SECTION 2.0

PREHISTORIC SETTLEMENT PATTERNS IN UPLAND SETTINGS:
AN ANALYSIS OF SITE DATA IN WATERSHED D
(CONEMUGH RIVER-BLACKLICK CREEK)

Beverly A. Chiarulli

2.1 Introduction and the Physical Setting

2.2 Research Issues

2.3 The PASS File Database for the Conemaugh-Blacklick Watershed

2.4 Site Significance and Survey Priorities

One final question remains to be addressed. It is "Are there enough upland sites in our sample from the Conemaugh-Blacklick watershed to address current research questions?" Trying to answer research questions with data that may or may not exist is somewhat like the debates that flourished in the Middle Ages over how many angels could dance on the head of a pin. It may be an intellectually interesting question, but it doesn't help us decide if the survey priorities policy is a useful approach to managing the archaeological database. There will always be more sites that could be found if our surveys are intensive enough. I know of a middle school in central Pennsylvania that decided to have their sixth grade conduct an archaeological excavation in the cornfield next to the school. The sixth graders excavated and screened the soil from sixteen 10x10- ft squares and found one point and 5 flakes. Their effort, however, did not find a "significant" site or a site that could provide us with very much information that would add to our understanding of the cultural patterns in that watershed.

The approach in this section is to look at examples of research questions that apply to the Conemaugh-Blacklick watershed and then look at whether there is information currently available in upland archaeological sites to answer these questions. In this section, I am not going to try to develop all the questions that could be investigated in the watershed. Instead, I am going to look at several examples that seem to demonstrate that our current upland site database is inadequate to address these questions. The research questions that can be examined in a watershed like the Conemaugh-Blacklick where there is a relatively high site density include those on cultural identity, subsistence and migration, chronology and the spatial organization of prehistoric settlement; and lithic resources.

Cultural Identity

Previous research in this region suggests that the Conemaugh-Blacklick watershed is crossed by a boundary between Late Woodland and/or Late Prehistoric cultures influenced by Monongahela cultures from the southwest and non-Monongahela cultures from the Upper Allegheny Valley. We know that either within the watershed or just outside the watershed are Late Prehistoric villages that seem to be part of different cultural traditions. The Johnston and Squirrel Hill sites, which are in the watershed along the Conemaugh River, have shell-tempered pottery and have been classified as Monongahela villages although ceramics from the McFate culture in Northwestern Pennsylvania are found as minor types in both assemblages. Sites adjacent and to the northwest of the watershed, such as the Mary Rinn Site located in the Crooked Creek drainage, do not have typical Monongahela ceramics.

While all of these sites are in floodplain or terrace settings, there are at least three known Late Prehistoric villages in upland settings in the watershed in the peninsula between the Conemaugh River and Blacklick Creek. Other Late Prehistoric villages are located along the Catawba Path, which extends from New York to the Carolinas. As we look at the distribution of Late Prehistoric sites in the watershed, we see that the sites are restricted to a small part of the area (Figure 32). We might hypothesize that one of the reasons that so few surveys have found sites in the watershed is that this area served as a buffer between some of the Late Prehistoric Monongahela and Upper Allegheny Valley Late Woodland cultural traditions. Or is this lack of sites a result of a sampling error? In addition, it seems that some of the Late Prehistoric sites types found in adjacent watersheds, such as hamlets, are not present in this database.

Data needed to address questions of cultural identity could be found in both upland and riverine sites and is discussed below.

1. Because few villages are found in the northern half of the watershed, surveys should sample the appropriate settings to confirm that there is an unoccupied buffer zone in the watershed. The difference in projectile point styles observed by Butch Laney may reflect this boundary and can be investigated through the detailed metrical analysis of points from as many occupations as possible throughout the watershed. Research question #1: Is there a cultural barrier in the Conemaugh-Blacklick watershed that separates influences from the Upper Allegheny Valley groups from the Late Prehistoric Monongahela cultures?

a. Current database: The current distribution of sites seems to show clusters of sites around two resources areas - the Loyalhanna chert source and the Catawba Path.

b. Needed data: Are other parts of the watershed under populated because of cultural barrier? Data needed to address this question might include systematic surveys concentrated on other parts of the watershed, which currently show few sites. Analyses of artifacts from these sites might look at stylistic markers such as the length and thickness of projectile points from all time periods from the north and south parts of the watershed.

2. Research question #2: Is there a standard distance between Late Woodland villages?

a. Current database: Late Woodland villages are found in upland (n=3) and riverine settings (n=5). Two sites identified as Archaic villages are also listed, one in a saddle and one on a terrace. Four sites are identified as villages, but have no chronologic identification. Upland villages are located on hilltops (n=2) and saddles (n=1) and are in the upland section of the watershed between the Conemaugh River and Blacklick Creek. Two of the riverine villages, the Johnston and Squirrel Hill sites, have been investigated.

b. Needed data: More village sites are needed to answer this question, as is information from villages between the Conemaugh and the Blacklick. Figure 32 shows that the sites are located at an interval of approximately 2600 meters (approximately 1.5 miles) between village sites.

3. The Catawba Path was an important path from New York to the Carolinas in the Historic Period. If the route also function during the pre-European period, earlier sites may be clustered along this route. Research question #3: Is there a higher density of sites located within a ½-mile corridor surrounding the path than at greater distances from the path?

a. Current database: More than 21% of sites are located within ½-mile of the Path, including most of the village sites, are located within 1.5 miles (2.5 kilometers) of the Path (Figure 32).

b. Needed data: Systematic surveys are needed of topographic settings in which sites are found could provide us with a way to evaluate this hypothesis.

4. Another part of this research might focus on the movement of groups from the Somerset Plateau into the Allegheny Valley. Both the Conemaugh River Valley and Blacklick Creek could have functioned as transportation routes. As with the Catawba, data to address this research area might be found in sites in all topographic settings in a corridor following the waterways.

a. Current database: Few sites are located in the headwaters of the Conemaugh or Blacklick.

b. Needed data: Systematic surveys of a ½-mile zone around each stream.

Subsistence and Migration

Other questions derive from the research presented in the Meyersdale project report. In the report on those investigations the Late Prehistoric occupation was interpreted as that of two clusters representing different populations. It is also hypothesized that the Late Prehistoric people abandoned this area around AD1400 because of a climatic deterioration. Given the variety of Late Prehistoric occupations within and around the Conemaugh-Blacklick watershed, is this same pattern of abandonment seen? It doesn't seem so because the radiocarbon dates at some of the sites are later than the AD 1400 dates from the Somerset Plateau. Does the Conemaugh-Blacklick watershed perhaps contain populations who have moved from the Somerset Plateau? This is another question for which we need information from more sites to answer. Some of the same areas for research examined above would also be appropriate for this question.

In addition to investigations of village sites, we might expect a variety of other sites types for the Late Prehistoric period including hamlets, short-term encampments, and special activity sites. Only one single-component upland Late Woodland (Prehistoric) site is currently listed in the database. More of these sites are needed to address this question.

Chronology and the Spatial Organization of Prehistoric Settlement

As described in Section 2.2 above, there are several serious problems with the chronology used for this watershed. The most difficult issues are about the dates of bifurcate points, and the terminology used for the Late Woodland or Late Prehistoric period from AD 1000 to 1630. If, as has been suggested, the assignment of sites to cultural periods depends on their relationship to either the Upper Allegheny Valley cultures or to the Monongahela, then one of the ways to resolve the problem is to investigate additional sites. In many watersheds in the state, depositional patterns have buried and preserved deeply stratified sites. Stratified sites provide ideal sources of information that can be used to develop regional chronologies. So far, deeply stratified sites have not been identified in the Conemaugh-Blacklick drainage, so the chronology has been developed from surface sites with well preserved features. Stream benches, hilltops, and upland flats are topographic settings in which these types of sites have been found. Additional surveys of these areas, especially in those parts of the watershed that are under represented in the site database will add information to this research area.

Lithic Resources

Another set of major research questions in this watershed derive from the investigations of the use and distribution of the Loyalhanna chert and other lithic types in the watershed. The sites associated with the exploitation of this raw material types are very likely eligible for the National Register, as has been the case with other lithic procurement systems. For example, there is a National Register district based on the exploitation of jasper quarries in central Pennsylvania. For this research question, sites that contain artifacts of Loyalhanna chert would be significant sites and would add to our understanding of the exploitation of this resource.

Predictive Models for Site Locations

As discussed above in Section 2.2, predictive models have not been generally successful in finding significant prehistoric sites in this watershed. It appears, however, there are two main variables that appear to have affected the settlement pattern within the watershed. These are the presence of Loyalhanna Chert source and the Catawba Path. In addition, the gaps through Chestnut and Laurel Ridges are probably also major conduits for the movement of prehistoric people. While there are probably many small sites in locations like spring heads and hollows, sites associated with these other major categories of resources are probably more likely to contain significant information.

The review of the literature on previous survey approaches suggests that the most useful information is produced from systematic surveys of large areas. These surveys can be conducted as part of a compliance project like the Meyersdale project or as research program, like the Crooked Creek survey. Table 24 compares the density of sites found through systematic surveys.

Table 24: Density of Archaeological Sites found through Systematic Survey

Table of Contents

Top of Page

Recommendations

This review of the PASS file database for the Conemaugh-Blacklick watershed has provided an opportunity to conduct a detailed examination of the archaeological information related to this area from a variety of sources. Beyond the discussion above about research issues that could be addressed by upland sites, other problem areas have been identified. These include the information in the database, the information contained in CRM reports, and suggestions on ways to improve our archaeological database.

The PASS File Database

The PASS files are an important source of archaeological information for the Commonwealth of Pennsylvania, but there are several problems with this resource. Because of the way that the PASS files were originally collected, the information is very uneven in quality. Table 25 lists the methods by which sites were recorded for the watershed. The majority of sites (almost 75%) were found through unsystematic surveys. Site locations are unverified, in many cases. I recommend that the archaeological community begin a comprehensive review and correction of this and other watershed databases, especially for those watersheds that have been identified as high site density areas. If the survey priorities policy is to work, it needs to be based on accurate information. It is important to have accurate site locations, information on the number of sites that are now destroyed, and accurate chronologic assessments. There should be a mechanism for updating information about the condition of sites in the database.

The CRM Report Database

Archaeologists believe that the "gray" or unpublished site reports that have accumulated on agency shelves as a result of CRM investigations will one day be an important source of information for our syntheses of the prehistory of various regions of the state. There are a number of problems with the reports, one of which is accessibility. Although there are three designated report repositories in the state (at the BHP, the Anthropology Department of the Carnegie Museum, and Temple University) access to this information is still limited by distance and availability. In addition, it is difficult to know how many reports are available for an area and to then find those reports. It would be useful to have a list of all of the reports that is updated on a yearly basis. There are also new technologies such as scanning and creating Adobe Acrobat (pdf) files of old reports that could then be used to produce compact discs (CD) of the literature for each watershed. The CD format would provide an inexpensive and easily duplicated way to distribute the archaeological literature. If the literature was more available, we would all be able to contribute to the refinement of research questions for the watersheds.

Beyond their basic availability, the reports are difficult to use because they reflect different approaches to survey and approaches that have changed through time. While it might be best if all surveys were conducted with a single standard methodology, in both upland and riverine settings, this may be an impractical requirement. It would be very useful to have a systematic way to record the locations of areas that have been surveyed and to record the survey results by topographic feature.

Modifications to the Survey Priorities Model for the Conemaugh-Blacklick Watershed

This investigation has identified upland topographic settings that often contain significant archaeological sites. These include:

•Stream benches
•Hilltops
•Springheads

Sites are found in other upland settings like hillslopes and upland flats, but are generally less substantial occupations than those from these three locations.

It has also suggested that the most useful survey information is derived from large aerial surveys, either conducted as research, like the Crooked Creek Survey or as part of a CRM investigation, like the Meyersdale Bypass. Through surveys of large areas, archaeologists are able to investigate the spatial organization of prehistoric societies. We gain a sense of the how prehistoric people used their landscape. The examination of other approaches to developing predictive models has shown that simplistic predictive models will never be completely accurate, because the prehistoric use of the landscape was influenced by multiple complex factors.

Archaeologists might find it useful to adapt an approach, used by historians in their evaluation of the built landscape. In this approach, contexts are developed to provide a framework of the types of research questions and resources that are found in an area. The context statement could then be used as a basis to develop individual management plans, designed for each watershed. In this approach, we would evaluate each of the recorded sites in a watershed to determine their state of preservation and potential data yield and combine these evaluations with targeted areas in which additional survey is needed to answer specific questions. We would also have a better sense of which sites are most significant and require preservation and of what parts of the watershed are likely to be most affected by development. Through this approach, we would be proactive rather than reactive in our efforts to protect and understand the pre-European heritage of Pennsylvania. This investigation could be a start for the development of a management plan for the Conomaugh -Blacklick watershed.

References Cited

Archaeological and Historical Conultants (AHCS)

1987 Phase I Archaeological Survey and Phase II Workplan. Proposed Indiana Bypass (L.R. 1112, Sections 005 and 006) Indiana County, Pennsylvania. Report submitted prepared for the Pennsylvania Department of Transportation.

1989 A Study of Six Early/Middle Woodland Sites in Westmoreland County, Pennsylvania. Pennsylvania Archaeologist 59(2):46-99.

Boyd Varna G, Gary F. Coppock, Kathleen Furgerson, Benjamen R. Fischler, Bernard K. Means and Frank Vento

2000 Prehistoric Archaeological Syntheses: The U.S. 219 Meyersdale Bypass Project. Prepared for the Pennsylvania Department of Transportation.

Bureau for Historic Preservation (BHP)

1996 The Development of Prehistoric Settlement Pattern Research Priorities in Pennsylvania. Pennsylvania Historical and Museum Commission, Bureau for Historic Preservation, Harrisburg, Pennsylvania.

Burkett, Kenneth

1999 Prehistoric Occupations at Fishbasket. Pennsylvania Archaeologist 69(1):1-96.

Carr, Kurt W.

1998 Archaic Site Distribution and Patterns of Lithic Utilization During the Middle Archaic in Pennsylvania. Iin 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

Chiarulli, Beverly, Paul Raber, Christopher Stevenson, and Michael Glascock

1999 The Use of Neutron Activation Analysis to Identify Raw Material Sources for Prehistoric Artifacts in Western Pennsylvania. Report Submitted to the Pennsylvania Historical and Museum Commission, Harrisburg.

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.

1982 Archaeological Survey in West Central Pennsylvania, Region VII. Report Submitted to the Pennsylvania Historical and Museum Commission.

Dragoo, Don

1953 Blairsville Before Blair. Carnegie Magazine 27(4):117-120.

1954 Excavations at the Johnston Site, Indiana County, Pennsylvania. Pennsylvania Archaeologist 25(2):85-141.

Duncan, Richard B., Thomas C. East and Kristen A. Beckman

1996 Evaluation of the Crooked Creek Predictive Model. Report submitted to the Pennsylvania Turnpike Commission

George, Richard L.

1974 Monongahela Settlements Patterns and the Ryan Site. Pennsylvania Archaeologist 44(1-2):12-49.

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

1983 The Gnagey Site and the Monogahela Occupation of the Somerset Plateau. Pennsylvania Archaeologist 53(4): 1-97.

1984 Spruce Run, a Spring-Head Camp Site in Somerset County, Pennsylvania. Pennsylvania Archaeologist 54(1-2):2-25.

1996 The Winters Knuckles Site and the Mystery House on the Youghiogheny. Pennsylvania Archaeologist 66(2):53-70.

1997 McFate Artifacts in a Monongahela Context: McJunkin, Johnston, and Squirrel Hill. Pennsylvania Archaeologist 67(1):35-44.

George, Richard L., Jay Babish, and Christie Davis

1990 The Household Site: Results of a Partial Excavation of a Late Monongahela Village in Westmoreland County, Pennsylvania. Pennsylvania Archaeologist 60(2):40-70.

Johnson, Meredith E.

1925 Topographic and Geologic Atlas of Pennsylvania: Greensburg Quadrangle. Pennsylvania Geological Survey. Harrisburg, Pennsylvania.

Johnson, William C.

1999 Tracing the Glaciated Allegheny Plateau Tradition: A suggested Culture History Sequence for the Late Woodland Period (ca. AD 1000-1600) in the Glaciated Allegheny Plateau of northwestern Pennsylvania (Five Years Later). Paper presented at the Sixty-ninth Annual Meeting of the Society for Pennsylvania Archaeology, Brookville.

n.d. The late Prehistoric and Protohistoric Period Monongahela Culture and the Case for an Iroquoian Connection. In Societies in Eclipse, edited by David S. Brose, Robert C. Mainfort, Jr., and C. Wesley Cowan. Smithsonian Institution Press, Washington, D.C. In press.

Johnson, William C., and William P. Athens

1998 Late Prehistoric Period Monongahela Culture Settlement Patterns in the Appalachian Plateaus section of the upper Ohio River Valley: The Case for a Risk Reduction Subsistence Strategy. Paper presented in the Pennsylvania Archaeological Council Symposium on the Late Woodland Period in Pennsylvania at the Spring 1998 Meeting of the Pennsylvania Archaeological Council, New Cumberland.

Johnson, Richardson, and Bonhert

1979 Archaeological Site Survey in Northwestern Pennsylvania, Region IV. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania

Lantz, Stanley

1985 The Paleo-Indian Period. In A Comprehensive State Polan for the Conservation of Archaeological Resources, Volume II, edited by Paul Raber. Pp. 164-80. Pennsylvania Historical and Museum Commission.

Mayer-Oakes, William J.

1955 Prehistory of the Upper Ohio Valley: An Introductory Archaeological Study. Annals of Carnegie Musuem, Anthropological Series No. 2. Pittsburgh, Pennsylvania.

McHugh., William

1982 Phase I Investigations of the Proposed Stony Run Mine. Report submitted by GAI Consultants to the Bureau for Historic Preservation

Means, Bernard K.

1998 An Archaeology of Archaeology: Recent Investigations into the 1938 Martz Rock Shelters Excavation. Pennsylvania Archaeologist 70(1): 45-80.

Miller, Benjamin Leroy

1934 Limestones of Pennsylvania. Pennsylvania Geological Survey Harrisburg, Pennsylvania.

Myers, Andrew J.

1997 An Examination of late Prehistoric McFate Trail Locations. Pennsylvania Archaeologist 67(1): 45-53.

Neusius, Sarah W. and Brett Giles

1999 A Preliminary Assessment of the Importance of the Mary Rinn Site, Indiana County, Pennsylvania. Paper Presented at the 70^th Annual Meeting for the Society for Pennsylvania Archaeology, April 25, 1999. Brookville, Pennsylvania.

Neusius, Sarah W. and Phillip D. Neusius

1989 A Predictive Model for Prehistoric Settlement Systems in the Crooked Creek Drainage. Report submitted to the Pennsylvania Historic and Museum Commission. IUP Archaeological Services, Indiana, Pennsylvania.

Neusius, Sarah W. and Robert Watson

1991 Testing the Crooked Creek Upland Settlement Predictive Model: 1990-1991. Report submitted to the Pennsylvania Historic and Museum Commission. IUP Archaeological Services, Indiana, Pennsylvania.

NPW Consultants

1988 Phase II Investigations of Four Sites in the Stony Run Mine, Westmoreland County. Report Submitted to the Bureau for Historic Preservation.

Oshnock, Robert

2000 Prehistoric Use of Loyalhanna Chert. Paper Presented at the 71^st Annual Meeting for the Society for Pennsylvania Archaeology, Williamsport.

PADEP Notebook

Pennsylvania DEP Watershed Notebook - Ohio Basin,Subbasin Number 18 (The Lower Allegheny) Watershed D (Conemaugh River - Blacklick Creek) Page 18D Internet site: http://www.dep.state.pa.us/dep/deputate/watermgt

Raber, Paul A. (editor)

1985 A Comprehensive State Plan for the Conservation of Archaeological Resources. 2 vols. Pennsylvania Historical and Museum Commission, Harrisburg, Pennsylvania.

Robson, John

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

Rovegno, Mary Elizabeth

1981 Archaeological Investigations of Defined Areas at Mahoning Creek Lake, Indiana and Armstrong Counties, Pennsylvania. Report prepared by the section of Man, Carnegie Museum of natural History, Pittsburgh. Submitted to The U.S. Army Corp of Engineers, Environmental Planning Section, Pittsburgh. Contract No. 59-80-C-0063

Shaffner, M.N.

1958 Topographic and Geologic Atlas of Pennsylvania: New Florence Quadrangle. Pennsylvania Geological Survey. Harrisburg, Pennsylvania.

Shultz, Charles H.

1999 The Geology of Pennsylvania. Pennsylvania Geological Survey, Harrisburg, Pennsylvania.

Solecki, Ralph

1951 An Appraisal of the Archaeological Resources of the Morgantown Locks in West Virginia and the Conemaugh and Clarion East Branch Reservoirs in Pennsylvania. Smithsonian Institute River Basin Surveys Manuscript. Washington, DC.

Stevenson, Christopher M.

1982 Patterns of Hollow Exploitation Along the Allegheny Front, Center County, Pennsylvania. Pennsylvania Archaeologist 52(No. 3-4):1-16.

Stewart, Michael and Judson Kratzer

1989 Prehistoric Site Locations on the Unglaciated Appalachian Plateau. Pennsylvania Archaeologist 59(l):19-36.

United States Department of Agriculture

1968a Soil Survey of Indiana County, Pennsylvania. United States Department of Agriculture, Soil Conservation Service, USGPO, Washington, D.C.

1968b Soil Survey of Westmoreland County, Pennsylvania. United States Department of Agriculture, Soil Conservation Service, USGPO, Washington, D.C.

1982 Soil Survey of Somerset County, Pennsylvania. United States Department of Agriculture, Soil Conservation Service, USGPO, Washington, D.C.

1985 Soil Survey of Cambria County, Pennsylvania. United States Department of Agriculture, Soil Conservation Service, USGPO, Washington, D.C.

1965 Indian Paths of Pennsylvania. The Pennsylvania Historical and Museum Commision, Harrisburg.

Table of Contents

Top of Page

Summary of Information on Selected Sites in
Westmoreland, Indiana, and Somerset Counties

Westmoreland County, Archaic/Woodland

The Durika I Site (36WM277)

This is one of four sites investigated during a Phase II evaluation of a proposed Surface Mine Permit by NPW Consultants, Inc (1988). The sites were originally recorded by Robert Oshnock, a local vocational archaeologist, and were investigated during a Phase I Survey by GAI Consultants, Inc (McHugh 1982). The investigation of this site included a controlled surface collection and stripping of 5-10% of the topsoil to expose the base of the subsoil in a strategy designed to locate pre-European features. The investigations recovered a limited number of lithic artifacts and debitage including 2 Brewerton points, 1 Perkioman point, and 1 Levanna point and seventy-eight pieces of debitage. Over 90% of the debitage was identified as Loyalhanna Chert. Thirty percent of the flakes were primary flakes. Two features were located during the Phase II investigation. One of these, a bark lined pit, produced charcoal, which was dated to 2450 +/- 85 BC; the second produced a Late Woodland date of AD 950+/-35.

Westmoreland County, Early Archaic through Late Woodland

The Durika II Site (36WM278)

This is a large multicomponent site, which contains occupations dating from the Early Archaic through the Late Woodland. The majority of artifacts suggest that the primary occupation was during the Late Archaic period, which is supported by four radiocarbon dates of 2915+/- 35 BC; 1775 +/-125 BC; AD 1055 +/-100; 2650 +/- 80 BC; 2490 +/- 70 BC; 1190 +/- 80 BC Charred hickory nutshells were recovered from 7 features. From one of these features a charred hawthorn seed was recovered as well. Two features contained charred bark. The site shows evidence for the use of Loyalhanna Chert for lithic reduction. Seventy-four percent of the recovered lithic debitage is this chert type. Recovered projectile point types include: Brewertons, Vosburgs, Lamokas, Bare Islands, and Normanskills from the Late Archaic; Perkiomen broad spears from the Terminal Archaic period; Meadowood points from the Early Woodland; Snyders points from the Middle Woodland; and Jack's Reef, Levanna, and Madison Points from the Late Woodland. The site was estimated to cover approximately 30,800 square meters in area.

Westmoreland County, Archaic

The Durika III Site (36WM282)

This is the smallest of the sites investigated in the proposed Stony Run Surface Mine. Only two projectile point fragments were recovered along with a small number of pieces of lithic debitage. Eleven of the sixteen recovered artifacts were Loyalhanna chert indicating that the occupants of this site, like the others in this area, were exploiting the local chert sources.

Westmoreland County, Archaic/Woodland

The Durika Site IV (36WM283)

This site yielded only two diagnostic artifacts indicative of a Late Archaic occupation. One small conical feature did produce a radiocarbon date of AD 1285 +/-50 suggesting that it was also occupied during the Late Prehistoric period.

Somerset County, Archaic/Middle and Late Woodland

Spruce Run Site (36SO153)

The Spruce Run site, 36SO153, was located in1978 by following an unmapped feeder stream of Spruce Run from Laurel Summit Road (George 1984). The location of the site has an elevation of 818 m, and is approximately sixty m east and sixteen ft below the Laurel Hill summit, which has an elevation of 2700 feet. This feature would have afforded the site's occupants some protection from the weather. The underlying sandstone is of the Pottsville Formation of the Pennsylvanian age, and the soil is Hazelton Channery loam.

During the ten weeks of excavation directed by Richard George, eighty-nine square meters were excavated with six square meters taken to sterile soil. All of the 1271 artifacts found during the excavation were Prehistoric. Twenty-three features were recorded, all of which were hearths or hearth-related. A charcoal sample from Feature eleven, a stone hearth at forty-five cm, yielded a radiocarbon age of 4540 +/-300 BC.

There were eleven features attributed to the Late Woodland component. Feature 12 yielded a radiocarbon age of AD 1630 +/-140. Feature types included large, disturbed areas with amorphous shapes; basin-shaped pits; and one surface firepit. The large, shallow amorphous shaped features (1,3,4,8) were characterized by scatterings of ash, charcoal, and burnt stone. The six shallow, basin-shaped features (2,5,6,12,16,23) had sloped sides and the bottoms were generally flat. They contained stained, burned soil, charcoal, and little else. Feature 20 was a surface hearth surrounded by firecracked stones, associated with a hammerstone and three chert flakes. Features 20 and 23 were associated with a flake scatter. The flakes were recorded in an arc one meter or less from a hypothesized knapper who would have been seated facing northwest in an area devoid of debitage. Plotted artifacts within the concentration included 301 chert flakes, eight utilized flakes, one biface tool fragment, and two triangular flakes. The following artifacts were attributed to the Late Woodland to Late Prehistoric components: seven triangular points, two point tips, one preform fragment, twelve utilized flakes, thirteen hammerstones, and one 'utilized' slate fragment. Pottery was absent. It is believed that the features in the Late Woodland/Late Prehistoric component of the Spruce Run site represent a series of temporary visits. The suggested function of the site was a hunting camp where products of the hunt were butchered and cooked.

A single artifact and, tentatively, two features that are similar in content and configuration represent the Middle Woodland component of the site. Feature 18 consisted of a roughly circular grouping of burnt sandstone surrounding a pit containing fire-stained silt and charcoal. Feature 19 was another stone hearth, which also included fire-altered sandstone. A thin narrow, corner notched point was found at a depth of 10 cm. It is made of blue/gray chert with elongated quartz inclusions. Two limestone-tempered pottery body sherds were found at a depth of 7seven to twelve cm.

The Archaic component included eight features found at fourteen point five cm and below. Feature 6 was determined to be a hearth, Features 9, 11, and 22 were stone hearths, while Features 10, 14, and 21 were stone groupings. One feature,15, was indeterminate. At a depth of eighteen cm a contracting stemmed point was recovered, and was misidentified as a Stanley Stemmed Point. The point is actually a later Morrow Mountain I form, which would be appropriate for the 4540 BC radiocarbon age. One large utilized flake, eight hammerstones, and two sandstone abraders were the only other artifacts found that could be assigned to the Archaic component.

(Early Archaic - Late Prehistoric Period Sites)

Six prehistoric sites were found in Westmoreland County near the town of Collinsburg. All were within an area of four point two km (Ballweber 1989). The area lies within the unglaciated Allegheny Plateau section of the Appalachian Plateaus province. It is characterized by relatively rugged terrain with deep and narrow valleys. Site 36WM601 was located on both a Wisconsinan and a Holocene terrace. Sites 36WM602 and 36WM453 were located on a Pleistocene terrace while 36WM605 was located on a lower Holocene terrace. Site 36WM453 was located on a low hilltop saddle over-looking the confluence of Sewickley Creek and the Youghiogheny River, while 36WM628 was located on a high interior bench on the divide between Sewickley and Little Sewickley Creek, several miles from the Youghiogheny River. Site 36WM453, the Backstrum site, was previously excavated by both NPW and Richard George and will be discussed later. All six sites yielded artifacts that suggested a strong Early, Middle, and early Late Woodland presence in the area. Overall, these six sites witnessed a series of occupations spanning the Late Archaic through the Late Woodland and Late Prehistoric periods. Two of these sites were located on upland benches near the Youghiogheny River, while the other four occupied upper order stream terraces. These topographic settings appear to have been favored locations for Early through Late Woodland populations of southwestern Pennsylvania.

Early, Late, and Terminal Archaic

Site 36WM454

Site 36WM454 is situated on a relatively level hilltop saddle (Ballweber 1989). The investigations included the excavation of five linear trenches. Twenty-two features were located, but only twenty were designated as cultural. The features were classified as hearths and pits, with pits being divided into two basic types: shallow, circular basins with smooth, rounded bases, and circular, semi-elliptical, and irregularly shaped pits with conical bases. Fifteen posts were identified in the trenches, but because of the placement of the units, no patterns were found. All but one of the posts was considered to be prehistoric in origin, and all posts were tapered at their bases. The majority of the artifacts found at the site were found during controlled surface testing. A total of 700 chert pieces were found, sixty of these were tools including five points or point fragments. These included an Early Archaic Kirk-like, a Late Archaic Brewerton-like, a Terminal Archaic Ashtabula-like, and two untyped points. Other chert tools included bifaces, bifacially flaked tools, unifacially flaked tools, utilized flakes, hammers, unifacial block tools, and a pebble tool. Two groundstone tools were also identified from the site: a hammerstone and a slate gorget fragment. Two very small sherds of grit-tempered pottery were recovered from Feature 10, a hearth. They were too small to determine whether they were decorated or cordmarked. The evidence suggests that this is a multicomponent site, ranging from Early Archaic to the Woodland period. It was interpreted as a habitation location for semi-sedentary groups.

Westmoreland County, Early Archaic and Early Woodland

Site 36WM628

Site 36WM628 is located on the southwestern edge of an upland nose at approximately 345 m MSL, three point three five miles for the Youghiogheny River or three point eight five miles from the confluence of Sewickley Creek and the Youghiogheny River (Ballweber 1989). The twenty features were divided into hearths and pits. The pits were again sub-divided into two basic shapes: circular and irregularly shaped basins with three classes of base shapes- conical, irregular, and rounded. Eight of the pits were irregular and eight were rounded in shape. Of the irregular pits, six had conical bases, and of the circular pits, four were conical and four were rounded. The four remaining features at the site were classified as hearths because of associated burning activities. Feature 5 yielded a radiocarbon date of 100 +/-55 BC, which suggests an Early Woodland settlement. Twenty-nine posts were identified. No cultural materials were found in any of the postmolds, but small pieces of charcoal flecks were found in all of them. One hundred sixty-six chert pieces were recovered, including two projectile point fragments. One is a stemmed point and one is a bifurcated bass point, but both are too fragmented to distinguish a specific type. Other tools included bifaces, unifacial tools, utilized flakes, and a possible bladelet fragment. A single groundstone artifact (muller) was recovered from the site. The presence of both local and foreign cherts suggest a multicomponent site, while the low percentages of foreign cherts probably indicates at least minor Early and/or Middle Woodland occupation. It is unclear as to what the principle occupation was since one bifurcate is from the late Early Archaic and the second is unidentified. However, a radiocarbon date from Feature 5 indicates that the features were associated with an Early Woodland occupation.

Westmoreland County, Early Woodland

Site 36WM601

This site is located northwest of Collinsburg in Rostraver Township, Westmoreland County, Pennsylvania, and is situated on both a Late Wisconsinan and a Holocene terrace of the Youghiogheny River (Ballweber 1989). The testing consisted of the excavation of seven 2-x 2-m units.

A total of sixteen features were mapped and excavated at the site. Five features were found within the excavation units of Test Area A and the remaining eleven were buried within the alluvial deposits of Test Area A. Feature 13 was determined not to be of cultural origin and Feature 7 was classified as a historic feature. The rest of the features were classified as hearths or pits. All hearths were characterized by heat reddened soil, charcoal and ash concentrations, and frequent use of clustered rocks. Two of the three hearths from Zone III lying in close proximity, generated radiocarbon or C14 dates of 450 +/-55 BC and 420 +/-55 BC Feature 15, a rock hearth, yielded two radiocarbon dates of 1010 +/-55 BC and 900 +/-110 BC, which places the features at the Terminal Archaic/Early Woodland period boundary. Four of the features at the site were identified as pits. Two were irregularly shaped and had irregular bases; one of these was a large, amorphous, relatively deep pit, one was a small ovoid pit with a rounded base, and the third was a circular pit with a conical base.

The excavations of this site yielded 1380-flaked stone artifacts, over 95% of which was local chert. Of the remaining artifacts, 1220 were flaking debitage, cores, or totally unmodified blocks, while the remaining 160 artifacts fell into ten categories: utilized flakes, unifacial flake tools, bifacial flake tools, unifacial pebble tools, unifacial shatter tools, flake scrapers, spokeshaves, unifaces, flake bladelets, and bifaces. The test excavations also yielded 479 ceramic sherds, almost all of which were tempered with crushed shale or a combination of shale and sandstone. The inventory included six rims, three basal sections, and three body sherds with appended lug handles. Most of the sherds were found within and directly beneath Feature 12/3, a large cobble hearth. The location, surface and lip treatments of these sherds suggest a single depositional episode involving more than one vessel. The ceramics at this site were divided into two categories: Half-Moon "Cordmarked," and Half-Moon "Smoothed." The former of these represents 97% of the total ceramic inventory at the site. It was manufactured by rolls of coiled clay, tempered mostly with crushed shale, with small inclusions of sandstone. The cordmarking was applied vertically or obliquely to the rim. No decoration was observed on any of the sherds. Half-Moon Cordmarked is directly associated with Adena burial mounds and habitation sites.

Feature 1 was associated with a date of 450 +/-55 BC, suggesting an Early Woodland occupation. An Early Woodland Forest Notched point was recovered in a horizon with some Half-Moon Cord-Marked sherds ca. 70-80 cm above the two dated Early Woodland features. Feature 12 yielded another Early Woodland date of 420 +/-55 BC, while Feature 15 yielded dates of 1010 +/- 55 BC and 900 +/- 110 BC, representing the earliest of the site's datable occupations. However, no diagnostics were associated with Feature 15.

Westmoreland County, Early and Middle Woodland

Site 36WM602

This site is situated approximately point two km west of Collinsburg, on a wide, relatively level Pleistocene terrace that faces north and overlooks the Youghiogheny River (Ballweber 1989). The elevation of the bench is 262 m. Each test unit at this site yielded chert artifacts including primary flakes, secondary flakes, blocks, shatter, utilized flakes, a scraper, a biface, and a point fragment. None of these lithic artifacts were diagnostic. Two hundred thirty five of the artifacts were made of Vanport Chert, which is what led Ballweber to determine that this was an Early to Middle Woodland site. One cultural feature was also discovered. It consisted of a concentration of fire-cracked and reddened rocks. The large number of primary and secondary flakes found in a limited number of test pits indicated that the bench may have been utilized as a lithic reduction area.

Westmoreland County, Early Woodland, Late Woodland, Late Prehistoric

Site 36WM605

This site is located on a broad, relatively level, open grass-covered Holocene terrace bordering Sewickley Creek, Huntington Township (Ballweber 1989). A total of 99 artifacts were recovered during controlled surface testing at the site. Of the assemblage, 6 were projectile points or point fragments and seven were bifaces and utilized flakes. The points appear to range from an Early Woodland Forest Notched, to two early Late Woodland Raccoon Notched, to a Late Prehistoric triangle (preform). Four test trenches yielded two features. Feature 1 was a small basin-shaped feature with a small postmold located approximately fifty cm west of it. Feature 2 was similar in size and soil texture to Feature 1, but the sides of this pit were more vertical and the base more defined. It contained greasy soil, small pieces of fire-cracked sandstone, and small fragments of nut hulls. These suggest that the features were trash pits. The terrace was probably the location of a series of short-term encampments. The large amount of lithic debitage indicates that tool manufacture and resharpening was occurring.

Somerset County

Martz Rockshelter No. 1 and No. 2

Sites 36SO14 and 36SO223

The first rockshelter recorded in Somerset County, the Martz Rockshelter, is located two miles northwest of Meyersdale, Pennsylvania (Augustine 1938, Means 2000). The rockshelter overlooks the north bank of the Casselman River. The opening faces south, and the dripline covers an area measuring seven-and-one-half feet from the wall of the shelter. The level of human occupation was indicated by charcoal and ash deposits. Numerous flint and chert spalls were found, along with cracked animal bone, projectile points, and skin scrapers. Only seven potsherds were found, and six of these were heavy cordmarked, limestone-tempered specimens. Excavations along the west wall uncovered a slight recess from which fragments of a human skull, a worn human molar, numerous cracked animal bones, projectile points, skin scrapers, and a flint knife were recovered.

Some of the artifacts found in the cave under the overhang and ledge include: a black chert point, eight potsherds, seven complete projectile points, sixteen broken projectile points, a human molar, human skull fragments, one bear canine, and approximately one hundred spalls. It is apparent from the number of spalls, projectile points, and animal bones that the cave was used extensively. The lack of potsherds may indicate that the cave was used as a hunting camp.

Westmoreland County, Early Woodland, Early Late Woodland, Early Late Prehistoric

Backstrum No. 1 Site (36WM453)

The Backstrum No. 1 Site, 36WM453, is located in the Sewickley Creek-Youghiogheny River area of Westmoreland County (George 1992a). The test excavations revealed little about the site's occupants, but the work served to focus attention on the area's undefined early Late Woodland period. The investigator, Richard George, hoped that the Backstrum Side Notched point would serve as a time marker for the early Late Woodland in western Pennsylvania. The Backstrum Side Notched point has been found in western Pennsylvania burial mound contexts in both Washington and Allegheny Counties, in addition to this Backstrum Site. The points are described as "elongated, side notched points with excurvate blade edges and a flat base" (George 1992a: 64; see also George 1992a: 66, Table 1; 67, Table 2 for detailed metric descriptive data).

The Pleistocene terrace locus of the site has an elevation of 302 m. The local topography is slightly undulating, and at the eastern and southern perimeters of the site, spring-fed runs occur at the bottom of gentle slopes. The site was first investigated by NPW Consultants, Inc., and Phase I and II testing was conducted (Ballweber 1989:46, 63-70). The site was interpreted as a multicomponent camp. Subsequent test excavations by the Carnegie Museum of Natural History of 148 square meters in eight, two-meter wide trenches yielded eight features, seven postmolds, and 2,175 stone and ceramic artifacts. An NPW Consultants Phase II testing program confined to the northwest sector of the site resulted in the discovery of fifty-seven postmolds and sixty-seven features. The 1,101 recovered artifacts included five projectile points of Early Woodland, early Late Woodland, and early Late Prehistoric age, and 206 pottery sherds that were unable to serve as time markers. Hickory, walnut, and nut fragments were recorded in twenty-eight of the sixty-seven features. Three radiocarbon dates yielded 400 +/-50 BC, AD 460 +/-60, and AD 690 +/-50 (Ballweber 1989: 73, Table 6). The last date, which falls within the early Late Woodland time frame, was obtained from Feature 45, which contained no diagnostic artifacts. The evidence suggests that for centuries the site was a seasonal nut-gathering locus, and hunting was a likely site function in the early Late Woodland based on the number of Backstrum Side Notched points with their tips missing. George (1992a: 67-68) suggested that the AD 690 radiocarbon date obtained during NPW Consultants' Phase II investigations would probably date the Backstrum phase occupation at the site. In fact the AD 460 radiometric age also obtained during NPW Consultants' Phase II testing might also date the Backstrum phase occupation. Other early Late Woodland period diagnostics include a Lowe Flared Base and nine Garvers Ferry Corner-Notched points (George 1992a: 25, Figure 14; 38, Figure 19). A Middle Woodland occupation is also indicated by the presence of 2 Manker Corner-Notched points and several prismatic blade fragments (Ballweber 1989: 70, Figure 21d; George 1992b: 25, Figure 14; 38, Figure 19). The small number of limestone-tempered sherds recovered could be of either Middle or Late Woodland provenience. The mixed shell and limestone-tempered sherds and shell-tempered smooth surfaced sherds including sherds decorated with incised and punctated decorations (George 1992b: 24, Table 6; 25, Figure 14) are clearly of initial Drew phase Monongahela-Late Prehistoric period provenience despite George's (1992b: 24) claim that they are attributable to the early Late Woodland Backstrum phase occupation.

Westmoreland County, Early Late Archaic, Middle Woodland, Early Late Woodland, Early Late Prehistoric

Winters Knuckles Site (36WM432)

The Winters Knuckles Site (36WM432) is located on a high, narrow Holocene terrace of the Youghiogheny River in Rostraver Township, Westmoreland County, opposite the mouth of the Sewickley Creek (George 1996). The main reason for excavation at the site was the potential destruction of the site by erosion. The excavation led to the discovery of a rectangular house pattern in one of the upper strata. The house is twelve point seven m long and four point five m and six m wide depending on which end of the pattern is measured. Only three features were associated with the house structure. Feature 11, an irregularly shaped hearth, was located in the center of the northwest sector of the house. The only artifacts found in this feature were six small chert flakes. There was no charcoal. Charcoal was found, however, in Feature 8, located thirty cm south of Feature 11. A radiocarbon sample was submitted for dating, but the assay of 5530 +/-280 BC is considered unrealistic and probably the result of rodent activity or fossil fuel contamination. The third feature was Feature 1, a stone hearth that abutted the north wall. This feature consisted of a seemingly random grouping of fire-reddened sandstone slabs, interspersed with a few charcoal fragments and lithic artifacts, among which were a primary cutting flake, two utilized flakes, and 341 pieces of chert debitage.

Descriptions of artifacts from the Winters Knuckles Site are divided into two categories: 1) biface and uniface artifacts associated with the Feature 1 lithic workshop and 2) stone and ceramic artifacts in the area of the house postmold pattern. The lithic specimens in the first category totaled 2,416 and included five cores, 156 primary flakes, 413 secondary flakes, 1,741 tertiary flakes, and 101 pieces of shatter. There were no finished biface tools, such as projectile points, found in the Winters Knuckles Site artifact assemblage. Of the 380 secondary flakes recovered, thirty-three exhibit obvious edge wear attributes resulting from utilization.

Eleven 3 m square units and a portion of another were excavated to uncover the house pattern at the Winters Knuckles site. Nineteen bifacial artifacts were recovered here and all but two of these are made from locally derived cherts. Two of the bifacial tools, made from non-local chert, were given the tentative designation of Brewerton Side-Notched of the Late Archaic. Since then, these tools have been called Otter Creek (George 1996: 55, Figure 2; 56), however, both Brewerton Side-Notched and Otter Creek date to the early Late Archaic. A Backstrum Side-Notched point attributed to the early Late Woodland period, ca. AD 600, was also recovered. Two corner-notched points were also found, the first attributed to the Garvers Ferry Corner Notched type (Middle Woodland period), and the second to the Middle Woodland Manker Corner Notched type (George 1996: 64, Figure 8e; 64, Figure 9b-c; 67). A small contracting-stemmed point was found, comparable to the Kiski-Notched points that are also assigned to Middle Woodland/early Late Woodland. An expanded-stemmed point was recovered and its resemblance to the Forest Notched type suggested that it may be of Early Woodland origin. The excavated sample from the house structure also included four preform fragments and two "complete" preforms. Seven uniface tools were also recovered from within the house structure. Of these, four are scrapers. The other materials recovered from the house pattern were ceramics, represented by barely a handful of sherds. The eight small body sherds were limestone tempered, displaying cordmaking on their exteriors. The sample is tentatively classified as Watson Cordmarked ware, which could be of either Middle Woodland or Late Woodland provenience. There were also thirty-eight shell tempered body sherds of Late Prehistoric origin, thought to be attributable to the Drew phase of Monongahela, that were found in the plow zone or within the plow scars (George 1996: 56, 65). The botanical remains at the site that are attributable to prehistoric origin are only six in number. Two hazelnut and three hickory nut fragments were recovered and one maize cupule.

The only radiocarbon assay (see above) was unrealistically early for its stratigraphic position and the artifact associations. The latter suggests a Middle Woodland or early Late Woodland temporal placement. Also attributable to that era are four corner-notched points found in surface collections. The presence of a single Backstrum Side-Notched point could have a direct relationship to the age of the Winters Knuckles house, and although it has not been directly radiocarbon dated, George speculated that it dates to approximately AD 600.

Westmoreland County, Late Woodland Monongahela

Household Site (36WM61)

The Household Site (36WM61) is located in Cedar Creek Park, Rostraver Township, Westmoreland County, on a level early Pleistocene terrace overlooking a meander of the Youghiogheny River (George et al. 1990). The excavation in 1980 resulted in the exposure of about 3,658 square meters of subsoil. Six house patterns were revealed, along with forty-four features, including eighteen burials. An inner stockade postmold pattern and a stockade trench were also found. The inner stockade consisted of sixty-eight postmolds placed approximately sixty-one cm apart. Approximately one point five m separated the inner stockade pattern from the surrounding trench, designated Feature 16. The trench had an average width of 99 cm and an average depth of twenty-three cm. This feature contained only 13% of the total lithic debitage, but 44% of the ceramic sherds. All of the houses had central hearths, and five of the six had attached appendages. It has been hypothesized that House 1 was a charnel house rather than a true dwelling based on the sixteen burials and one possible unused burial pit within the postmold pattern of that structure. Seventeen of the eighteen burial pits contained skeletal material, and the pits were oval in shape with straight sides and flattened bottoms. At least six of the burial pits had sandstone slabs in association, but none of the burials were found undisturbed. All six hearths had central locations in the houses. Besides charcoal, all but one contained typical Monongahela artifacts and subsistence related debris. The house appendages were given feature numbers of 21, 22, 23, 25 and 29. Aside from the previously mentioned features, Features 31 and 32 were defined as charred mats. Feature 1 was a basin-shaped pit, Feature 35 was a large, relatively deep pit that probably functioned as an earth oven, and Feature 43 was classified as a smudge pit.

Of the ceramics found at the Household Site, 3,201 items or 93% are considered representative of the Late Prehistoric Monongahela occupation at Household. The ceramics that were believed to be products of earlier time periods are a single grit-tempered sherd and 249 limestone-tempered sherds. The 3,154 Monongahela sherds found at the site were shell tempered and subdivided into categories based on their surface treatment. Most of the sample is classified as Monongahela Cordmarked. Of a sample of 100 sherds, 4% were "S" twist and the remaining 96% were "Z" twist. There were no incised or punctuated sherds and no examples of lugs or strap handles. Lip treatment represents the only decorative medium. Of the study of fifty-one rim sherds utilized for analysis, one was cord impressed, two were plain, and forty-eight were tool impressed. Three ceramic pipe fragments were recovered from the site: one bowl lacking a stem, and, in Burial 6, a portion of a bowl and the proximal end of the stem. The 249 limestone-tempered sherds are products of the utilization of the locale by Middle and/or Late Woodland peoples predating the Monongahela village.

There were 8,038 chert specimens found, including 343 tools and 7,695 pieces of debitage. The discussion in this article focused only on the lithics that were considered Late Prehistoric or probable Late Prehistoric artifacts. Two varieties of triangular points were found at Household; one type was Monongahela and one not. The latter is represented by eleven specimens for which a Fort Ancient derivation is suggested. It may be significant to mention that the Household Burial 4 had a point of this type associated in the area of the cervical vertebrae. Twenty-five of the artifacts found are assigned to the triangular preform category, while five examples of uniface microblades and five biface knives were also recovered. One hundred and sixteen utilized flakes were found, and all exhibit use scars on one or more edges. A single example of a Ryan Notched knife was also recovered. Of the ground and pecked stone artifacts, only a small percent can be definitively assigned to Monongahela. Five of these are biconcave discoidals, seven are sandstone disks, five are celt fragments, and six are utilized quartz pebbles. Most of the non-diagnostic artifacts probably belonged to the Monongahela component, but cannot be given definitive assignment because no clear pattern of attributes can be noted.

One hundred seventy nine bone artifacts were also found, among them a whole turtle shell cup and fragments of others, antler tools, and awls and pins. Sixty-six artifacts made of marine and freshwater shells were also present. All but four of these artifacts had been put into the category of beads. Also found were seven tubular copper beads and a copper bead fragment found in Feature 26, believed to be a recently-disturbed child burial pit. The beads were not tested to determine whether they were made from Native or European copper.

Cursory examination of the faunal sample at the site suggests that the Household Site hunters concentrated on white-tailed deer, but that they were not averse to taking other edible species. Mollusk remains are plentiful, and fish scales were recovered through flotation from Feature 21, a house appendage. Maize was also found, but was seldom profuse as was common bean.

A wood charcoal sample from Feature 22, an attached appendage, was submitted for assay. The charcoal was assayed at AD 1625 +/- 80.

More recently, Johnson (2001 or n.d. or in press) has suggested that the Household site is part of his tentatively posited Youghiogheny phase, based upon the high frequency of "pie crust" lip decoration on the rim sherds, high frequency of final Z twist cordage impressions on the cord-marked ceramics, the late associated radiocarbon date, and the presence of a charnel house. The Youghiogheny phase is centered on the lower Monongahela River and adjacent lower Youghiogheny River valleys. It spans the terminal late Prehistoric Middle Monongahela and early Protohistoric periods, ca. AD 1475/1550-1615/1620). Depending on whether the copper beads associated with the infant burial were made from native or European copper, the Household site would fall on one side or the other of the appearance of European trade goods in the lower Upper Ohio Valley, ca AD 1580/1590.

Indiana County, Late Woodland

Johnston Site (36IN2)

This site was discovered by Ralph Solecki of the River Basin Survey during its Upper Ohio Valley Archaeological Survey (UOVAS) (Dragoo 1954: 1). The site is located near the town of Blairsville and within the flood zone of the Conemaugh River Reservoir. The site is situated on the first terrace on the north bank of the Conemaugh River. Test trenches dug in June/July of 1952 found evidence of an oval or round stockade. The structural features of the village include stockades and associated features, houses, storage structures, refuse pits, and fire pits.

The postmolds in association with the stockade were traced for 150 feet. A second stockade line was found eight to ten feet inside the larger stockade. Much of the stockade line on the south side of the village had been eroded away by the Conemaugh River. Dragoo estimated that the two parallel stockades, one inside the other, measured 450 feet and 430 feet in diameter, respectively. Three small semi-circular structures were found connected to the stockade. They appear to have been bastions. They measured ten feet in width and extended four to five feet from the stockade wall. Dragoo suggests that the two stockades were contemporaneous.

The people of the Johnston site built their houses in a smaller circle inside the stockades. The central part of the village appears to have been an open plaza with few fire pits. There were eight house structures found. The typical house was eighteen to twenty feet in diameter, made of saplings placed in the ground and arched over the top. There may be evidence to support the presence of low platforms used for beds inside the structures. Fire pits were placed near the center of the house. Refuse pits and burials were also found within the subsoil under the houses. Seven storage structures were found in the vicinity of, or attached to, the house structures. Cultural debris was scarce in all of these pits. The only class of artifact found in these pits was milling stones. Large amounts of decomposed organic materials were found here, however. Refuse pits of various sizes and shapes were found scattered throughout the site. Most contained animal bones, mussel shells, and potsherds. Some intruded into burials. A number of fire pits were found around the village as well. A few had been used for long periods of time and displayed brick-red earth around their exteriors, while others only contained earth blackened from the charcoal. Few cultural items were found in the charcoal and organic debris of the pits.

More recently, George (1978: 41, 44; 1997) named and defined a Late Middle Monongahela period Johnston phase with the Johnston site as its type station.

Westmoreland County

The Squirrel Hill Site (36WM35)

The Squirrel Hill Site is located on the floodplain of the Conemaugh River east of Chestnut Ridge, the westernmost ridge of the Alleghenies (Robson 1958). This large site appears to have been an oval stockade with an open-center plaza, similar to the Monongahela-type village. Scattered over the entire area were the artifactual remains of the Late Prehistoric villagers. Over one thousand triangular points were found, generally under an inch in length and finely worked made of chert. Of the ten pits excavated over four years, the following artifacts were found: ash and charcoal, FCR, bone and teeth, sandstone discodials, shell-tempered potsherds, ceramic pipes, bird bone beads, gorgets, and rubbed and polished red hematite pieces.

Six burials were found, including four adults, all tightly flexed on their sides, and two burials of infants in extended positions. All potsherds found at the Squirrel Hill Site were shell-tempered. Along with Monongahela Cord-Marked ware, recovered ceramics include a high frequency of McFate Incised ceramics and a companion type in which McFate Incised ware motifs were executed in a pseudo-corded fashion using a narrow edged tool tightly wrapped with fine diameter cordage.

Recently, George (1978: 41; 1997) has included the Squirrel Hill site in his newly defined Johnston phase, based on the presence of McFate Incised ceramics recovered there. Johnson (1999) has named and formally described the pseudo-cord decorated ceramic variety, Conemaugh Cord-Impressed type, with the Squirrel Hill site as the type locus. The Conemaugh Cord-Impressed type is most frequent at the Squirrel Hill site where it exceeds McFate Incised in relative frequency. It occurs in lower percentages at the downstream Johnston site and at the McKinley Earthwork and McFate sites in northwestern Pennsylvania.

Indiana County, Late Woodland

The Mary Rinn Site (36IN29)

The Mary Rinn Site is located adjacent to Crooked Creek in Indiana County near the town of Creekside in Washington Township, and has long been recognized as a Late Woodland site of some importance (Neusius and Giles 1999). Believed to be a stockaded village site, Mary Rinn is located between better-defined cultural groups such as the Monongahela to the southwest and the Mead Island and Allegheny Valley Iroquois of the upper Allegheny River to the north. The preliminary assessment of this site is based on the largely intact records of Virginia Gerald's 1973 and 1975 excavations, as well as the surface materials in the Boyer Collection. On the preliminary basis, it is suggested that the site is closely affiliated with the Fishbasket complex of sites located along Redbank Creek in Armstrong and Clarion Counties.

The only known excavations at Mary Rinn are the limited test excavations conducted by Virginia Gerald in the 1970's. From the records of Gerald's excavations we know that 51% of the sherds were limestone tempered, 27% were quartz tempered, and 22% were grit tempered. We also know that twenty-three whole or fragmented projectile points were recovered, including a fairly equal number of triangular, side-notched, corner-notched, and stemmed varieties. It is believed that the greatest densities of artifacts were found at the center of the field, where there is a slight rise in the ground surface. It is worth noting that there is a possible Archaic period component in the northeast corner of the field. Over the two years that Gerald excavated, 323 features were identified and excavated, including 199 postholes, sixty-seven pits, eight trench features, and nine hearth or burned areas as well as forty features of miscellaneous types. Gerald believed she had excavated at least two houses and part of a stockade line.

Mr. Boyer's collection from the Mary Rinn site originally included forty-five complete points and 121 total lithics as well as 172 catalogued ceramics and more than thirty pounds of uncatalogued sherds. We no longer have all of the rim sherds and projectile points identified in the Boyer collection, and it is suspected that they may have been integrated with the now missing excavated materials from Gerald's excavations. The Boyer collection currently contains 2,450 sherds. The aplastic in the limestone-tempered sherds is almost all leached, and may be less durable than the quartz- or grit-tempered ceramics. Large numbers of the sherds are also worn, making surface treatment hard to discern. Where cordmarking is identified, the final twist direction is often indeterminate.

The most common tempering agent is limestone, representing just over 70% of the temperaplastic. The next most common is quartz, found in 20% of the sherds. Quartz here is defined as the deliberately crushed felsic rock rather than sand particles that may be called grit by other archaeologists. Less than one percent of the sherds were tempered with grog, and just fewer than 1.5% of the sherds were tempered with what appears to be hematite or red ochre.

Of 2,156 non-decorated body sherds, 1,232 were cordmarked, six were fabric impressed, and 770 were plain. 125 were coded as indeterminate due to the complete erosion of at least one of their surfaces. Of the 195 decorated body sherds, fifty-four were decorated over a surface finish. In nineteen of these cases, final twist direction could be determined. Three exhibited Z twist and 16 exhibited S twist. In all, 107 of the 195 decorated sherds were incised, sixty-two sherds displayed cord-wrapped stick impression, 5% exhibited dentate impressions, 3% displayed punctuations, and 1.5% have decorations stamped or impressed into the surface. The range of decoration found on these body sherds is similar to that found in other early Late Woodland assemblages in northwestern Pennsylvania, including Mead Island and the Fishbasket Series. Forty-four rim sherds were recovered and placed in the Boyer Collection. Of the undecorated sherds, eleven are indeterminate, thirteen are plain, and eight were cordmarked. Five rim sherds exhibit incising, two have punctations, two are stamped, and three were cord-wrapped stick impressed. Rim profiles breakdown as follows: sixteen are indeterminate, five are outflaring, and one has an expanded lip profile.

The lithics from the site have not been studied yet, but the most common point type appears to be triangular including both Madison and Levanna points. Other point types include both Brewerton Side-Notched, a possible Genesee point, and at least one Raccoon Notched point. These suggest secondary Late Archaic, Terminal Archaic and late Middle Woodland components that would be consistent with Gerald's impression that there was a spatially segregated Archaic component on the northeastern margin of the site.

Previous references have suggested that the Mary Rinn Site represents a southern extension of the Mead Island culture. One attribute in the Mary Rinn Boyer Collection that suggested a Mead Island affinity is the added rim strip often mentioned in descriptions of Mead Island pottery. On the other hand, Mead Island ceramics are largely grit tempered with small portions of shell tempering, while the Mary Rinn ceramics are mostly limestone and quartz tempered. A high frequency of limestone tempering is found at the Fishbasket and the Fishbasket North sites. In the Boyer Collection, there were no diagnostic Clemson Island or Owasco sherds, nor was there really much evidence for Monongahela ware. The authors of this paper believe that there is a strong cultural connection between the Fishbasket sites and the Mary Rinn Site, but also a possible connection to the people who made the Mead Island ceramics. Johnson (comments on the draft report) believes that there is very little if any relationship to Mead Island. Instead, he suggests that Mary Rinn, Kimmel Mine and other sites are a distinct currently unnamed phase that are probably related to the Late Woodland (Upper Allegheny terminology AD 700-850), site at the confluence of the Little Mahoning and Mahoning Creek Fishbasket Sites.

The Archaeological Conservancy purchased the Mary Rinn Site and the IUP Archaeological Field School began investigations of the site in the summer of 2000. The results of those investigations are not yet available.

Westmoreland County, Late Prehistoric Monongahela

Ryan Site (36WM23)

The Ryan Site (36WM23) is located in Penn Township, Westmoreland County, Pennsylvania. The site is on a natural saddle with hills to the north and south, and occupies one of the two relatively level areas (George 1974). Ryan is bracketed to the east and the west by two spring-fed runs that flow into a branch of Brush Creek. The site appears to be of the typical Monongahela pattern, an oval stockade surrounding a "circle" of houses with a "plaza" or open area in the center. Eight round houses are located on the Ryan site map. Six of these are twenty feet in diameter, and one is slightly smaller. The eighth is much larger and was referred to as the Big House by George (1974). Contained within this house, whose diameter was thirty feet, were four features and a number of postmolds. A slightly irregular square-shaped structure was found at the northwest edge of the village. Postmolds were regularly widely spaced, about three feet apart, suggesting a structure that was open on all sides but roofed. In this structure was Feature 46, significant because it was one of the few non-burial features found at the site that was considered to have contained material that had not been accidentally deposited by either nature or man.

Enclosed storage pits are among the most common of Monongahela attributes. Twelve of these pits were found on the Ryan site, and they had postmolds both inside and around the perimeters. The most common attribute at the Ryan Site were relatively small, round pits sixteen of them. These were designated as potholders after Feature 48 was excavated, containing sixteen basal fragments from one vessel. Among the thirty remaining pits, there were few consistencies and many of them had been destroyed by rodents. Large post molds were represented by four examples and only one feature was found that could beyond a doubt be classified as a firepit - Feature 46 from the irregularly square-shaped structure. None of the features had been lined with stone.

There is ample evidence at the Ryan Site that its inhabitants traveled to the Allegheny River for an important raw material, the chert that they utilized in the making of their tools. Evidence of seasonal plant collection, fishing, and hunting was noted throughout the course of the excavations as were maize kernels, which indicate that agriculture was also a means of subsistence for the inhabitants of Ryan. This evidence suggests that the Ryan Site was occupied throughout the year and that a variety of environments were being exploited.

Three wood charcoal samples were submitted for radiocarbon assay. These yielded dates of AD 690 +/-80 for Feature 30, AD 770 +/-80 for Feature 40, and AD 1120 +/-80 for Feature 46. The latter of these dates is the one that appears most valid because this sample was based on large pieces of charcoal in Feature 46, whereas the other dates were obtained from aggregates of small wood fragments scattered about the features. Also, the first two dates were too early for Monongahela and would be considered Late Woodland in any other context. The Monongahela component at the Ryan site is assignable to the Early Monongahela Drew phase.

Somerset County, Late Prehistoric, Monongahela Somerset Phase

The Quemahoning Site (36SO15)

The Quemahoning Site (36SO15) is located in Somerset County, on the north bank of Stony Creek (Robson 1958; George 1983: 93-97). Steep, tree-covered hillsides surround the entire bottomland, forming a protective wall. The site consists of twelve rows of features forming a huge rectangle. These were identified as fire or refuse pits. Twelve fire pits were excavated in December of 1956. They averaged 36 inches in diameter and 24 inches in depth. They contained ash, charcoal, fire-cracked rock, animal and bird bones, limestone-tempered potsherds, flint chips, and freshwater clamshells. Eight refuse pits were excavated, with similar contents plus additional artifacts. Sherds were numerous, ceramic pipestems, crude triangular points, small celts of hematite, a small stone bowl, a bone comb, flint flake knives and scrapers were all found. On another dig, a perfectly smooth, dark-red 'floor' was discovered. It was oval in shape, appearing to be heavily burnt although no ash or charcoal was found. Robson does not venture a guess as to the origin of this feature. More recently (1975-1976), a portion of the site was excavated by members of the Conemaugh and Somerset chapters of the Society for Pennsylvania Archaeology (George 1983: 93).

The main village area of the site produced primarily limestone-tempered potsherds, but they were so abundant that a final count was never made. Nine partially reconstructed bowls were the yield from these sherds. Fifty yards to the west of the main village, seven firepits were found. Just north of these was a pit that yielded grit-tempered sherds, and from these a large bowl was partially reconstructed. The bowl bears deep cordmarking over its entire surface. Just south of the main village a firepit yielded three large rim sherds and a six-inch rhyolite spear. The rim sherds have broad smooth lips with much flare and are stamped with punch marks near the top in the interior. The tempering agent is not known, but it appears to be abundant thin lenses of a microcrystalline material that sparkles when moved in the light. On the sherds from the main village, the tempering is limestone and decorations are restricted to the rims and neck areas. The decorations are usually lines incised before the bowl dried and with no distinct pattern. Occasionally there are rows of deep punch marks or fingernail marks. The rims bear appendages that are numerous and varied. Those on the rim are usually scalloped. The refuse pits also produced several pipe stems. They are oval, round, or rectangular in cross-section.

The remains of an Early Archaic camp were found overlooking the bend in the river and directly opposite the village site. Evidence of earlier cultures are present in the village site, but bear no similarities to those on the bluff.

Projectile points are the most numerous stone artifacts found at the Quemahoning site. Large and small triangular points are the most abundant . The cache blades are of jasper, the long fishtail spears are of rhyolite or black flint, and the small corner-notched specimens are also chert. Scrapers, knives, and blanks are numerous throughout the entire site. Celts are the most numerous of the groundstone artifacts. Gorgets are also numerous, usually made of fine-grained green sandstone. There was one bannerstone. No bi-pitted stones were found. The site produced two granite discs and one smooth sandstone discodial. A fragment of one bone comb was the only bone artifact from the site.

The 1975-1976 excavations uncovered evidence for a single palisade line (without an associated stockade trench or borrow pits) and at least five house structures of which four were round and the fifth rectangular (George 1993: 93). Other possible partial house postmold patterns are evident as well as other discontinuous linear postmold patterns suggesting stockade segments (see George 1993: 93, Figure 45). Numerous culture features were also documented.

The ceramic industry displays a high frequency of limestone-tempered sherds (79%) and cord-marked surfaces (89%). Rims sherds displayed a number of low added-on collars, which characteristically occur on at least some other Somerset phase Monongahela sites. Other rim and neck sherds, however, display various appliqués including lugs, spouts, bifurcated castelations associated with spouts and semi-lunar lugs, and loop handles and incised and/or punctuated design motifs (George 1983: 95-96). The appliqué and incised and punctuated decoration applied to these latter sherds is more characteristic of Early Monongahela period Drew phase sites. This suggests that, while the Quemahoning site village component is best ascribed to the Somerset phase because of its geographical location and the high frequency of limestone-tempered pottery and added-on collars, it - like the Montague site (36SO4) on the Youghiogheny River - was occupied at a time when Drew phase decorative horizon attributes were penetrating the Allegheny Mountain section. Although George (1983: 72) suggests that the Quemahoning site village represented a unit-site intrusion from the Montague site, this seems to be unlikely given the distance. As well, the Quemahoning site shares many ceramic decorative attributes with that of the downstream Early Monongahela Kiskiminetas phase site centered along the lower Kiskiminetas River, including the presence of vessels with added-on collars and sherds decorated with cord-wrapped paddle edge or tool-stamped plats and obliques identical with those illustrated by George (1983: 97, Figure 51).

Finally, an AD 1245 +/-65 MASCA radiocarbon date is associated with the Somerset phase occupation of the Quemahoning site (George 1983: 97). Regrettably the uncalibrated age was never published; so, the actual radiocarbon age is not known nor can it be calibrated according to newer more precise methods.

Somerset County, Late Prehistoric Monongahela, Somerset Phase

Peck No. 1 Site (36SO1)

The Peck Farm is located in Somerset County (Augustine 1938). Two sites were found on this farm. Traces of five occupations were found, all of them near the mouth of the creek on the south or west bank of the Casselman River.

At the Peck No. 1 site, slightly more than one acre was completely topped. In this area were found three stockades superimposed upon each other. The pottery and cultural artifacts found here indicate that the occupations were identical. Thirty-eight circular house outlines were determined from the postmolds uncovered and one house pattern that lay outside of the protective palisades. 126 fire, refuse and storage pits were found, but they had been badly damaged by a subsoiling machine that had been used to drain the land. Approximately three bushels of pottery sherds were found, one small-unbroken vessel, and three restored vessels. Limestone temper was predominant although an occasional shell or sand hardened specimen was found. Decoration was limited to impressed and incised lip designs, and cordmarked and plain exteriors were common. Also found were 126 triangular projectile points, 107 notched projectile points, broken pottery pipes, bone tools, and other chert and groundstone artifacts.

Five burial disturbances were also recorded, but skeletal material was so greatly decomposed that little more than traces were observed for recording.

Somerset County, Late Prehistoric, Monongahela, Somerset Phase

Peck No. 2 Site (36SO8)

A second site was found on the Peck Farm, Peck No. 2 site (Augustine 1938). The site consists of a stockaded village, three hundred feet in diameter. Sixty-one circular house outlines were found, averaging in diameter from twelve to eighteen feet. Eleven pear shaped storage houses were also recorded, but artifacts in these houses were rare. Fifty-five fire and refuse pits were recorded with generally heavy oxidation and ash and charcoal deposits. The following artifacts were recovered from pits on the field: stone and pottery pipes, flint drills, stone disks, bone awls and skewers, projectile points, tooth pendants, knives, gorgets, beads, and various other bone and stone artifacts. Unbroken pottery vessels were rare. Only one partly broken vessel was associated with the twenty-four burials.

The predominant culture at both this and other sites in this area of Somerset County appears to be Algonquin with varying degrees of Iroquois influence.

Somerset County, Late Prehistoric, Monogahela, Somerset Phase

Powell No. 1 Site (36SO11)

Surface indications were found at a Native American site on an upland saddle on the Sylvester Powell Farm in Shade Township, Somerset County (Augustine 1938).

In the first of the Powell sites, twenty-six house outlines, all circular or oval, were traced from more than one thousand postmolds. Forty-six fire, refuse, and storage pits were recovered and ten burials were recorded. Artifacts were rare. The ten burials were all highly fragmentary. In none of the burials was the person buried with his head facing west. Heavy charcoal deposits surrounded the skeletal remains, but no evidence of cremation was found in any of the burials. No associated artifacts were recovered from any of the burials.

The village was constructed in the form of a great crescent with houses built around three sides of the circle with the southern side open and unprotected. Openings to the houses faced the center of the village, and houses were ten to twenty-seven feet in diameter. Burial disturbances inside the house outlines were common. The site included thirty-three fire and refuse pits. The artifacts recovered included axes, a ceramic cup, three broken ceramic vessels, pipes (both broken and not), triangular points, scrapers, and various tooth and bone artifacts. A bushel and a half of potsherds was recovered, tempered with limestone, sand and shell, shell temper being rare.

Somerset County, Late Prehistoric, Monongahela, Somerset Phase

Powell No. 2 Site (36SO12)

As noted before, Powell No. 1 site was located on an upland saddle on the Sylvester Powell Farm (Augustine 1938). The top of this saddle is thickly wooded, and on the west side is where Powell No. 2 site was found, although the precise location, 1.5 miles from the Powell No. 1 site, is problematic. The occupation found here was light. There were only twenty-one pits found and nineteen roughly circular house outlines were traced from the postmold features, within a palisade that was circular in shape. The house outlines were found close to the palisade with the central part of the fortification remained clear. No burials were found at Powell No. 2 site. The pits and houses found at this site were similar to Powell Site No. 1, as were the potsherds that were recovered. There was only a light artifact representation at this site. The sites are assumed to be identical as to the time of occurrence. The artifacts that were found were a small stone cup, two broken pottery pipes, a celt, a sandstone axe, and both flint and native stone projectile points, notched and triangular.

The excavations referred to in this article were part of the WPA project sponsored by the County Commissioners. The project was an archaeological survey of Somerset County. The practice of the survey was to locate sites by surface indications, as there were no written documents to indicate any village whereabouts.

Somerset County, Late Prehistoric, Monongahela, Somerset Phase

Emerick Site (36SO10)

One of the most spectacular sites in the county was discovered on the late William Emerick farm in Southampton Township, located on a saddle between Big Savage and Little Allegheny mountains. Springs below the site flow directly into Wills Creek and Gladdens Run, a tributary of Wills Creek (Augustine 1938). A complete prehistoric "Algonkian" town was exposed in which the outlines of a complicated system of fortification, thirty-seven circular houses, fifty fire, refuse, and storage pits, thirty-four burials, and an enviable collection of artifacts were recorded. The palisade enclosed an area 160 feet in diameter. The diameters of the houses were twelve to twenty feet and presented much the same appearance as an Eskimo igloo. The house outlines were near the palisade and all faced in toward the central part of the enclosure. The artifact assemblage included various stone and bone implements, such as axes, knives, projectile points, hoes, pestles and mortars, and drills. Stone and pottery tobacco pipes were found, as were many beads of various animal bone. Large quantities of broken pottery were found in the refuse pits. The ceramic specimens associated with the site were classified as "Algonquin".

Of the thirty-six burials found at the Emerick site, two were double burials. Artifacts associated with burials were rare from this area in this time period, found in about one in ten burials. Pottery vessels were associated with seven of the burials. Burial No. 5 was found with a roughly worked native stone point penetrating one of the vertebrae deep into the spinal canal. Such a wound would have produced instant paralysis.

Somerset County, Late Prehistoric, Monongahela, Somerset Phase

Troutman Site (36SO9)

Another site found in Somerset County during the WPA excavations was the Troutman Site (Augustine 1938). This site is located 4.15 miles southwest of the Emerick Site on a bench off Big Savage Mountain and drained by Gladdens Run. It is plainly visible from Emerick. Like Emerick, Troutman is several miles from any stream or river, although fresh water springs are plentiful here as well. Troutman was identified by Augustine as a fortified stronghold on a high knoll on the east side of Savage Mountain.

Like other sites in the county, Troutman was surrounded by a palisade or fortification, the largest to be found in Somerset County. The outline was an irregular circle, with an opening eight feet wide on the northwest side. Twenty-four house outlines were distinguished from the hundreds of postmolds that were staked. Because so many more postmolds were found than could be associated with any house outline, it is estimated that there were three times the number of houses than were actually determined.

Pottery at the site was rare. Only one complete vessel was found, a small limestone-tempered, cordmarked, pointed base cup. Approximately three bushels of potsherds were found in the pits. No pottery was found associated with the burials. Sixty fire, refuse and storage pits were recorded on the eastern side of the site. They were lightly oxidized and sparingly produced potsherds, animal bones, and fresh water shells. Among the artifacts collected were celts, a stone hoe, pestles and hammerstones, eight smoking pipes, and several broken pottery pipe stems and bowls. Shell objects of adornment were plentiful. Bone objects included a fishhook, projectile points, pendants, awls, and turtle shell spoons and cups. Stone projectile points were comparatively rare. Of the points recovered, triangular flint points predominated.

Fifty burials were recorded; twenty-seven were infants or children under the age of ten and twenty-three were adolescents or adults. All of the burials were found inside the palisade. Like the other sites that included burials, the bodies of the deceased never headed west and the usual burial position was flexed with the knees drawn up sharply. Again, artifacts were found in association with about one in ten burials.

Table of Contents

Top of Page

Appendix 2

Summary of CRM Projects In and Adjacent to the
Conemaugh-Blacklick Watershed

ER 84-0879-21-D

This survey was conducted under the author's direction by GAI Consultants Inc in 1984. The proposed right-of-way crosses the divide between the Conemaugh - Blacklick Creek drainage and the adjacent watershed to the east, which flows into the Susquehanna drainage. Two prehistoric sites were recorded, although one (Bloom site 36CB141), consisted only of a single flake. Site #2 is the Manor site (36CB142), where one shovel test pit yielded twenty moderately sized flakes, suggestive of a lithic reduction center. The report recommended that a Phase II survey be conducted for 36CB142. However, no additional investigations were conducted, because the project was to be avoided by the proposed construction.

Cambria County

ER 93-0421-021

No sites were found in this project by Christine Davis Consultants during a Phase I Survey. The area is located in the Colver Quad, Cambria County and covers 1.01 acres. No further archaeological work was conducted.

Cambria County

Blacklick Township

ER 91-508-21

A Phase I survey conducted by GAI Consultants found no sites in this project area. The area is located in Blacklick Township, Cambria County, on the Colver Quad. No additional investigations were recommended.

Cambria County

Jackson Township

ER 90-0481-021

Vandel Enviro-Industrial Consultants, Inc. was hired to conduct a Phase I Survey in January 1991 for the proposed Vintondale Water System Project. The major drainage in this area is Shuman Run, a tributary of the south branch of Blacklick Creek. The topography of the area consists of moderate to steep valley slopes along a small steep gradient. There was prior disturbance of the soils in three of the four segments of the survey. No sites were discovered; no further archaeological testing was required.

Cambria County

Jackson and East Taylor Townships

ER 92-3847-021

Pellegrini Engineers hired Heberling Associates to conduct a Phase I archaeological survey for their Sanitary Sewer Collection System. The project covered a distance of 16,000 linear feet and was located in the Nanty-Glo quad. The topography of the area is an upland mid-stream setting disturbed by urban development. No further archaeological investigation was recommended

Cambria County Projects adjacent to the Conemaugh - Blacklick Watershed

ER 86-1261-021

In August of 1996, a Phase I archaeological survey was conducted of a natural gas pipeline right-of-way (LN-2427) by Kelly Lynn Cosgrove and Ronald Michael of NPW Consultants, Inc. for the Consolidated Gas Transmission Corporation. The area is located near the West Branch of the Susquehanna River sub-basin and is adjacent to, but not in the watershed. The project area was 6.1 kilometers long. No sites or artifacts were found, and the report recommended that no further investigation of the area be conducted.

ER 91-2008-021

Two sites were discovered during archeological investigations of the location for the West Branch Sewer Authority's proposed Sanitary Sewage Facility. The work was conducted by Christine Davis Consultants for L. Robert Kimball and Associates. The Kennedy Avenue and Germantown sites are located within the 170,000 foot linear project area. These sites can be found in Spangler Township, Cambria County on the Barnesboro USGS Quadrangle. The major drainage in the area is Fox Run.

The Kennedy Avenue site (36CB161) is located on a stream bench. The site is non-stratified, and only a Meadowood Projectile Point and debitage fragments were found.

The Germantown site is located on a terrace that is severely eroded to bedrock. It is also a non-stratified site, and only debitage was recovered in the Phase I survey. The site is determined to be the remains of a large prehistoric settlement that has been previously impacted. The recommendations for both sites were that no further archaeological work was needed.

ER 92-1024-021

There were no sites found during the Phase I survey of this project area, when the Johnstown Area Regional Industries hired Christine Davis Consultants in December 1992. The project area is located in Richland Township, Cambria County, in the Geistown Quad. No further archaeological work was recommended.

Indiana County

Armstrong Township

ER 86-0938-063

This Phase I survey was conducted by Skelly and Loy Inc in August 1991. One site was found on the project right-of way, which is 2.8 miles long. The area is located in Armstrong Township, Indiana County, in the Ernest Quad. The major drainage in the area is Curry Run. The site was named the McQuowen Site and was identified as a lithic scatter that contained no diagnostic material. It was concluded that no further archaeological work was needed.

Indiana County

Blacklick Township

ER 95-0723-063

This Phase I archaeological survey was conducted on the site of the proposed replacement of the Aultman's Run Bridge S.R. 3031 Section 450. This survey was conducted by Christine Davis Consultants for the Pennsylvania Department of Transportation. It covered an area of .8 hectares on the Conemaugh River drainage. No sites were discovered.

Indiana County

Blacklick Township

ER 93-4193-063-B

A Phase I archaeology survey was conducted on the site of the O'Brian Methane Treatment Plant. No sites were found in this project area. The survey was conducted by IUP Archaeological Services for Devon Resources in June of 1994. The topography is an upland area, between the Conemaugh and Blacklick Rivers.

Indiana County

Burrell Township

ER 92-1710-063A

No sites were found within the project survey area, located in Burrell Township, Indiana County, in the Bolivar Quad. The project area totaled 40.9 hectares, in the Blacklick Creek drainage area. IUP Archaeological Services conducted this survey for Earthtech, Inc.

Indiana County

Burrell Township

ER 92-1710-063-B

A Phase I archaeological survey was conducted in Burrell Township by GAI Consultants of the Strangford Area project for the Burrell Township Sewer Authority. The survey was conducted for a U.S. Army Corp of Engineers, Pittsburgh District permit. The project is in the Blacklick drainage and consists of a twelve to fifteen meter wide right-of-way covering a distance of 14.8 miles. Nine sites were discovered in the project area.

However, the lack of diagnostic artifacts and low artifact densities resulted in recommendations that no further work would be necessary. A brief description of each of the sites follows. The Cornell site, 36IN379, was a disturbed lithic scatter situated on a lower slope contained no diagnostic artifacts. The Pender Avenue site, 36IN378, was located on upper slopes and also contained no diagnostic artifacts. The Gundy site, 36IN377 consisted of twenty-one chert flakes and one unidentifiable biface. It was located on upper slopes. 36IN376, the Lasser site, was on an upland flat and upper slopes. It contained nine flakes and one unidentifiable biface. The Corporate Campus site, 36IN375, yielded nine flakes and was located on lower slopes. Site 36IN374, the Demarines site, was located on a ridge top within a plow zone. The Elder Road site, 36IN373, was located in a mid-slope setting and contained only debitage. The Brown Road site, 36IN372, was in a mid-slope setting and has only four flakes. The Kennel site, 36IN371, was on lower slopes and contained fifteen flakes.

Indiana County

Burrell Township

ER 93-2598-063

Only one site was found within this twenty-one hectare project area. Of this total, approximately ten to fifteen percent was previously disturbed. The project was in the Bolivar and New Florence Quads in which Blacklick Creek is the major drainage. Wilbur Smith Associates conducted the survey, and one Early Archaic site was found. Additional testing was requested, but no report of those investigations was found in our research.

Indiana County

Center Township

ER 89-0768-063-D

Phase I archaeological investigations in the Homer City Coal Refuse Disposal Site Study Area were conducted by the Cultural Resource Management Program from the University of Pittsburgh Anthropology Department for Schneider Engineers. It covered an area of 32.4 hectares in the Cherry Run drainage. The topographic setting of this area includes hilltops and saddles, valley slopes with benches and valley floors. One flake was recovered, but no sites were identified as a result of this survey.

Indiana County

Center and Blacklick Townships

ER 2000-289-063-C

A Phase I archaeological survey was conducted for the Discharge Pipeline Project in Indiana County. This project was conducted by Christine Davis Consultants for FERC. It covered an area of 4.4 hectares in the Two Lick Drainage. The only site was an isolated find along an upland terrace, one secondary flake. No diagnostic artifacts were found, and no PASS number was assigned.

Indiana County

East Wheatfield Township

ER 84-1638-063

A Phase I archaeological survey and testing of Legislative Route 337, Section 9 and a Phase II archaeological testing of the Blacklick Site (36IN355) Buena Vista Bridge Replacement Legislative Route 337 was performed by NPW consultants for Westmoreland Engineering Co. Inc. The initial survey identified one site, 36IN355, based on the presence of eighteen primary flakes and three secondary flakes. The site is located on a topographic bench in the Blacklick drainage. The Phase II survey identified the site as a small non-intensive occupation; no diagnostic materials were found. It was suggested that no further work was necessary.

Indiana County

Green Township

ER# not given

Christine Davis Consultants conducted a Phase I archaeological survey, of T-916 over Baker's Run, for the Pennsylvania Department of Transportation. No archaeological sites were found within the .089034 hectares project area. No recommendations for additional archaeological work were made.

Indiana County

Rayne Township

ER 86-1534-063

There were no sites found when a Phase I Survey of this project area was conducted by IUP Archaeological Services in May 1987, hired by Mears Enterprises Inc. The project area is located in Rayne Township, Indiana County, in the Clymer Quad. The area consists of two ponds encompassing an area 210 ft by 50 ft. The major drainage is Rayne Run to Crooked Creek. It was concluded that no further archaeological work was needed.

Indiana County

West Wheatfield Township

ER 91-4114-063-F

A Phase I cultural resource survey was done for the Penelec Conemaugh Facility in West Wheatfield Township, Indiana County. Ebasco Environmental conducted this survey for the U.S. Corp of Engineers. It covered an area of 133.6 hectares and resulted in the identification of two sites. The major drainage in the area is the Conemaugh River. A Phase II survey was conducted on 36IN12. This site was estimated to date to the Archaic to Woodland periods. It was described as a disturbed, ephemeral occupation.

Indiana County

West Wheatfield Township

ER 93-2437-063

An abbreviated Phase I archaeological report was done for the Climax Water Service Extension Project. This survey was conducted by David Russell Archaeological Contracting for the Indiana Office of Planning and Development. It covered an area of .9 hectares in the Conemaugh river drainage. No sites where identified.

Indiana County

White, Center, and Armstrong Townships

ER 86-0938-063

Phase I Survey of the Proposed Indiana Bypass

Four sites were found within this survey project area located in White, Center, and Armstrong Townships, Indiana County, in the Ernest, Elderton, Indiana, and Whitesburg Quads by Archaeological and Historical Consultants, Inc. The topographic setting of the area is an undulating upland terrace, and a small area in the floodplain of Curry Run. The first site was Anderson Knoll (36IN331), a mid-late Archaic site that lay outside of the project area, and yielded mid-late Archaic Lamoka points. The second site was 36IN105, and also lay outside of the right of way. Site #3 was I.B. 2 (36IN330) and its chronology is unknown. The artifacts recovered at this third site were debitage. The last site was 36IN57 and has been designated a late Archaic through Early Woodland hunting camp. All four sites were recommended for further testing.

Indiana County

White Township

ER 86-0938-063

Phase II investigations were conducted of the Simpson Site in the Proposed Indiana Bypass. The Simpson site (36IN57) was named after the major drainage in the area, the Simpson Spring. The site was located in Indiana County, White Township during a field reconnaissance by Archaeological and Historical Consultants, Inc. for the proposed Indiana Bypass. The topography in the area of the site is a relatively level upland saddle, and the drainage is the Simpson Spring. Since the site's discovery, Phase I and Phase II surveys have been conducted. The site was found to be eligible for the National Register of Historic Places and additional Phase III investigations were recommended.

The Phase III Investigations were conducted by Skelly and Loy in response to recommendations in the Phase II report. During the Phase III Investigation the site area was stripped to the subsoil. The investigations are described in a later section of this report.

Indiana County

White Township

ER 88-1213-063-A

A survey of the Indiana County Transit Authority Maintenance and Office Facility was conducted by IUP Archaeological Services for the Indiana County Transit Authority in January of 1989. It covered an area of one hectare in an area drained by McCarthy Run. The topography is an upland ridge. No sites were found in the survey area.

White Township

ER 86-0938-063

Two sites were found during this Phase I Survey conducted by Penn DOT hired Skelly and Loy, Inc. in September 1993. The area of located in Indiana County, in the Indiana Quad. The first site is named Paul Beck #2, which yielded sixteen non-diagnostic flakes, whose chronology is unknown. It is located on a floodplain and a terrace. The second site, the Fenton Site, whose chronology is also unknown, is located on a foot slope and yielded fourteen non-diagnostic flakes. In the case of both sites, all of the artifacts were found in the plow zone, and it was determined that no further archaeological work was needed.

Indiana County

White Township

ER 92-1627-063-A

No sites were found within the survey project area when the Dept. of Environmental Resources hired Christine Davis Consultants in July 1993 to conduct a Phase I Survey of an Act 537 permit. The project area is located in White Township, Indiana County, in the Brush Valley and Clymer Quads. The total project area covers 46.2 acres. The major drainage in the area is Two Lick Creek. No sites were located; no further archaeological work was recommended.

Indiana County

White Township

ER 95-0644-063

No sites were found within the project area during the Phase I Survey conducted by Skelly and Loy Inc in October 1996. The project area is located in White Township, Indiana County, in the Brush Valley and Indiana Quads. The total project area is five acres, the major drainage is Two Lick, and because the Phase I Survey yielded no sites, no further archaeological work was recommended.

Indiana County

White Township

ER 95-0688-063

This was a Phase I archaeological survey proposed by the Cherry Run Sewer System and conducted by Christine Davis Consultants for the Department of Environmental Resources. The linear area consisted of 24,500 linear feet. No sites were found. No further archaeological work was recommended.

Indiana County

White Township

ER not given

A Phase I archaeological survey was conducted for the SCI Pine Grove, Pennsylvania maximum-security facility. The survey was conducted by Christine Davis Consultants for the Department of General Services. It covered an area of 33 hectares in the Two Lick drainage. No sites were located.

Indiana County

White Township

ER Not Given.

This was a Phase I survey project proposed by the Townfair Center Development and conducted by Christine Davis Consultants for the Department of Environmental Resource 537 in White Township. A total of 14.5 hectares were covered and there was no further archaeological work needed.

Indiana and Cambria Counties

East Wheatfield and Jackson Townships

ER# not given

Indiana county, U.S. Route 22 Improvement Project S.R. 0022 Section 491: Phase I Cultural Resources.

The Pennsylvania Department of Transportation contracted with Skelly and Loy to conduct archaeological investigations for this project. The project area spanned three municipalities in Indiana Co. These were Burrell, east Wheatfield and west Wheatfield. No sites were discovered.

Indiana and Cambria Counties

East Wheatfield and Jackson Townships

ER 93-0246-063-C

The Indiana County U.S. Route 22 Improvement Project S.R. 0022, Section 495 Phase I cultural resources survey was conducted by Skelly and Loy for the Pennsylvania Department of Transportation. It covered an area of 85.8 hectares in the Blacklick drainage. No sites were identified by the survey.

Indiana and Westmoreland Counties

Fairfield and Wheatfield Townships

ER 92-1204-063

A Phase I Archaeological Survey was conducted of the Bolivar Bridge Replacement Project. This survey, which was conducted by Christine Davis Consultants for the Maguire Group Inc, and SAT Consulting Engineers in December of 1997, identified twelve archaeological sites. Eleven of these were identified as prehistoric, and of these eight are potentially eligible for the NRHP. The topography in the area consists of a river valley with high, forested floodplains and gentle relief defining ridges and is approximately 60 acres.

Westmoreland County

Derry Township

ER 84-0744-129

A total of 15 sites were discovered during a Phase I survey conducted for the Latimer Construction Company by GAI Consultants for a cultural resources survey of the Stony Run Strip Mining sites. These sites; Sulkosky Farm #1, Ourika, Sulkosky #2, Dunlap (36WM360), Sulkosky #3, Sulkosky #4, Dunlap sites (36WM356, 357, 358, 360, 361), and Durika sites (36WM277, 278, 279, 280, 281, 282, 283, 284), are located within project area 84-0744-129. The sites are found in Westmoreland County, Derry Township on the Blairsville USGS Quadrangle; and the major drainage in the project area is Stony Run.

The Sulkosky Farm Site #1 (GAI site A1) is located on a terrace. After the Phase I investigation was completed, the cultural material recovered included Brewerton Corner-Notched and Brewerton Side-Notched projectile points, determining the cultural component as Archaic. Further testing was suggested of this site.

The Durika site (36WM279) is situated on a terrace. Recovered artifacts date to the Late Archaic component. The cultural material recovered from the survey included a number of Brewerton projectile points.

The Sulkosky Site #2 (36WM311, 312) is located on a terrace. The site is a multicomponent occupation and has evidence of Archaic and Middle Woodland components.

The Dunlap site (36WM359) is not stratified, but contained surface evidence of both Early and Middle Woodland components. The investigation conducted on this hillside slope produced artifacts that included chert debitage.

A survey of the hilltop above the Dunlap site (36WM360) produced evidence of an Early Woodland occupation, including pitted stones and lithic material.

The Sulkosky Site #3 (36WM523) is located on a bench overlooking Stony Run. It contained evidence of more intense prehistoric occupation than some of the sites previously discussed. Components include Late Archaic, Transitional, and Middle Woodland occupational periods, as identified by the recovery of Broadspear, Brewerton, and Jacks Reef projectile points. The Sulkosky Site #4 (36WM646) is a terraced, Early Archaic site that produced Lecroy projectile points. The Dunlap sites (36WM356, 357, 358, 360, 361) as a whole indicate Archaic, Early Woodland, and Middle Woodland occupations scattered across a bench and terrace.

Finally, the Durika sites (36WM277, 278, 279, 280, 281, 282, 283, 284) are situated on a terrace where an Archaic and Middle Woodland chronology is evident. Further testing is necessary.

Westmoreland County

Derry Township

ER 84-0744-129

Phase II archaeological testing was conducted on some of the Durika sites (36WM277, 36WM278, 36WM282, and 36WM283) by NPW Consultants in November 1988. The investigations were conducted in a surface mine permit area in which fifteen sites were initially located in the Phase I Survey. The major drainage in the area is Stony Run, and the topographic setting is a terrace. The sites yielded significant data concerning feature function. This data, along with diagnostic artifacts, contributed greatly to the cultural significance of the sites, and, therefore, made them candidates for Data Recovery. A Phase III Data Recovery of site 36WM278 was conducted in November 1992.

Westmoreland County

Derry Township

ER 87-0679-129

A Phase I survey was conducted of a Three-Well Drilling Site, at Lake Conemaugh, by GAI for Keystone Energy Oil and Gas, Inc. in January of 1987. The major drainage in the area is Stony Run, and the setting is a terrace. No new sites were found, although the previously recorded Gamelands site was found at the extreme northern edge of Well Pad 5, and the gas well was moved to avoid the site. Because no new sites were recorded, no further archaeological work was recommended.

Westmoreland County

Derry Township

ER 92-2969-129

A Phase I Survey was conducted for PennDot District 12-0 by Wilbur Smith Associates in October of 1992. The major drainage in the area is Stony Run, the topographic setting is a terrace, and the size of the survey project area covered 0.33 acres. No sites were found.

Westmoreland County

Derry Township

ER 96-0594-129

A Phase I cultural resource investigation at the Proposed Conemaugh Hydroelectric Project was conducted by the University of Pittsburgh Center for Cultural Resource Research for the Pennsylvania Renewable Resources Associates in October of 1994. The major drainage in the area is the Conemaugh River, and the site covers 2.8 acres. Due to a cut bank and exposed bedrock, no subsurface testing was necessary. No archaeological sites were found.

Westmoreland County

Derry Township

ER 96-2868-129

There were no sites found in this project area when the Gibson-Thomas Engineering Co. hired Thomas Baker in September 1997 to conduct a Phase I Survey. The area is located in Derry Township, Westmoreland County, in the Derry Quad. The drainage is Miller Run, and the project area was 76.2 m long and 6.1 m wide.

Westmoreland County

Fairfield Township

ER # 96-2132-129-c

A Phase I archaeological survey of the Ross Mountain park Timber Access Road and Tubmill Creek (Bridge 1) and Lick Run (Bridge 2) Bridge replacement project was conducted by Thomas Baker, hired by the Ross Mountain Club. The project area spanned .744648 hectares in St. Clair, Westmoreland Co. The topographic setting is mountainous terrain near the western base of Laurel Hill. The mountainous terrain includes sideslopes, terraces, and a floodplain. The major drainages in the area are Tubmill Creek and Lick Run. No archaeological sites were found.

Westmoreland County

Fairfield County

ER# not given

Hendricks Creek Bridge Replacement Project

Skelly and Loy conducted archaeological investigations for the Pennsylvania Department of Transportation Hendricks Creek Bridge Replacement Project. The project area, in Westmoreland County affected 0.28329 hectares. The major drainage in the area is the Conemaugh River. No archaeological sites were found.

Westmoreland County

St. Clair Township

ER# not given

A Phase I archaeological survey on the New Florence Water Filtration Plant was conducted by Heberling Associates for the Department of Environmental Protection in October of 1993. The investigations took place in St. Clair, Westmoreland County on 0.8094 hectares of land. The Conemaugh River is the main drainage for this area. No archaeological sites were found; no additional investigations were conducted.

Projects Located in Adjacent Townships in Westmoreland County

Westmoreland County

Derry Township

ER 81-1015-129

A Cultural Resource Survey of Loyalhanna Lake.

This survey was conducted by Christine Davis Consultants for the U.S. Army Corp. of Engineers. The survey covered an area of 162 hectares and consisted of both Phase I and II investigations. Of the twenty-six total sites found in the project area, only five were given ER numbers. These are: 36WM572, a Late Archaic site on a terrace whose integrity has been destroyed; 36WM571, located on a bench; 36WM570, also located on a terrace; 36WM560, a stratified site located on a terrace: and 36WM557, a late Archaic stratified site also located on a terrace. The report recommended that some of the sites need further testing.

Westmoreland County

Hempfield Township

ER# not given

Phase I Cultural Resource Survey of the Proposed Expansion of the New Stanton Water Pollution Control Plant

Christine Davis Consultants conducted a Phase I archaeological investigation of the Department of Environmental Resources Act 537 permit area. The project area covered 4.047 hectares of land in the drainage of the Youghiogheny River in Hempfield, Westmoreland Co. No sites were found; no additional archaeological investigations were conducted.

Table of Contents

Top of Page