Phase II and Phase III Archaeological Database and Inventory
18PR94 Indian Creek V
Diagnostic Artifact Data
Features
Summary
The Indian Creek site (also known as Indian Creek V or 18PR94) is an Archaic Period seasonal base camp near Beltsville in Prince George’s County, Maryland. One section of the site contained the remains of a well-preserved, short-term habitation from Early and Late Archaic Periods, used during the annual settlement migration in Maryland and Pennsylvania. It was frequently reoccupied to procure available plant resources from the surrounding floodplains and wetlands, resulting in one of the largest excavated Archaic Period Assemblages in Maryland. The site is located on multiple parcels of land, either part of the U.S. Department of Agriculture (USDA) Research Complex in Beltsville or owned by the Washington Metropolitan Area Transit Authority (WMATA). It is situated on low terraces adjacent to the broad, marshy floodplain of Indian Creek. The Indian Creek site is within the Bibb-Tidal Marsh soil association, which is generally characterized by poorly drained floodplain soils and marshes that are subject to frequent flooding.
The Indian Creek Site was first identified by Dennis Webb, a local collector, in 1972. Webb’s collection from 18PR94 included a bannerstone fragment, a bifurcate-based point, and various Archaic points made of quartz and rhyolite. The first professional attention paid to 18PR94 was a brief site visit in 1978. At that time, a survey was being conducted within the corridor of a proposed water main along the outer perimeter of I-495. While the site was determined to be well outside the project impact area, the investigators did observe a soapstone fragment and various quartzite, quartz, and rhyolite flakes on the surface of the site. The investigators suggested that the primary environmental features that would have attracted prehistoric settlement in the area were the presence of secondary cobble and gravel deposits, and broad floodplain and marsh habitats.
Major excavation work was first conducted at Indian Creek in late 1986 and early 1987, when a Phase I survey examined the area for a proposed public transit infrastructure improvement. At that time, WMATA was evaluating the location of their Greenbelt railcar storage and maintenance facility. The Phase I work involved an intensive survey of the entire 70 acre area, including surface collection, systematic shovel testing, backhoe trenching, and limited test excavations. Fieldwork began with a transit survey of the project area, to establish a grid for horizontal provenience control. First, survey control stakes were set at the intersection points of a 91.44 X 91.44 meter (300 X 300 ft) grid, which was then divided into 22.86 meter (75 ft) grid blocks. All visible prehistoric material was then collected within the 22.86 meter grid blocks. Surface collection was confined to the cultivated portion of the survey area, and the majority of this area had been planted in alfalfa, which afforded fair ground surface visibility. A soybean field in the extreme southern end of the survey area provided much less favorable ground visbility.
The surface collection was followed by shovel testing at 22.86 meter intervals, with tests placed at the grid intersection points used for the surface collection. With the exception of swamp, landfill, or otherwise disturbed areas, the shovel testing program covered the entire project area, including the area that previously had been surface collected. Shovel test pits (STPs) were circular in plan and approximately 30.48 cm (1 ft) in diameter. These tests were excavated according to natural strata and all soils were screened through hardware cloth. All shovel tests were advanced at least 9-10 cm into subsoil, depending on the recovery of cultural material. The survey work demonstrated a nearly continuous lithic scatter across the project area.
Based on the research findings from the surface collection and shovel testing, four areas were identified that exhibited relatively greater concentrations of prehistoric material. These areas were then subject to more intensive subsurface testing, including additional shovel tests and small test squares. Within the areas identified for more intensive testing , STPs were excavated at 4.57 meter (15 ft) intervals within the established grid system. Then, after the areas of greatest concentration were defined, test units were excavated to determine the depth of the deposits, the presence of cultural material in sub-plowzone contexts, and the presence of features. Like the STPs, the test squares were excavated according to natural stratigraphy. Except for the plowzone, soil strata were subdivided into arbitrary 9.14 cm (.3 ft) levels for more precise vertical control, and excavation proceeded until two successive sterile levels had been removed. Vertical profiles and photographs were made of each unit. A total of 18 test squares were excavated during the initial survey phase of the project. For the most part, the units were placed within the four areas of concentration, although two units were placed in other areas of interest. With one exception, all test squares measured 91.44 X 91.44 cm (3 X 3 ft) in plan.
Excavation of the test squares was followed by the excavation of four backhoe trenches. These trenches were placed at selected locations in order to examine deep stratigraphic profiles and to test buried cultural deposits or features. The backhoe trenches were placed in areas of archeological interest, but generally at the margins of the cultivated area in order to minimize crop damage. No attempt was made to screen the spoil from the backhoe trenches before the units were backfilled. The trenches were generally around 61 cm wide, and they ranged from approximately 4 to 5 meters in length.
Phase II testing was then carried out in a portion of the site identified as “Area 3”, which had demonstrated the presence of apparently well-preserved prehistoric deposits. The Phase II work included the hand excavation of twenty 1.524 X 1.524 meter (5 X 5 ft) squares, comprising approximately a 1% sample of the site area as defined by the shovel tests. The grid established for testing was aligned with the survey baselines, but a new system of north and east grid coordinates was established. Excavation of the test squares proceeded according to arbitrary 9.14 cm levels within strata. In undisturbed subsoils, levels were divided and excavated by 76.2 X 76.2 cm (2.5 X 2.5 ft) quadrants to permit a more fine-grained horizontal provenience control. It was not immediately apparent that the surface soils had been plowed, and the A-horizon soils in the first several units were taken out in arbitrary levels. After it became obvious that the surface soils had been cultivated, the plowzone was removed as a unit, rather than by levels and quadrants.
Flotation samples were generally taken from each undisturbed level that contained prehistoric material, as well as from feature contexts. The standard procedure for the taking of flotation samples was to remove a one quart sample from the northeast corner of each unit, to provide a more or less continuous column sample. After air-drying, all flotation samples were processed through a water separation system. Visible cultural material was extracted from the processed samples prior to submission of the samples to an analyst.
The Phase II work in Area 3 revealed activity areas and four prehistoric features that contained fire-cracked rocks. Approximately, 70% of the materials were recovered from undisturbed subsoil contexts. Specialized tests indicated blood residues on lithics, and flotation studies revealed relatively well-preserved floral remains. Based on these findings, Phase III data recovery was recommended at the site.
Phase III work was conducted between January and April of 1989 and entailed the excavation of 124 test units (1.524 X 1.524 meters each) situated in 6 areas of block excavation in Area 3. Combined with the total area exposed during Phase I and Phase II work, the excavated sample from Area 3 encompasses an area of approximately 334.5 square meters. The Phase III excavation strategy was based on a sampling plan that included three principal components: 1) scattered, exploratory units within previously untested areas of the site, 2) block areas centered on features identified during the Phase II testing program (3 fire-cracked rock concentrations, two lithic work areas, and a unit where 2 Early Archaic points were recovered) and 3) block areas centered around newly identified features. The exploratory units were scattered throughout the site to provide a spatially more representative sample, emphasizing the core area of the site as defined during the preceding investigations.
The field methods used during the data recovery program were virtually identical to those of the Phase II program, to permit integration of the results of both phases of work. Test units were excavated by quadrants once the plowzone had been removed. Excavation proceeded in arbitrary 9.14 cm levels within strata screened through hardware cloth. When possible, tools in subsoil levels were point-plotted and assigned artifact numbers to enable reconstruction of activity areas. Features and soil profiles were drawn to scale and photographed. Soil samples were removed for flotation processing, soil chemistry analysis, and pollen analysis. Flotation samples were taken from feature contexts and from subsoil levels of other units distributed across the site for subsequent processing and analysis. The standard procedure for the taking of flotation samples was to remove a two liter soil sample from the northeast corner of a selected unit, to provide a more or less continuous column sample. Off-site soil samples were also obtained for use as control samples. Soil samples intended for chemical analysis were taken primarily from feature contexts, but these also included non-feature contexts and off-site contexts. Small soil samples (25-30 ml) were also retained for pollen analysis; these were obtained from feature contexts and from the dried flotation samples. After processing, cultural material was picked from the flotation samples and catalogued with the remainder of the artifact collection.
Historic material was recovered during the Phase II excavations but it was determined to be representative of dumping activity rather than residential refuse. During Phase III fieldwork, historic material was noted and discarded in the field. The historic remains will not be discussed further here.
Lithic tools, debitage, and fire-cracked rock make up the entire assemblage of prehistoric artifacts recovered at the Indian Creek V site during all phases of archeological work prior to construction of the Greenbelt railcar storage and maintenance facility. These 59,665 objects reflect the various activities that occurred at 18PR94, including tool manufacture and maintenance and the procurement and processing of food stuffs. The combined Phase I-III archeological investigations revealed 35 features, including 31 fire-cracked rock concentrations, two lithic flaking/knapping areas, a charcoal concentration, and a cache of unfinished quartzite bifaces. Activity areas, representing a full range of lithic production, appear to have been focused around the fire-cracked rock features. Artifacts diagnostic of the Late Archaic were most abundant in the plow zone and first subsoil level, while Early Archaic materials were mostly found in the first and second subsoil levels.
The artifact assemblage from Phases I-III consisted of 568 bifaces, 18 unifacial tools, 228 cores, 76 modified flakes, 34,313 other flakes, 639 chunks, 25 hammerstones, 7 hammerstone/multi-use cobble tools, 1 anvil, 2 steatite vessel fragments, a groundstone ornament fragment, 8 miscellaneous groundstone tools (3 pestles, 2 manos, 2 metates, and an abrading stone), 4 unidentifiable groundstone object fragments, 2 limonite chunks with ground surfaces (for producing pigment), 23,770 pieces of fire-cracked rock, and 2 probable manuports (pieces of petrified wood). The possible ornament, limonite chunks, and petrified wood were also counted in the table above as “uncommon objects”. Identifiable projectile point types within the biface assemblage included 2 Palmer points, 6 Kirk stemmed, 2 Kirk corner-notched, 3 St. Albans, 8 LeCroys, 9 Kanawha, 1 Stanly, 3 Morrow Mountain II points, 11 probable Otter Creek points (could be Brewertons as well), 38 Vernons, 14 Clagett points, 26 Savannah River points, 5 Holmes points, 4 Lackawaxens, 4 Calverts, and 63 un-typed projectile points. Other bifacial tools included 6 drills, 26 early stage bifaces, 56 middle stage bifaces, 28 late stage bifaces, and 150 unidentified bifaces. Lithic raw materials were primarily quartz and quartzite, with lesser quantities of rhyolite, sandy chert, chert, chalcedony, jasper, and argillite.
Two types of residue analysis were conducted on lithic tools (primarily bifaces). The first test assessed the presence of blood, which was positive on 49 of the bifaces. Family level testing on those 49 bifaces then suggested that deer, elk, and various small game animals were being hunted at Indian Creek. The high acidity of the soils caused prehistoric faunal materials to disintegrate, resulting in none being recovered. Thus, the blood residue analysis provides an important, otherwise missing, puzzle piece for discussions of site subsistence.
An assemblage of seeds, nutshell fragments, macrospores, and small charred wood fragments was recovered from flotation samples at the Indian Creek V site. Over 10,000 fragments from many different plant species were recovered from subsoil levels and features, representing a wide variety of fruit, tubers, starchy seeds, nuts, shoots, and leaves. A detailed description of the flotation assemblage is available in the linked ethnobotany profile for 18PR94. These plants would have been used as a dietary resource and also as medicines, smoking material, and insect repellant. In addition to the flotation data, a pollen core from a nearby peat deposit provided a vegetation record for the terminal Pleistocene and Holocene epochs, allowing a detailed environmental reconstruction.
The Indian Creek V site spans the entire Archaic period, with two intervals of intensive site occupation. The initial intensive occupation is represented by the Palmer, Kirk and bifurcate-base points and based on radiocarbon dates from other sites spans the period from ca. 7,800 to 5,300 BC. This initial period of occupation fits within the Early Archaic timeframe. The second period of intensive occupation was in the Late Archaic and is represented by the Morrow Mountain, Otter Creek, Vernon, Clagett, Savannah River, Lackawaxen, Holmes, and Calvert points. Radiocarbon dates from other sites place these points in the timeframe from 4,000 to 1,000 BC.
Considering the great age of the site, preservation at 18PR94, at least in “Area 3”, was excellent. While faunal remains were largely obliterated by the acidity of the soils, charred plant remains were recovered through flotation, which provide one of the largest collections of prehistoric wetland resources in Maryland. The evidence points to seasonal use of the site to exploit the resources of the nearby floodplains. Intact features, activity areas, and diagnostic artifacts were also encountered. Most Archaic period sites do not bear evidence of distinct activity areas, making 18PR94 a unique resource for yet another reason.
Following data recovery operations in the late 1980s and early 1990s, the WMATA railcar facility was constructed atop major portions of the site. Most, if not all, of Area 3 would have been destroyed during construction work. However, questions remained about other portions of the site (specifically to the northwest) which were less-well documented during the aforementioned studies. In 1992, a survey was conducted because portions of the site and surrounding area within the USDA’s Beltsville Agricultural Research Complex were being considered for the development of an office/research facility that would include at least 2 large buildings, extensive parking lots, and the associated infrastructure to support them. Such development would have extensive impacts on any archeological resources in the area and survey was required under Section 106 of the NHPA of 1966.
Site 18PR94 was examined during Phase I as part of the 26 acre Survey Area D. Methods employed in the testing of Survey Area D included a systematic surface collection conducted along transects spaced at 5 meter intervals, a second surface collection along transects spaced at 1 meter intervals (specifically within the 18PR94 site area), the excavation of 260 shovel tests placed at 20 meter intervals and the excavation of a 1 X 1 m test unit (also within 18PR94). All subsurface tests were excavated at least 10 cm into “culturally” sterile subsoil.
The 1992 Phase I survey yielded a total of 296 artifacts from an area at the site approximately 150 meters wide by 260 meters long. This artifact assemblage consisted of debitage (93%), formal stone tools (5%), and fire-cracked rock (2%). The complete assemblage included 10 bifaces, 2 unifaces, 4 cores, 235 quartzite flakes, 34 quartz flakes, 1 rhyolite flake, 3 hammerstones, and 7 fragments of fire-cracked rock. Twenty percent of the artifacts were recovered from the surface of the site, with the remainder coming from the plowzone level of STPs and the single test unit excavated adjacent to one of the most productive STPs.
Based on these findings Phase II testing was recommended on the USDA-owned portion of 18PR94. Although “significance” was not deemed to be an issue in this case, there were a number of technical issues which needed to be addressed, including exact site boundaries, stratigraphic integrity (horizontal and vertical), and the research potential of the deposits. In the Fall of 1992, researchers returned to the site to conduct the Phase II work.
The Phase II testing in 1992 included the systematic surface collection of 90 whole or partial 20 m collection blocks and 93 ten meter blocks. This was followed by the machine-stripping of 9 blocks, providing 450 square meters of horizontal exposure and the excavation of eight 1 X 1 m test units down to an average depth of 60 cm.
A total of 404 artifacts were recovered in the course of testing work, including 39 tools, 314 flakes, and 57 fire-cracked rocks. The complete 1992 Phase II assemblage consisted of 22 bifaces (15 quartzite, 7 quartz), 6 other tools, 18 cores (15 quartz, 3 quartzite), 74 chunks (71 quartz, 3 quartzite), 222 flakes (155 quartzite, 63 quartz, 4 other), 11 hammerstones, and 51 pieces of fire-cracked rock. The surface collection yielded 380 of these artifacts and the remaining 24 artifacts were all recovered at or within 5 cm of the plowzone/subsoil interface. Diagnostic artifacts include 5 Bare Island-like points, 2 points of general Archaic form, 1 corner-notched point of Late Archaic/Early Woodland form, and 3 triangular points.
Subsurface investigations failed to produce any evidence of features or stratified deposits in this area. A total of 15 soil anomalies identified in the course of backhoe stripping were cross-sectioned with a trowel, and all of them turned out to be non-cultural and comprised of tree stains and animal burrows. Test unit excavation also failed to yield evidence for in situ deposits and produced only a few artifacts just at or below the interface of the plowzone and the subsoil, and are perforce suspect in terms of integrity.
Phase II research in 1992 revealed that the portions of Site 18PR94 on USDA property did not retain sufficient integrity to address research questions regarding Maryland prehistory. Subsequent construction of the aforementioned office complex would have largely destroyed this portion of the site. As has already been noted, significant portions of 18PR94 on WMATA property were also destroyed following data recovery operations. Thus, it is unlikely that significant portion of the Indian Creek V site remain preserved and its research potential is largely exhausted.
|
