Phase II and Phase III Archaeological Database and Inventory
18ST570 Thomas Point
Diagnostic Artifact Data
Features
Summary
Thomas Point (18ST570) is a multicomponent prehistoric site with short-term occupations starting in the Late Archaic period, and includes two Middle to Late Woodland period shell middens. It is located on the lower Patuxent River in St. Mary’s County. Thomas Point is situated on a promontory of the Myrtle Point peninsula. Myrtle Point is the 20th century place name for the narrow, flat headland bound by the deep outlet channels of Cuckold Creek to the north and Mill Creek to the west, and Little Kingston Creek to the southeast. Most of the land surface is level to gently sloping from 4-7 m above the river, and is vegetated by tangled thickets of mature red cedar, pines, holly, wild grape, and a verdant understory of escaped English ivy, poison ivy, and wax myrtle. Two drainages have cut through the former eastern embankments to drain into the tidal marsh. One of these drainages is a ravine, which bisects the site. Soils in the site area are Matapeake silt loam and Sassafras sandy loam.
The lower Patuxent River was an area occupied by prehistoric peoples for more than 9000 years. Archeological evidence suggests that populations grew rapidly in Southern Maryland during the Early Archaic period (7,500-6,000 BC). Increasing populations and resource diversity occurred during the Late Archaic period in the lower Patuxent. The earliest dated oyster shell middens along the lower Potomac River are dated to that period. The Early Woodland (1,000 BC – AD 200) is marked by increasing sedentism and the appearance of ceramics, particularly of Accokeek pottery in the lower Patuxent. The Middle Woodland period in the Patuxent drainage is characterized by the appearance of large special purpose sites, a rise in intra-component artifact diversity, and evidence of inter-regional exchange networks. By the Late Woodland period, and into the 17th century, there appeared to be trends toward residential sedentism accompanied by political nucleation and status differentiation throughout the Chesapeake estuary. The exception seems to be in the Patuxent drainage and north along the western shore of the Chesapeake where social ranking and stratification appear to be less prevalent.
In January 1642/43, Nicholas Harvey was granted the patented for 1,000 acres of land called St. Joseph’s Manor. The Manor was originally located within Mattapanient Hundred, lands held by the Jesuits. Harvey and his family fled to Virginia during the “time of troubles” precipitated by Richard Ingle’s conquest of the colony in 1645. At about that time, it has been suggested that raiders burned Harvey’s house down. The records indicate that after Harvey’s death, a man named Edward Lloyd purchased St. Joseph’s Manor from Thomas Green, who had married Harvey’s widow. There are no further mentions of Lloyd, Green, or the widow. Documents suggest that ownership of the Manor passed to Harvey’s daughter, Frances, who had married George Beckwith. The Beckwiths developed the property into a tobacco plantation and rebuilt the dwelling house burned down by Ingle’s men. The Beckwiths also kept and maintained a ferry boat used for crossing to and from Point Patience. By the end of 1659, approximately 425 acres on St. Joseph’s Manor had been subdivided and sold. By the 1720’s, all of the Manor lands had been sold out of the Harvey and Beckwith families. For several years following the Beckwiths’ deaths, the land remained unpatented. Then in 1727, George Plater II obtained the patent for St. John’s Manor totaling 1,250 acres. Sometime later, a 50 acre portion of the Manor was sold to Samuel Jenifer and became known as Harveytown, the early colonial port which functioned from ca. the 1660’s to the early 18th century. The community of Harveytown was located in the settlement called Harvey Hundred. By 1798, a total of 34 individuals or estates owned land in Harvey Hundred. Early 19th century maps identified the peninsula as Town Point (now located at the southern abutment of the Thomas Johnson Bridge to the east) and Thomas Point (the northeastern most promontory of the peninsula). The record taken of Harvey Hundred in 1810 listed 5 owners of parts of St. Joseph’s Manor. From the early 19th century through the mid-20th century the land had several owners who made various improvements to the property. The US Coast and Geodetic Survey map of 1848 indicated that the area around site 18ST570 was deforested and under cultivation during the 19th century. By 1896, the site area had been transformed to pasture. Reforestation apparently began only in the 1930’s. In 1966, a portion of the Manor land that included the site area was sold to Patuxent River Farms, Inc., who then sold it to Route 347 Realty Corporation in 1986.
The Thomas Point site was first identified during a Phase I archaeological survey of the 211-acre Patuxent River Farms/Route 347 Realty Corporation property conducted from the summer of 1986 through the spring of 1987. Work was conducted ahead of proposed construction of an intensive townhouse, single family residential, and recreational development. The purpose of the survey was to locate historic and prehistoric sites in the areas of potential impact. The preliminary survey identified 37 sites, including 9 which were determined eligible for the National Register of Historic Places. Two eligible sites could not be avoided during construction, the Thomas Point site (18ST570) and the nearby Myrtle Point site (18ST569).
Field methods included dividing the property into 13 areas defined on the basis of topography, ground cover, and applicable survey techniques. The areas were assigned non-consecutive letters A through P. Site 18ST570 was located across Areas E and F in the northeastern portion of the study zone. Shovel test pits (STP) were excavated at 10 m intervals along transects laid out 30 m apart. STP sampling started along a central baseline in Area E which began along the fence line bordering the Area P field. In total, 274 STPs were dug in Areas E and F. Of those, approximately 113 STPs relate to site 18ST570.
The site was identified as a concentrated distribution of prehistoric artifacts along the eastern edges of Area E and F facing the modern tidal marsh. It was determined to be a large shell midden and campsite that extended for about 220 m along the edge of the bank overlooking the tidal marsh, and for about 160 m west/inland of the marsh bank. The highest shell concentrations occurred in STPs E26, 29, 33, F4 and 7, close to the marsh bank. Nearly all of the shell was broken. It was recovered from the plowzone which measured from 9 to 28 cm deep.
At STP E33, an area of whole layered shell was encountered. This was interpreted as an area of preserved midden directly south of the ravine that bisected the site. Layered shell was observed to a depth of 32 cm, and shell continued deeper. Late Woodland pottery was recovered from the STP. Within the site, the distribution of debitage suggested some degree of spatial or functional differentiation. The ravine that bisects the main area of oyster and artifact distribution has been progressively filled by colluvial wash from the surrounding higher terrace. Quartz debitage was less prevalent in the ravine but was more common on flat areas north and south of the drainage. This indicated the possibility that tool production and oyster processing occurred in different areas of the site.
Colluvial wash apparently buried a second area of intact shell midden at the drainage outlet into the brackish marsh. A deep unmapped test of the marsh midden encountered excellent preservation of snails, small bone fragments, and nut hulls in a 563 milliliter flotation sample. This wetland portion of the Thomas Point site was not to be impacted by development and was nominated to the National Register. The midden on the upper terrace was to be greatly impacted by proposed construction plans, therefore, Phase II and Phase III excavations were recommended for the site.
The artifact inventory for Areas E and F did not provide a breakdown of site and non-site materials. According to the site boundaries established by the excavators, only 113 STPs related directly to 18ST570. Therefore, only the artifacts from those STPs, rather than from across the entirety of Areas E and F, are presented here and in the table above. A total of 137 prehistoric artifacts were retrieved from the site. There were 87 pieces of debitage (62 quartz flakes, 10 quartzite flakes, 1 rhyolite flake, 15 jasper/chert flakes). There was 1 projectile point fragment, a Holmes point which is typically dated to the Late Archaic period. There were 19 fire cracked rocks (FCR). A total of 30 ceramic sherds were found including 1 Pope’s Creek, 1 Accokeek, 1 quartz-tempered, 17 Late Woodland type shell-tempered, 2 grit-tempered, 1 sand- and grit-tempered, 3 sand- and shell-tempered, and 4 unidentified tempered sherds. The shell-tempered sherds most likely correspond to the Townsend series and are cataloged as such in the table above. The excavators indicated that assigning the shell tempered ceramics to the Mockley ware series was not preferable due to the fact that Selby Bay points and rhyolite tools, artifacts generally associated with Mockley, were not found at the site.
A historic component was identified during the investigations in Areas E and F as a concentration within the southern portion of 18ST570. It was designated with its own site number, 18ST580, and called ‘Abell’s Point’ after a former landowner. 18ST580 was not discussed in the text of the original report and in subsequent excavations at prehistoric site 18ST570, no further mention was made of 18ST580. Historic artifacts were recovered during the later Phase II/III study and were recorded in the artifact inventory as part of the overall site assemblage. Therefore, the artifacts from 18ST580 will be considered here as a component of 18ST570. A total of 72 historic artifacts were collected during the Phase I. There were 52 architectural items (30 bricks fragments, 1 cut nail, 3 wire nails, 1 unidentified nail, 4 fragments of mortar, 13 asphalt shingles). There was 1 clothing item (a brass buckle). There were 14 kitchen related items (3 pieces of table glass, 2 lead-glazed earthenware, 1 tin-glazed earthenware, 2 unidentified earthenware, 1 whiteware, 2 creamware, 1 pearlware, 1 Staffordshire ware, 1 brown salt-glazed stoneware, 1 possible Delftware sherd). There were 2 miscellaneous items (1 unidentified metal object, 1 unidentified object). There were 3 tobacco items (white clay pipe stem fragments). The historic component was determined to be a 19th century artifact concentration possibly representing a non-extant structure.
It was determined that the proposed residential development would severely impact the site. The site was determined culturally significant for its potential to yield intact subsurface features. Therefore, in 1987/88 Phase II and Phase III investigations were conducted at the Thomas Point site. The goals of the investigations were to determine the site boundaries, locate any undisturbed buried components and provide a representative sample of the stratigraphy and artifact assemblages from the area of the site which was planned for development. Phase II study consisted of the excavation of 48 one-m² units placed at 20 m intervals across the site. All soils were screened through ¼” mesh and all artifacts were retained except oyster shell, which was weighed in the field and discarded. Phase III efforts focused on the investigation of the shell midden through the excavation of 76 one-m² units which transected the midden and tested the surrounding area. All soils were water-screened through 3 grades of screen. Six 25-cm column samples, spaced at 2 m intervals within the midden trench system, were excavated separately. All fill from the column samples was processed with a flotation system. Oyster shell from the samples and the feature contexts was analyzed for size, shape, predation and seasonal indicators. Oyster shell from the midden was sorted and discarded in the field. Soil samples were collected from 3 locations at the site. These included the upper terrace backhoe trench (Trench A), the upper terrace midden area (120S/58W), and a backhoe trench on the lower terrace gully position (Trench B). Three transects were also selected to determine the extent of changes in the pH with distance from the major portion of the shell midden. The data is available in Appendix C of the original report.
An analysis of the faunal remains collected during the 1987/88 study was conducted. The remains were first divided into 2 broad categories: oyster shell, and non-oyster shell, chitin and bone. Samples (n=12) of 25 whole oyster shells were selected from column samples and features. Results of the oyster shell analysis indicated that the oysters were collected primarily during 2 seasons: June-August and November-April.
Analysis of the non-oyster shell, chitin and bone indicated that 85% of the NISP were comprised of land snails, marine gastropods, scaphopods, and pelecypods which entered the archeological record as incidental inclusions or by way of attachment to oysters. There were also 2,533 other faunal remains that represent food procurement/processing at the site including 78 fish bone fragments, 35 turtle, 99 deer, 319 bird remains, 39 identifiable small mammal, and 1,963 unidentified mammal bones. These were included in the ‘unmodified faunal’ category in the table above. The full report is available in Appendix D of the original report.
The archeobotanical analysis of the plant remains from the site consisted of flotation samples taken from each stratum and level of the column samples and from each feature, 33 samples in total. A total of 677 remains from various species of nuts, cultigens, fruits, and commensals or miscellaneous species were identified. There were 53 maize remains (8 cupule and 45 kernel), 48 hickory shell and 108 acorn shell fragments, 31 fruit seeds (23 persimmon, 1 plum, 1 plum/cherry, 2 elderberry, 2 blueberry, 2 grape), 437 commensals and miscellaneous (3 chenopodium, 1 cheno/am, 2 knotweed, 2 pokeweed, 3 bedstraw, 1 grass, 2 juniper, 47 unidentified striated, and 9 unidentified seeds, 354 unidentifiable seeds and shell fragments, and 13 unknown). Identifications of potential food remains suggest a subsistence strategy which combined foraging with small scale plant cultivation during at least the summer and fall seasons. The cultivation of maize appears to have been practiced as part of a foraging economy which exploited a diverse array of fruits and seeds and relied heavily on acorn mast and hickory nuts. The full report is available in Appendix E of the original report.
During the Phase II study, 46 of the 48 units were placed on the upper terrace. Most of the artifacts were recovered from the A-horizon which was incorporated into the historic plowzone, usually less than 20 cm thick. Two units excavated on the lower terrace during the Phase II investigation sampled the small shell midden. The horizontal extent of the midden was determined by a soil probe sample, spaced along a 1 m² grid. Along the lower terrace, which was less than 1 m above the high-tide level, excavation encountered ground water seepage in the excavation just above the shell bearing stratum. No further testing was conducted in the location as it was not immediately threatened by the proposed development.
In addition to the upper and lower terrace middens, 2 additional features were encountered during the Phase II study. A historic trench was located at the contact of the B horizon in Unit 1S/67W and Unit 1S/68W. As that area of the site was not scheduled for development, no further excavations were conducted. The second feature (Feature 5) was identified in the corner of Unit 101S/43W. It was a prehistoric basin shaped pit that contained an abundance of oyster shell and other food remains. Excavations were extended during Phase III around Feature 5 revealing several other prehistoric pits (Features 9, 10 and 11).
Based on the Phase II study, 4 areas (Areas A-D) of artifact concentrations were identified. A discussion of those findings in this synthesis is tentative because the references to the particular areas in the original report are vague, and in some places contradictory. Generally, the lower terrace midden had low frequencies of all lithic artifact classes and appears to have been the location of domestic activity focused on oyster processing. The upper terrace had more early-stage biface flaking debris than expected and appeared to be an area not only used for shellfish processing but also for biface production from raw materials likely procured from the shoreline. Two areas on the upper terrace seemed to represent specific activity loci. The first is an area of habitation where limited domestic activities occurred and where lithic reduction focused on the production of finished tools. The second area, which was adjacent to the shell midden, appears to have been utilized for lithic reduction although there was a notable absence of bifaces.
A total of 2,548 prehistoric artifacts were collected during the Phase II investigations. Totals were taken from Appendix A, artifact tabulations, in the original report, except for flake tools and hammer/anvil counts, which were taken from a table within the text as they were not listed in the Appendix artifact catalog. There were 1,300 pieces of debitage including cores and core debris. There were 17 flake tools. There were 37 biface tools 14 of which were identifiable projectile points including 11 Potomac/Yadkin, 1 Madison/Hamilton, 1 Vernon or Calvert (listed under Vernon in the table above), and 1 Early Woodland narrow lanceolate. In the text in the original report “10 Levanna or Madison type” points were mentioned. No Levanna were identified in the Appendix but there were Potomac/Yadkin and Madison/ Hamilton types. Those are listed under Madison/Potomac in the table above. There were 15 use-modified stone tools, either hammers or anvils, which were recorded in the ‘other lithics’ category in the table above. There were 62 fire-cracked rocks. Quartz and quartzite make up the bulk of the materials used in tool production. A large amount (n=309) of chert debitage (flake and core) suggested that there would be a proportionally high number of chert tools, but this was not indicated in the assemblage. Only 16 chert biface and flake tools were recovered from the site. A small number of rhyolite objects (n=22) were also recovered. A total of 1,117 prehistoric ceramics were recovered during the Phase II investigations. There were 2 Accokeek plain, 91 Rappahannock (34 plain, 57 fabric impressed) (recorded as Townsend in the table above), 2 Sullivan Cove cord marked, 3 Mockley cord marked, 5 Mockley plain, 4 Townsend cord marked, 3 Townsend incised (in the text these are referred to as Rappahannock incised), 1 Yeocomico, 2 Potomac Creek plain, 3 unidentified sand-tempered, 1 unidentified shell and crushed quartz-tempered, and 1 unidentified grit tempered, and 999 unidentified shell-tempered sherds. A total of 647 oyster shell fragments were recorded during Phase II. None were retained therefore they are not included in the unmodified faunal box in the table above.
A total of 85 historic items were retrieved from the site during the Phase II investigations. These totals were taken from Appendix A, artifact tabulations, in the original report. In general, the artifacts were scattered across the site. However, there was a slight clustering of nails and metal in the northern portion of the site, suggesting that in the past a structure may have existed in the location or nearby, and a clustering of brick and window glass in the southern portion of the site. There were 53 architectural items including 35 brick fragments, 6 window glass shards, 1 wrought nail, 8 cut nails, 3 wire nails. There was 1 clothing item (1 wooden button). There were 12 kitchen-related items including 2 bottle glass fragment, 4 earthenware, 3 lead-glazed earthenware, 1 creamware, 1 salt-glazed stoneware (unclear if this is white salt-glazed), 1 tin-glazed earthenware. There were 8 miscellaneous items, all unidentified metal objects. There were 10 tobacco items including 9 white clay pipe fragments and 1 terra cotta pipe fragment. There was 1 arms-related item (a gun flint).
During the Phase III component, 76 units were excavated on the upper terrace in the location of the shell midden, 48 in the shell zone and 28 adjacent to the shell. The plowzone was dry screened through ¼” mesh and the remaining strata were water-screened. A single backhoe trench was excavated at the terrace shoulder near the shell zone. The stratigraphy consisted of 4 strata. The shell layer was identified in stratum II and varied in depth from 10-60 cm below the ground surface. Six features (including Feature 5 identified in Phase II) were encountered in the general area of the midden. Features 5, 9, 10 and 11 were a cluster of shallow, basin shaped pits situated about 10-12 m north and east of the main shell concentration. They were associated with a high frequency of deer bone, oyster and clam suggesting that they were used as roasting pits. Feature 9 (F9) contained a concentration of FCR, oyster shell and charcoal. Feature 11 (F11) contained no FCR but did contain numerous soft-shelled clam clustered in a dense charcoal lens. Charcoal samples were collected from F9 and F11 for radiocarbon dating. Feature 10 was heavily plow disturbed. Features 12 and 13 (F12, F13) were identified beneath the shell midden in the southwestern portion of the site. They appeared to be tree-fall depressions which were utilized as refuse pits. A sample of carbonized wood charcoal was collected from F12 for radiocarbon dating.
Two additional units (61S/20W and 81S/20W) were excavated during Phase III in the marshy terrain at the foot of the terrace. Unit 61S/20W exposed an oyster shell midden buried beneath approximately 70 cm of colluvium.
Five radiocarbon dates were obtained from carbonized plant material recovered from the upper terrace shell midden area and from 1 sample from the lower midden. Two samples were acquired from small refuse-filled pits (F9 and F11) adjacent to the midden. F9 produced a calibrated date range (2 sigma) of AD 936-1223 and F11 produced a date range of AD 1219-1333. The sample recovered from F12 below the shell midden produced a calibrated date range of AD 1016-1272. One of the column samples taken from the shell midden matrix (Stratum II, Level 2) fell into the date ranges of AD 1116-1272. All of these date ranges fall comfortably within the Late Woodland period and correspond to the artifacts recovered from these locations. Another of the column samples taken from the shell midden matrix (Stratum II, Level 1) fell into the date ranges of AD 861-1042, which falls more closely within the late Middle Woodland period. It is unclear why the upper stratum would provide an earlier date than the stratum below. The presence of late Middle Woodland Mockley ceramics at the site does suggest occupation at that time. A sixth date range of AD 314-614 was obtained from wood charcoal from a column sample (Stratum II, Levels 9-10) in the shell deposit at the foot of the terrace. The shell deposit in that area was buried under almost a meter of colluvium and Levels 9 and 10 were below the water table. The first sample retrieved was too small so a second sample was acquired at a later time and added to the first. The date range falls within the Middle Woodland period and the artifacts recovered from the associated unit included several small shell-tempered sherds, possibly Mockley ware.
A total of 5,847 prehistoric items were collected during the Phase III investigations. Totals were taken from Appendix B, artifact tabulations, in the original report, except for flake tools which was taken from a table within the text as they were not listed in the Appendix artifact catalog. There were 1,558 pieces of debitage including cores and core debris. There were 11 flake tools. There were 55 biface tools of which 14 were identifiable projectile points including 7 Potomac/Yadkin, 4 Madison/Hamilton, 1 Bare Island/Poplar Island, 1 Levanna, 1 unidentified Late Archaic broad blade. There were 9 use-modified stone tools, either hammers or anvils, which were recorded in the ‘other lithics’ category in the table above. There were 421 fire-cracked rocks. As noted in the Phase II discussion, quartz and quartzite make up the bulk of the materials used in tool production. Chert debitage (flake and core) and a small number of rhyolite objects were also recovered. A total of 3,793 prehistoric ceramics were recovered during the Phase III investigations. There were 1 Accokeek plain, 435 Rappahannock (146 plain, 289 fabric impressed), 134 Sullivan Cove cord marked, 92 Mockley cord marked, 7 Mockley plain, 4 Mockley net-impressed, 3 Moyaone cord marked, 4 Moyaone plain, 7 Townsend cord marked, 45 Townsend incised (in the text these are referred to as Rappahannock incised), 1 Townsend punctate, 4 Potomac Creek plain, 10 unidentified sand-tempered, 3 unidentified shell and crushed quartz-tempered, and 3,043 unidentified shell-tempered sherds.
No historic artifacts were recorded in the inventory for the Phase III investigation.
An intensive survey of 18ST570 was undertaken in 1990 ahead of construction of the proposed Washington Gas Light Company Patuxent Pipeline installation. The original plans indicated that construction would impact the site. Fieldwork consisted of a pedestrian survey and surface reconnaissance and/or subsurface sampling of all previously undisturbed portions of the project area. A total of 16 shovel test pits were excavated at 10 m intervals at the site, which was located in Test Unit 5. It was decided that Thomas Point would be avoided by redesigning the pipeline right-of-way and no further work was suggested for the site.
The excavation of 7 STPs located just to the northwest outside the designated site boundaries of 18ST570 yielded prehistoric (and a few historic) materials. The excavators postulated that they may be related to 18ST570 but did not definitively state so. They may also be related to 18ST569, a prehistoric site with a dense Late Woodland component located about 65 m northwest of 18ST570. Given this, those artifacts were not included in the totals in the table above.
A total of 394 prehistoric artifacts were recovered during the 1990 testing at the site. There were 10 pieces of debitage (7 quartz, 1 quartzite, 2 chert). There were 4 unidentified ceramic sherds (2 quartz tempered sherds, 1 mixed tempered sherd, and 1 limestone tempered sherd). A total of 379 oyster shell fragments were recorded along with 1 animal bone fragment.
A few historic items (n=5) were also collected including 1 brick fragment and 1 window glass fragment (architectural), 1 gray salt-glazed stoneware sherd (kitchen), and 2 unidentified metal fragments (miscellaneous).
The data collected from the Thomas Point site (18ST570) indicated that the site was first occupied in the Late Archaic period (ca. 4,000-2,000 BC). More frequent re-occupation during the Middle Woodland period (ca. 200 BC-AD 900) was evidenced by the presence of Mockley sherds in the midden. Occupation intensified during the Late Woodland period (ca. AD 900-1600) as indicated by the formation of a continuous midden and the presence of associated artifact types. Increased sedentism associated with the Late Woodland period was reflected by food remains which indicated that maize agriculture was an integral part of the economy. Therefore, the Thomas Point site has yielded important information on the transition from the Middle to the Late Woodland periods. The proposed residential development of the late 1980’s never became a reality but the pipeline project was completed with little adverse effect to the site. The area where the site is located is currently the Myrtle Point Park, maintained by St. Mary’s County, Maryland.
|
