Parker E. and Kathleen E. Miller
1977 Late Quaternary Environment and Man in Western New York. In: Amerinds and Their Paleonevironments in Northeastern North America. Walter S. Newman and Bert Salwen (eds.). Annals of the New York Academy of Sciences 288: 297-315.

Carbone, Victor A.
1974 The Paleo-Environment of the Shenandoah Valley. In The Flint Run Paleo-Indian Complex: A Preliminary Report 1971-73 Seasons, edited by William M. Gardner, pp. 84- 99. Occasional Publication Number 1, Archaeological Laboratory, Department of Anthropology, The Catholic University of America, Washington, D.C.

Carpenter, Edmund S.
1942 Archaeological Reconnaissance in the Upper Allegheny Valley. Pennsylvania Archaeologist 12(1):20-23. Milton.

1950 Five Sites of the Intermediate Period. American Antiquity, Vol. 15, No. 4, pp. 298-314. Menasha.

1950a Four Hopewellian Tumuli in Western New York. Journal of the Washington Academy of Sciences, Vol. 40, No. 7, pp.209-16. Washington.

1956 The Irvine, Cornplanter and Corydon Mounds, Warren County, Pennsylvania. Pennsylvania Archaeologist 26(2): 89-115.

1971 The Irvine, Cornplanter and Corydon Mounds, Warren County, Pennsylvania, pp. 267- 286. In Foundations of Pennsylvania Prehistory. Anthropological Series of the Pennsylvania Historical and Museum Commission No. 1., Harrisburg.

Carpenter, Edmund S. and Harry L. Schoff
1951 The Nelson Mound, Crawford County, Pennsylvania. Pennsylvania Archaeologist 21(3- 4): 57-59.

Carr, Kurt W.
1998 Archaic Site Distribution and Patterns of Lithic Utilization During the Middle Archaic in Pennsylvania. In The Archaic Period in Pennsylvania: Hunters-gatherers of the Early and Middle Holocene, edited by Paul A. Raber, Patricia E. Miller, and Sarah M. Neusius. Pennsylvania Historical and Museum Commission.

Carr, K. W. and J. M. Adovasio
2002 Paleoindians in Pennsylvania. In Ice Age Peoples of Pennsylvania, edited by Kurt W. Carr and James M. Adovasio, pp. 1-50. Recent Research in Pennsylvania Archaeology No. 2. Pennsylvania Historical and Museum Commission, Harrisburg.

Chapman, Jefferson
1977 Archaic Period Research in the Lower Little Tennessee River Valley. University of Tennessee Department of Anthropology Report of Investigations No. 18.

Childs, Ronald
1989 History of Forest County 1867-1967. Forest Press, Tionesta.

Clark, Neil A., Stanley W. Lantz and William J. Robinson
1960 The Danner Mound. Pennsylvania Archaeologist 30(2): 37-45. Gettysburg.

Coe, Joffre L.
1964 The Formative Cultures of the Carolina Piedmont. Transactions of the American Philosophical Society 54 Part 5, Philadelphia.

Cowin, Verna
1981 Archaeological Survey in West Central Pennsylvania, Region VII. Report submitted to the Pennsylvania Historical and Museum Commission.

1991 The Middle Archaic in the Upper Ohio Valley. Journal of Middle Atlantic Archaeology 7:43-52.

Crane, H.R. and J.B. Griffin
1958 University of Michigan Radiocarbon Dates, III. Science 128: 1117-1123.

Dean, Robert
n.d. U.S.D.A. Forest Service Cultural Resource Inventory Form Region 9. 04-135 Tionesta Mound.

Deller, D. Brian
1989 Interpretation of chert Type variation in Paleoindian industries, southwestern Ontario. In Eastern Paleoindian Lithic Resource Use, C.J. Ellis and J.C. Lothrop, eds., pp. 191-220. Boulder: Westview Press.

Deller, D. Brian, and Christopher J. Ellis
1986 Thedford II: A Paleo-Indian Site in the Ausable River Watershed of Southwestern Ontario. Report on file, Ontario Ministry of Citizenship and Culture, Toronto.

Dincauze, Dena F.
1968 Cremation Cemeteries in Eastern Massachusetts. Papers of the Peabody Museum of Archaeology and Ethnology, Harvard University 59(1). Cambridge.

1976 The Neville Site: 8000 Years at Amoskeag, Manchester, New Hampshire. Peabody Museum Monographs 4.

Donehoo, George P.
1928 A History of the Indian Villages and Place Names in Pennsylvania. Telegraph Press, Harrisburg, Pennsylvania.

Dragoo. Don W.
1959 Archaic Hunters of the Upper Ohio Valley. Annals of the Carnegie Museum 35: 139- 245.

1963 Mounds for the Dead: An Analysis of the Adena Culture. Annals of Carnegie Museum, Vol. 37. Pittsburgh.

1965 Archaeological Investigations in the Kinzua Area of the Allegheny Basin of Western Pennsylvania during 1965. Report submitted by Carnegie Museum, Pittsburgh, Pennsylvania, to the National Park Service, Region 5 Office, Philadelphia. Project No. N.E.R. 82.

1971 Excavations at the Johnston Site. In Foundations of Pennsylvania Prehistory, edited by B. Kent, I.F. Smith, III and C. McCann, pp. 545-587. Pennsylvania Historical and Museum Commission, Harrisburg.

1971 Adena in the Upper Ohio Valley. In Foundations of Pennsylvania Prehistory. Anthropological Series of the Pennsylvania Historical and Museum Commission No. 1, pp.203-222. Pennsylvania Historical and Museum Commission, Harrisburg.

1976 Prehistoric Iroquoian Culture in the Upper Ohio Valley. In the Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium Revised, edited by David S. Brose, pp. 76-88. The Cleveland Museum of Natural History, Cleveland, Ohio.

1977 Prehistoric Iroquoian Culture in the Upper Ohio Valley. In Current Perspectives in Northeastern Archaeology: Essays in Honor of William A. Ritchie, edited by Robert Ee. Funk and Charles F. Hayes III, pp. 41-47. Researches and Transactions of the New York State Archaeological Association 17(1). Rochester.

Dragoo, Don W. and Stanley W. Lantz
1967 Archaeological Investigations at the Onoville Bridge Site (30CA5) and at Other Locations in the Allegheny Reservoir in Pennsylvania and New York. Report to the U.S. National Park Service on Projects P.O. No. 14-10-0529-2878 and P.O. No. 950-78. Carnegie Museum of Natural History, Pittsburgh.

Ellis, Christopher and D. Brian Deller
2000 An Early Paleo-Indian Site Near Parkhill, Ontario. Archaeological Survey of Canada Paper 159, Mercury Series. Canadian Museum of Civilization, Hull, Quebec, Canada.

Dragoo, Don W. and Stanley W. Lantz
1969 Salvage of Selected Sites in the Allegheny Reservoir in Pennsylvania and New York, 1969-1971. Report to the U.S. National Park Service on Projects No. 14-10-5-950-28 and 14-10-5-950-48. Carnegie Museum of Natural History, Pittsburgh.

Fetzer, Elmer W. and William J. Mayer-Oakes
1951 Excavation of an Adena Burial Mound at the Half-Moon Site. West Virginia Archaeologist, No. 4, pp. 1-25. Moundsville.

Fitting, J.E. and C.E. Cleland
1969 Late Prehistoric Settlement Patterns in the Upper Great Lakes. Ethnohistory 16(4):289- 302.

Fogelman, Gary L., and Stanley W. Lantz
2006 The Pennsylvania Fluted Point Survey. Fogelman Publishing Company, Turbotville, Pennsylvania.

Fowler, M.
1959 Summary Report of Modoc Rock Shelter, 1953, 1954, 1955, 1956. Report of Investigations No. 8, Illinois State Museum. Springfield.

Fryman, R.F.
1982 Prehistoric Settlement Patterns in the Cross Creek Drainage. In Meadowcroft: Collected Papers on the Archaeology of Meadowcroft Rockshelter and the Cross Creek Drainage, edited by R.C. Carlisle and J.M. Adovasio, pp 53 68. Prepared for the symposium "The Meadowcroft Rockshelter Rolling Thunder Review: Last Act" Forty Seventh Annual Meeting of the Society for American Archaeology, 14 17 April, Minneapolis, Minnesota.

Funk, Robert E.
1976 Recent Contributions to Hudson Valley Prehistory. The University of the State of New York, The State Education Department, New York State Museum, Memoir 22.

1989 Some Contributions of Archaeology to the Study of Cave and Rockshelter Sediments: Examples from Eastern New York. Man in the Northeast 37:35-112.

Funk, Robert E. and Bruce E. Rippeteau
1977 Adaptation, Continuity, and Change in Upper Susquehanna Prehistory. Occassional Publications in Northeastern Anthropology No. 3.

Gardner, William M.
1978 Comparison of Ridge and Valley, Blue Ridge, Piedmont, and Coastal Plain Archaic Period Site Distribution: An Idealized Transect (Preliminary Model). Paper presented at the Annual Meeting of the Middle Atlantic Archaeological Conference, Rehoboth, Delaware.

1983 What Goes Up Must Come Down: Transhumance In the Mountainous Zones of the Middle Atlantic. In Upland Archaeology in the East: A Symposium, edited by C. Geier, M. Barber, and G. Tolley. Pp. 2-42. U.S.D.A. Forest Service Southern Region Cultural Resources Report No. 2. Roanoke, Virginia.

Gardner, William M. and William P. Boyer
1978 A Cultural Resources Reconnaissance of Portions of the Northern Segment of Massanutten Mountain in the George Washington National Forest, Page, Warren, and Shenandoah Counties, Virginia. Thunderbird Research Corporation, Front Royal, Virginia.

Gardner, William M., and Robert A. Verry
1979 Typology and Chronology of Fluted Points from the Flint Run Area. Pennsylvania Archaeologist 49(1):13-46.

George, Richard L.
1974 Monongahela Settlement Patterns and the Ryan Site. Pennsylvania Archaeologist 44(1- 2):1-22.

1978 The McJunkin Site, a Preliminary Report. Pennsylvania Archaeologist 48(4):33-47.

1998 The Early Woodland Thorpe Site and the Forest Notched Point. Archaeology of Eastern North America 26: pp. 1-32.

George, Richard L. and Harry Bassinger
1975 The Wadding Rockshelter, 36AR21. Pennsylvania Archaeologist 45(4): 1-21.

George, Richard L., and Christine E. Davis
1986 A Dated Brewerton Component in Armstrong County, Pennsylvania. Pennsylvania Archaeologist 56(1-2): 12-20.

Gordon, Samuel G.
1959 The Mineralogy of Pennsylvania. Special Publication No. 1. The Academy of Natural Sciences Philadelphia.

Granger, Joseph
1981 The Steward Site Cache and a Study of the Meadowood Phase “Cache Blade” in the Northeast. Archaeology of Eastern North America 9:63-102.

Graybill, Jeffery R.
1989 The Shenks Ferry Complex Revisited. In New Approaches to Other Pasts, Edited by W. Fred Kinsey, III and Roger W. Moeller, Archaeological Services, Bethlehem Conn.

Graybill, Jeffrey R., Henry Heisey and W. Fred Kinsey
1971 Shenk's Ferry Culture: A View From 1971. In Introduction. Pennsylvania Archaeologist, Vol. 41(4):1-6. Ann Arbor.

Griffin, James B.
1943 The Fort Ancient Aspect: Its Cultural and Chronological Position in Mississippi Valley Archaeology. University of Michigan Press, Ann Arbor.

Guthe, Alfred K.
1958 The Late Prehistoric Occupation in Southwestern New York: An Interpretive Analysis. Researches and Transactions of the New York State Archaeological Association 14(1). Albany.

Hart, John P., and Hetty Jo Brumbach
2003. The Death of Owasco. American Antiquity 68:737-752.

Hasentab, R.J.
1987 Canoes, Caches and Carrying Places: Territorial Boundaries and Tribalization in Late Woodland Western New York. The Bulletin-Journal of the New York State Archaeological Association 95:39-49.

Hatch, James
1981 A Shenks Ferry Vessel from the Upper Bald Eagle Drainage. Pennsylvania Archaeologist, Vol. 51 (1-2):8-10. Ann Arbor. Ann Arbor.

1983 A Stratigraphic Analysis of Late Woodland Material Culture Change at Fisher Farm. Pennsylvania Archaeologist, Vol. 53(1-2): 11-27.

Heisey, Henry
1971 An Interpretation of Shenks Ferry Ceramics. Pennsylvania Archaeologist, Vol. 41(4):44- 70.

Hemmings, E. Thomas
1978 Exploration of an Early Adena Mound at Willow Island, West Virginia. West Virginia Geological and Economic Survey, Morgantown.

Henretta, James E.
1929 Kane and the Upper Allegheny. Philadelphia, PA.

Herbstritt, James T. and Dave Love
1975 Archaeological Investigations of Split Rockshelter, 36EL4, Horton Township, Elk County, Pennsylvania. Pennsylvania Archaeologist 45(4): 22-44.

Herbstritt, James T. and Barry C. Kent
1990 Shenks Ferry Revisited: A New Look at an Old Culture. Pennsylvania Heritage 16(1):12- 17.

Hughes, H.D.
1981 Looking for Ephraim or Letters and Journals of Early Pioneers in Elk County. Elk County Historical Society.

Hughes, Richard and Paul B. Weissman
1982 Cultural Resources Assessment Study for the Selection of Power Plant Siting Areas in Western Maryland. Maryland Historical Trust Manuscript Series, No. 25. Annapolis.

Johnson, William C.
1975 The Late Woodland Period in Northwestern Pennsylvania: A Reappraisal and Update, 1975. Paper Presented at the Annual Meeting of the Eastern States Archaeological Federation, Columbus, Ohio, November 14-16, 1975.

1976 The Late Woodland in Northwestern Pennsylvania: A Preliminary Survey and Analysis for the Symposium on the Late Woodland Period in the Lake Erie Drainage, in: The Late Prehistory of the Lake Erie Drainage Basin: A 1972 Symposium Revised, D.S. Brose (Ed.), Cleveland Museum of Natural History.

1982 Chesser Notched Points. In the Prehistory of the Paintsville Reservoir Johnson and Morgon Counties, Kentucky. By J.M. Adovasio et al. Department of Anthropology, University of Pittsburgh.

1993 Those Other Iroquois: The Proto-historic Monongahela and the Case for an Iroquoian Connection. Paper prepared for the 44th Annual Meeting of the West Virginia Archaeological Society, Huntington, West Virginia.

1994a Tracing the Glaciated Allegheny Plateau Tradition: A Preliminary Culture History of the Late Woodland Period (ca. A.D. 1000-1600) on the Glaciated Allegheny Plateau Section of Northwestern Pennsylvania and Southwestern New York. Paper presented at the Sixty-Fifth Annual Meeting of the Society for Pennsylvania Archaeology, Pittsburgh, April 22-24, 1994.

1994b McFate Incised, Conemaugh Cord-Impressed, Chautauqua Simple-Stamped and Chautauqua Cord-Marked: Type Definitions, Refinements and Preliminary Observations on Their Origins and Distributions. Paper presented at the Sixty-Fifth Annual Meeting of the Society for Pennsylvania Archaeology, Pittsburgh, April 22-24, 1994.

1999 Tracing the Glaciated Allegheny Plateau Tradition: A Suggested Culture History Sequence for the Late Woodland Period (ca. A.D. 1000-1600) in the Glaciated Allegheny Plateau Section of Northwestern Pennsylvania (Five Years Later). Paper presented at the Sixty-Ninth Annual Meeting of the Society for Pennsylvania Archaeology, Brookville, April 23-25, 1999.

2001c The Protohistoric Monongahela and Case for an Iroquois Connection. In Societies in Eclipse: Archaeology of the Eastern Woodlands Indians, A.D. 1400-1700, edited by David S. Brose, C. Wesley Cowan, and Robert C. Mainfort, Jr., pp. 67-82. Smithsonian Institution Press, Washington, D.C.

n.d. Prehistoric Ceramics (36Fo106) Warrant Rockshelter

Johnson, William C, et al.
1978 A Cultural Resource Survey of the Erie National Wildlife Refuge, Crawford County, Pennsylvania. A report prepared under the supervision of James B. Richardson III, PhD. and submitted to the Heritage Conservation and Recreation Service, Interagency Archaeological Services, Atlanta, in fulfillment of Order Number A-5662(78), November 1989.

Johnson, William C., et al.
1979 Archaeological Site Survey in Northwestern Pennsylvania, Region 4. Report Prepared by the Carnegie Museum of Natural History for the Pennsylvania Historical and Museum Commission, Harrisburg.

Johnson, W.C., W.P. Athens, M.T. Fuess, L.G. Jaramillo, K.R. Bastianini, and E. Ramos.
1989 Late Prehistoric Period Monongahela Culture Site and Cultural Resource Inventory. Report submitted by the University of Pittsburgh Cultural Resources Management Program to the Pennsylvania Historical and Museum Commission, Harrisburg.

Johnson, William C. and Andrew J. Myers
2004. Population Continuity and Dispersal: Cordage Twist Analysis and the Late Woodland in the Glaciated Allegheny Plateau of Northwestern Pennsylvania. In Perishable Material Culture in the Northeast.New York State Museum Bulletin 500. Albany, New York.

Jones, Robert W.
1931 Report of Robert W. Jones, Wild Rose, Wisconsin: The Clemson Mound and the Book Mound. Fifth Report of the Pennsylvania Historical Commission, pp. 97-111. Harrisburg.

Justice, Noel
1987 Stone Age Spear and Arrow Points of the Midcontinental and Eastern United States. A Modern Survey and Reference. Indiana University Press, Bloomington, Indiana.

Kent, Barry C.
1984 Susquehanna’s Indians. Pennsylvania Historical and Museum Commission. Harrisburg.

Kent, Barry C., Janet Rice and Kakuko Ota
1981 A Map of 18th Century Indian Towns in Pennsylvania. Pennsylvania Archaeologist 51(4):1-18.

Kent, Donald H.
1991 The French Invasion of Western Pennsylvania 1753. Commonwealth of Pennsylvania. Pennsylvania Historical and Museum Commission, Harrisburg.

Kohler, Timothy A. and Sandra C. Parker
1986 Predictive Models for Archaeological Resource Location. In Advances in Archaeological Method and Theory Vol. 9, pp. 397-452. Academic Press, New York.

Kohler, Timothy A.
1988 Predictive Locational Modeling: History and Current Practice. In Quantifying the present and predicting the past: Theory, method, and application of archaeological predictive modeling, edited by W.J. Judge and L. Sebastian, U.S. Government Printing Office, Washington, D.C.

Kidd, Kenneth E. and Martha A. Kidd
1970 A Classification System for Glass Beads for the Use of Field Archaeologist. Canadian Historic Sites: Occasional Papers in Archaeology and History 1:45-89. Ottawa.

Kinsey, W. Fred
1972 Archaeology in the Upper Delaware Valley. Pennsylvania Historical and Museum Commission, Anthropological Series No. 2. Harrisburg.

Kinsey, W. Fred, III and Jay F. Custer
1982 Lancaster County Park Site (36LA96): Conestoga Phase. Pennsylvania Archaeologist 52(3-4): 25-56.

Kneberg, Madeline
1956 Some Important Projectile Types Found in the Tennessee Area. Tennessee Archaeologist 12(1):17-28.

Lantz, Stanley W.
1966b The Red Ocher Culture in the Kinzua Valley. Kinzua Chapter 18, Society for Pennsylvania Archaeology Newsleter 1(2):7-10.

1967a Paleoindian-Early Man in the Kinzua Valley. Kinzua Chapter 18, Society for Pennsylvania Archaeology Newsletter 1(2):11-12.

1969 Rockshelters (Trail Shelters) of the Upper Allegheny. Pennsylvania Archaeologist 39(1- 4):1-5.

1971 The Cold Spring Site-Cattaraugus County, New York. Kinzua Chapter 18, Society for Pennsylvania Archaeology Bulletin 1(4):1-32.

1975 The Buckaloons Site at Irvine, Pennsylvania. A Preliminary Report. Carnegie Museum of Natural History, Butler, PA.

1980 Seneca Cabin Site: Historic Component of the Venatta Site (30CA46). Pennsylvania Archaeologist 50(1-2):9-41.

1984 Distribution of Paleo-Indian Projectile Points and Tools from Western Pennsylvania: Implications for Regional Differences. Archaeology of Eastern North America 12:210-230.

1986 An Archaeological Investigation of the Dill Mine Bituminous Coal Stripping Operation, Application #0850-117, In Boggs Township, Armstrong County. Carnegie Museum of Natural History, Pittsburgh.

1989 Late Woodland Occupation in Western Pennsylvania. Paper prepared for the sixtieth Annual Meeting of the Society for Pennsylvania Archaeology, Edinboro, Pennsylvania, April 28-30, 1989.

2004 The Mead Island Tradition: A Complex Early Late Woodland Occupation in the Middle Allegheny River Valley. Paper Presented at the Seventy-Fifth Annual Meeting of the Society for Pennsylvania Archaeology, Clarion, Pennsylvania, April 24, 2004.

Lantz, Stanley W., et al.
1982 A Cultural Resource Survey of the Proposed Warren to Indianola Pennsylvania Pipeline Right of Way Through Allegheny National Forest Lands in Warren and Forest Counties, Pennsylvania. Carnegie Museum of Natural History, Pittsburgh, Pennsylvania.

Lepper, Bradley T.
2005 Pleistocene Peoples of Midcontinental North America. In Ice Age Peoples of North America. Environments, Origins, and Adaptations of the First Americans. Edited by Robson Bonnichsen and Karen L. Turmire, pp. 362-394. Center for the Study of the First Americans, Department of Anthropology Texas A&M University, College Station.

Luchterhand, Kubet
1970 Early Archaic Projectile Points, and Hunting Patterns in the Lower Illinois Valley. Illinois State Museum Report of Investigations No. 19.

MacNeish, Richard S.
1952 Iroquois Pottery Types: A Technique for the Study of Iroquois Prehistory. National Museum of Canada, Bulletin 124. Ottawa.

Matlack, Harry A.
1986 Mystery of the Fort Field, the Bell Site Dig. Privately Published.

1987 Pottery of the Bell site, Clearfield County. Privately Published.

1990 Indians in Clearfield County 10,000 B.C. to 1800 A.D. Privately Published.

Mayer-Oakes, William J.
1953 An Archaeological Survey of the Proposed Shenango River Reservoir Area in Ohio and Pennsylvania. Anthropological Series No. 1, Annals of the Carnegie Museum 33: 115- 124.

1955 Prehistory of the Upper Ohio Valley; An Introductory Archaeological Study. Anthropological Series No. 2. Annals of the Carnegie Museum, Pittsburgh, PA.

McAvoy, Joseph M.
1997 Archaeological Investigations of site 44SX202, Cactus Hill, Sussex County, Virginia. Virginia Department of Historic Resources, Research Report Series No. 8., Richmond.

McConaughy, Mark A. and Janet R. Johnson
2003. Sugar Run Mound (36WA359) and Village (36WA2): Hopewell/Middle Woodland in Warren County, Pennsylvania. In Foragers and Farmers of the Early and Middle Woodland Periods in Pennsylvania, edited by Paul A. Raber and Verna L. Cowin, pp. 101-116. Pennsylvania Historical and Museum Commission, Harrisburg.

McKnight, W.J.
1929 Pioneer History of Brady Township, Clearfield County, Pennsylvania. DuBois Courier and Daily Express.

McMichael, Edward V. and Oscar L. Mairs
1969 Excavations of the Murad Mound, Kanawha County, West Virginia and an Analysis of Kanawha Valley Mounds. Report of Archaeological Investigations No. 1, edited by Bettye J. Broyles. West Virginia Geological and Economic Survey, Morgantown.

Miller, Charles L.
1986 Locational Model of Cultural Resources for the Allegheny National Forest. Privately Published. Buffalo.

Mills, William C.
1922 Exploration of the Mound City Group. Ohio Archaeological and Historical Quarterly, Vol. 31, No. 4, pp. 423-584. Columbus.

Munson, Patrick J.
1966a An Archaeological Survey of the “Wood River Terrace” and Adjacent Bottoms and Bluffs in Madison County, Illinois. Illinois State Museum Preliminary Reports No. 8, Springfield.

1966b The Sheets Site: A Late Archaic-Early Woodland Occupation in West Central Illinois. Michigan Archaeologist 12(3): 111-120. Ann Arbor.

Murphy, James L.
1977 Radiocarbon Dates from Globe Hill Shell Heap (46HK34-1), Hancock County, West Virginia. Pennsylvania Archaeologist 17(1):19-24.

1971 An Archaeological Survey of the Wood River Terrace and Adjacent Bottoms and Bluffs in Madison County, Illinois. In An Archaeological Survey of the American Bottoms and Wood River Terrace. Illinois State Museum Reports of Investigations No. 21, Part 1.

1982 Marion, Black Sand, Morton and Havana Relationships: An Illinois Valley Perspective. The Wisconsin Archaeologist 63(1):1-17.

Murphy, James L.
1971 The Fairport Harbor Site (33LA5), Lake County, Ohio. Pennsylvania Archaeologist 41(3):26-43.

Myers, Andrew J.
1997 An Examination of Late Prehistoric McFate Trail Locations. Pennsylvania Archaeologist Vol. 67(1): 45-53.

2001 An Examination of Ceramics from the Dutch Hill Rockshelter. Pennsylvania Archaeologist 71(1): 43-68.

n.d. Dutch Hill Rockshelter: A Preliminary Report of Findings. Unpublished Manuscript.

Myers, Andrew J. and Malinda Moses Myers
2007 A Preliminary Report on the Paleoindian Assemblage from Indian Camp Run No. 1 (36FO65). Pennsylvania Archaeologist 77(2):1-33.

Norona, Delf
1950 The Grave Creek “Shepherd” Tablet. West Virginia Archaeologist, No. 2, pp. 4-6. Moundsville.

Oswald, Adrian
1975 Clay Pipes for the Archaeologist. British Archaeological Reports 14. Oxford, England.

Parkman, Francis
1926 Montcalm and Wolfe: The French and Indian War. 2 vols., Boston.

Peterson, James B., Robert N. Bartone, and Belinda J. Cox
2002 The Late Paleoindian Period in Northeastern North America: A View from Varney Farm. In Ice Age Peoples of Pennsylvania, edited by Kurt W. Carr and James M. Adovasio, pp. 123-143. Recent Research in Pennsylvania Archaeology No. 2. Pennsylvania Historical and Museum Commission, Harrisburg.

Prufer, Olaf H.
1981 Raven Rocks: A Specialized Late Woodland Rockshelter Occupation in Belmont County, Ohio. Kent State Research Paper in Archaeology No. 1.

Prufer, Olaf H., and Orrin C. Shane, III
1970 Blain Village and the Fort Ancient Tradition in Ohio. Kent State University Press.

Prufer, Olaf H. and C. Sofsky
1965 The McKibben Site (33TR57) Trumbull County, Ohio. Michigan Archaeologist 11(1):9-40.

Ritchie, William A.
1928 An Algonkian Village Site near Levanna, N.Y. Research Records of the Rochester Municipal Museum, No. 1. Rochester.

1934 An Algonkian-Iroquois Contact Site on Castle Creek, Broome County, N.Y. Research Records of the Rochester Municipal Museum, No. 2. Rochester.

1936a A Prehistoric Fortified Village Site at Canandaigua, Ontario County, New York. Research Records of the Rochester Museum of Arts and Sciences, No. 3. Rochester.

1937 Culture Influences from Ohio in New York Archaeology. American Antiquity, Vol. 2, No. 3, pp. 182-94. Menasha.

1938a Certain Recently explored New York mounds and their probable relation to the Hopewell culture. Research Records of the Rochester Museum of Arts and Sciences, No. 4. Rochester.

1938b A Perspective of Northeastern Archaeology. American Antiquity, Vol. 4, No. 2, pp. 94-112. Menasha.

1944 The pre-Iroquoian occupations of New York State. Rochester Museum of Arts and Sciences, Memoir No. 1. Rochester.

1961 A Typology and Nomenclature for New York Projectile Points. New York State Museum Science Service Bulletin 384.

1965 The Archaeology of New York State. The Natural History Press. Garden City, New York.

1969a The Archaeology of New York State. Revised Edition. Natural History Press.

1994 The Archaeology of New York State. Purple Mountain Press. Fleischmanns, New York.

Ritchie, William A. and Robert E. Funk
1973 Aboriginal Settlement Patterns in the Northeast. New York State Museum and Science Service. Memoir 20. Albany.

Robson, J.
1958 A Comparison of Artifacts from the Indian Villages Quemahoning and Squirrel Hill. Pennsylvania Archaeologist 28(3-4):112-126.

Roosa, William B.
1965 Some Great Lakes Fluted Point Types. Michigan Archaeologist 11(3-4):89-102.

Schenck, J.S. (ed.)
1887 History of Warren County. D. Mason and Co. Syracuse, New York.

Schock, J.M.
1974 The Chautauqua Phase and Other Late Woodland Sites in Southwestern New York. Ph.D. dissertation, SUNY, Buffalo.

Shane, Orrin C., III
1975b Appendix: Ohio Radiocarbon Chronology. In Studies in Ohio Archaeology Revised Addition. Edited by Olaf H. Prufer and Douglas H. McKenzie, pp. 329-356. Kent State University Press.

Shumway, George
1985 English Pattern Trade Rifles. In Proceedings of the 1984 Trade Gun Conference. Part II, Selected Papers. Rochester Museum and Science Center, Research Records 18, Rochester New York.

Smith, I.F.,III
1984 A Late Woodland Village Site in Northcentral Pennsylvania; Its Role in Susquehannock Culture History. Pennsylvania Historical and Museum Commission, Harrisburg.

Smith, Ira F., III and James T. Herbstritt
1976 Preliminary Investigations of the Prehistoric Earthworks in Elk County, Pennsylvania. Pennsylvania Historical and Museum Commission, Harrisburg.

Solecki, Ralph S.
1953 Exploration of an Adena Mound at Natrium, West Virginia. Smithsonian Institution Bureau of American Ethnology, Anthropological Papers, No. 40. Washington, D.C.

Stewart, R. Michael
1983 Prehistoric Settlement Patterns in the Blue Ridge Province of Maryland. In Upland Archaeology in the East: A Symposium, edited by C. Geier, M. Barber, and G. Tolley. Pp. 43-90. U.S.D.A Forest Service Southern Region Cultural Resource Report No. 2. Roanoke, Virginia.

Stewart, R. Michael and Judson Kratzer
1989 Prehistoric Site Locations on the Unglaciated Appalachian Plateau. Pennsylvania Archaeologist 59(1):19-36.

Storck, Peter L.
1988 The Early Palaeo-Indian Occupation of Ontario: Colonization or Diffusion? In Late Pleistocene and Early Holocene Paleoecology and Archeology of the Eastern Great Lakes Region, edited by Richard S. Laub, Norton G. Miller, and David W. Steadman, pp. 243-250. Bulletin of the Buffalo Society of Natural Sciences 33, Buffalo, New York.

Speiss, Arthur E., Deborah Wilson, and James Bradley
1998 Paleoindian Occupation in the New England-Maritimes Region: Beyond Cultural Ecology. Archaeology of Eastern North America 26:201-264.

Thomas, Cyrus
1894 Report on the Mound Explorations of the Bureau of Ethnology. Twelfth Annual Report of the Bureau of Ethnology 1890-91: 5-742. Smithsonian Institution, Washington, D.C.

Thompson, A.C.
1954 The Boyers Run Rockshelter: A Preliminary Report. Pennsylvania Archaeologist 24(3- 4): 115-126.

Tippins, William H.
2007 An Inventory of Western Pennsylvania Rockshelters. Pennsylvania Archaeologist 77(2): 34-69.

Tolley, George A.
1983 Blue Ridge Prehistory: Perspective from the George Washington National Forest. In Upland Archaeology in the East: A Symposium, edited by C. Geier, M. Barber, and G. Tolley. Pp. 104-15. U.S.D.A Forest Service Southern Region Cultural Resource Report No. 2. Roanoke, Virginia.

Turnbaugh, William A.
1979 Calumet Ceremonialism as a Nativistic Response. American Antiquity 44(4):685-691.

Vento, F.J. and H. Rollins
1989 Development of a Late Pleistocene-Holocene Genetic Stratigraphic Framework as it Relates to Atmospheric Circulation and Climate Change in the Upper and Central Susquehanna River Drainage Basin. Pennsylvania Historical and Museum Commission, Harrisburg.

Waddell, Louis M. and Bruce D. Bomberger
1996 The French and Indian War in Pennsylvania 1753-1763: Fortification and Struggle During the War for Empire. Pennsylvania Historical and Museum Commission, Harrisburg.

Wall, Robert D.
1981 An Archaeological Study of the Western Maryland Coal Region: The Prehistoric Resources. Maryland Geological Survey, Baltimore.

2000 A Buried Lamoka Occupation in Stratified Contexts West Branch Valley of the Susquehanna River, Pennsylvania. Pennsylvania Archaeologist 70(1): 1-44.

Wallace, P.A. W.
1987 Indian Paths of Pennsylvania. Third Printing. Pennsylvania Historical and Museum Commission, Harrisburg.

1991 Indians in Pennsylvania. Anthropological Series 5, Second Edition. Pennsylvania Historical and Museum Commission, Harrisburg.

Webb, William S., and Charles E. Snow
1945 The Adena People. University of Kentucky Reports in Anthropology and Archaeology 6.

Webb, William S., and Raymond S. Baby
1957 The Adena People No. 2. The Ohio Historical Society. Ohio State University Press.

Wellman, Beth and Kevin Hartgen
1975 Prehistoric Site Survey and Salvage in the Upper Schoharie Valley, New York State (Abstract). Eastern States Archaeological Federation Bulletin 34:15.

Weslanger, C.A.
1972 The Delaware Indians A History. 2nd Edition. Rutgers University Press, Brunswick, New Jersey.

Wessman, A.L.
1981 A History of Elk County. Knupp Printing Company, Sykesville.

White, M.E.
1961 Iroquois Culture History in the Niagra Frontier Area of New York State. Anthropological Papers 16, Museum of Anthropology, University of Michigan, Ann Arbor.

Witthoft, John
1954 Pottery from the Stewart Site, Clinton County, Pennsylvania. Pennsylvania Archaeologist, Vol. 24 (1): 23-29.

1971 Pottery from the Stewart Site. In Foundations of Pennsylvania Prehistory Number 1. Anthopological Series of the Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.

Witthoft, John and S.S. Farver
1952 Two Shenk’s Ferry Sites in Lebanon County, PA. Pennsylvania Archaeologist, Vol. 22(1):3-32.

Witthoft, John and S.S. Farver
1971 Two Shenks Ferry Sites in Lebanon county, Pennsylvania. In Foundations of Pennsylvania Prehistory Number 1. Anthopological Series of the Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.

Wray, Charles F.
1952 Archaeology of the Illinois Valley. In Archaeology of the Eastern United States, edited by James B. Griffin, pp. 152-164. University of Chicago Press, Chicago.

Wray, Charles F., Martha L. Sempowski, Lorraine P. Saunders, and Gian Carlo Cervone.
1987 The Adams and Culbertson Sites. Rochester Museum and Science Center, Research Records 19. Rochester, New York.

Wray, Charles F., Martha L. Sempowski, and Lorraine P. Saunders
1991 Tram and Cameron, Two Early Contact Era Seneca Sites. Rochester Museum and Science Center, Research Records 21. Rochester, New York.

Wright, G.A.
1967 Some Aspects of Early Mid-Seventeenth Century Exchanges Networks in the Western Great Lakes. Michigan Archaeologist 13(4):181-197.

Wright, Henry T., and William B. Roosa
1966 The Barnes Site: A Fluted Point Assemblage from the Great Lakes Region. American Antiquity 31(6):850-860.

Yarnell, R.A.
1964 Aboriginal Relationships Between Culture and Plant Life in the Upper Great Lakes Region. Anthropology Papers 23, Museum of Anthropology, University of Michigan, Ann Arbor.

Zakucia, John A.
1954 The Bollinger Site (36LR21): A Late Woodland Manifestation in Lawrence County, Pennsylvania. MS.

1956 The Byler Mound, A Middle Woodland Manifestation. Eastern States Archaeological Federation Bulletin 15:10-11.

1960 The Chambers Site, An Historcal Burial Ground of 1750-75. Eastern States Archaeological Federation Bulletin 19:12.

1974 New Radiocarbon Dates from the Upper Ohio Valley, Appendix A. In: The Boarts Site: A Lithic Workshop in Lawrence County, Pennsylvania. Pennsylvania Archaeologist 44(1-2):100-102.