A Phase I Intensive Archaeological Survey of the Stanwick Farm,

Aquasco, Prince George‘s County, Maryland,

Phase II Investigations of Garrett‘s Chance #2 (18PR703), and Phase II/III

Investigations of Garrett‘s Chance #3 (18PR704)

BY

JAMES G. GIBB ANDREW GARTE & ASSOCIATES 6285 Shady Side Road Shady Side, MD 20764

Submitted to

LANDESIGN Engineers • Land Surveyors • Planners 2905 Mitchellville Road, Suite 111 Bowie Professional Center Bowie, MD 20716

August 2004 Revised September 7, 2005 Revised February 9, 2006

Abstract LANDESIGN, Inc., proposes residential development of 113 acres of the 134-acre Stanwick Farm at Aquasco, Prince George‘s County, Maryland (Maryland Archeological Research Unit No. 9), the project to be called Garrett‟s Chance. In anticipation of concerns of the Prince George‘s County Department of Planning, and out of concern for the increased design costs that might be incurred to avoid potentially significant archaeological resources, LANDESIGN, Inc., commissioned this Phase I archaeological and architectural survey of the area of potential effects. Fieldwork was undertaken between May 9 and July 15, 2004, under the direct supervision of the Principal Investigator with one or two archaeological technicians. All other aspects of the project were undertaken by the Principal or by an assistant under his immediate supervision. The project area is contiguous to Swanson Creek and the Charles County line. Much of the eastern portion of the tract is cultivated, although severely eroded. Most of the western portion of the tract has reforested over the past 30 to 40 years. Surface reconnaissance and limited shovel testing of the forested areas revealed that they had been more severely eroded than the current fields, likely accounting for the farmers having taken them out of cultivation. The survey involved archival research (largely title research), architectural recording and assessment, controlled surface collecting of the plowed and disked fields, and shovel testing in some forested portions of the project area. Four archaeological sites and four standing structures and ruins were inventoried, including: a probably Late Archaic site (18PR702), an early 18th- century houselot (18PR703), a Paleoindian site (18PR704), a late 19th/early 20th-century trash scatter (18PR709), a c.1930 farmhouse with two small outbuildings, and three tobacco barns, one a ruin. Andrew Garte & Associates undertook Phase II investigations at Garrett‘s Chance #2 (18PR703) and Garrett‘s Chance #3 (18PR704), and a Phase III data recovery at Garrett‘s Chance #2 (18PR703). Garrett‘s Chance #1 (18PR702) will not be developed, and no further investigation is recommended for this site. Garrett‘s Chance #2 (18PR703) has been mechanically stripped and all but a few postholes excavated: no further investigation is recommended. Garrett‘s Chance #3 (18PR704) yielded one Clovis point, a Bare Island biface, and a number of flakes and fire- cracked rocks from three intensive surface collections after heavy rains, shovel testing at 25 ft intervals, and twelve 2½ by 2½ ft excavation units. The site is heavily eroded, the cultural material diffuse and largely undiagnostic, and the hilltop is severely eroded with a plowzone as little as two inches thick in places. No further investigation of Garrett‘s Chance #3 (18PR704) is warranted or recommended. Garrett‘s Chance #4 (18PR709) is a dense scatter of late 19th and 20th century domestic refuse, with very little architectural material, on a slope below the extant farmhouse. The site appears to be severely eroded and the evidence for architecture is limited to a few brick fragments and several pieces of possible window glass. The site appears to lack research potential and we recommend no further investigation. The three barns were built 9in the 20th century. One is a collapsed ruin, another is in fair condition. The third is in good condition and will remain. The house appears to date to the 1930s and to be in good to excellent condition. It is a two story frame vernacular dwelling with a late 20th century addition. All of the interior moldings around doors and windows is identical, with bulls eye bosses, even around the doors and windows of the recent addition, suggesting a classical revivalism that is also evident in the pair of fireplace mantels. Two small outbuildings—a workshop/storage building and a chicken coop—both appear to be mid-20th 2

century. The Compton family built the house and outbuildings. They lived in the area at least since the 18th century and were among the several locally prominent farm families. The house is of a type found through Prince George‘s County and Southern Maryland. The Compton houselot does not appear to meet any of the criteria for inclusion into the National Register of Historic Places. The late 19th-century trash scatter and the Paleoindian site both lack integrity and are ineligible for the National Register. The early 18th-century site has been excavated in its entirety and also no longer meets National Register criteria. The Late Archaic site adjacent to Swanson Creek may be eligible for National Register listing, but additional investigation would be necessary to make that determination. The site, however, is not slated for development.

3

Table of Contents Abstract ...... 2 Table of Contents ...... 4 List of Figures ...... 5 List of Tables ...... 7 Chapter 1. Introduction ...... 8 Chapter 2. Project Location and Environment ...... 9 Location ...... 9 Environment ...... 9 Chapter 3. Culture History ...... 15 Regional Prehistory ...... 15 PaleoIndian Stage ...... 15 Archaic Stage ...... 17 Woodland Stage ...... 18 Local Prehistory ...... 22 Regional and County History ...... 26 Colonial Period ...... 26 Post–colonial Patterns ...... 26 Tract History ...... 27 Summary ...... 34 Chapter 4. Research Design and Methods ...... 35 Research Design ...... 35 Methods ...... 36 Chapter 5. Results of Survey...... 37 Introduction ...... 37 Garrett’s Chance Field A ...... 38 Garrett’s Chance Field B ...... 39 Garrett’s Chance Field C ...... 40 Garrett’s Chance Field D ...... 40 Garrett’s Chance Field E ...... 40 Garrett’s Chance Field F ...... 41 Garrett’s Chance Field G ...... 41 Garrett’s Chance Field H ...... 43 Garrett’s Chance Field I ...... 43 Garrett’s Chance Field J ...... 44 Garrett’s Chance Field K ...... 44 Garrett’s Chance Field L ...... 44 Garrett’s Chance Field M ...... 45 Garrett’s Chance Field N ...... 46 Sample Area 1 ...... 46 Sample Area 2 ...... 47 Sample Area 3 ...... 48 Sample Area 4 ...... 49 Sample Area 5 ...... 49 Sample Area 6 ...... 50 Summary of Results ...... 51 Chapter 6. Garrett’s Chance #1 (18PR702) ...... 52 Introduction ...... 52 Results ...... 52 Analysis and Interpretation ...... 54 Chapter 7. Garrett’s Chance #2 (18PR703) ...... 56 Introduction ...... 56 4

Archival Record ...... 57 Research Design and Methods ...... 57 Field and Laboratory Results ...... 60 Analysis and Interpretation ...... 77 Chapter 8. Garrett’s Chance #3 (18PR704) ...... 82 Introduction ...... 82 Results ...... 83 Surface ...... 83 Shovel Testing ...... 85 Excavation Units ...... 88 Summary and Interpretation ...... 88 Chapter 9. Garrett’s Chance #4 (18PR709) ...... 91 Introduction ...... 91 Results ...... 91 Summary and Interpretation ...... 92 Chapter 10.Standing Structures ...... 94 Introduction ...... 94 Compton House and Houselot ...... 94 Barn 1 ...... 101 Barn 2 ...... 101 Barn 3 ...... 101 Barn 4 ...... 102 Summary and Interpretations ...... 102 Chapter 11. Summary, Conclusions, and Recommendations ...... 104 References Cited ...... 106 Appendix A: Shovel Test and Artifact Data ...... 112 Appendix B: Artifact and Sample Catalogue ...... 113 Appendix C: Archaeobotanical Report by Justine W. McKnight ...... 114 Appendix D: Credentials ...... 118

List of Figures FIGURE 2–1. MARYLAND ARCHEOLOGICAL RESEARCH UNIT MAP...... 10 FIGURE 2–2. USGS 7.5‘ TOPOGRAPHIC MAP, BENEDICT, MD (1953; PHOTOREVISED 1974). ....11 FIGURE 2-3. ELEVATION PROFILE OF PROJECT AREA...... 12 FIGURE 2–4. PROJECT AREA LOCATION MAP...... 13 FIGURE 2–5. PROJECT AREA TOPOGRAPHIC MAP...... 13 FIGURE 2–6. SOILS MAP...... 14 FIGURE 3–1. MARTENET MAP OF PRINCE GEORGE‘S COUNTY, MARYLAND, DETAIL (1861). 31 FIGURE 3–2. MARTENET MAP OF MARYLAND (1885)...... 32 FIGURE 3–3. HOPKINS MAP OF PRINCE GEORGE‘S COUNTY, MARYLAND, DETAIL (1878). ....33 FIGURE 3-4. PLAT RECONSTRUCTION OF PART OF PROJECT AREA (1844)...... 33 FIGURE 3-5. RECONSTRUCTED PLATS AND STONE AND COMPTON FARMS...... 34 FIGURE 5–1. SURVEY LOCATIONS A-N AND 1-6...... 37 FIGURE 5–2. CULTIVATED FIELD AT GARRETT‘S CHANCE...... 38 FIGURE 5–3. AREA K, LOOKING SOUTHEAST...... 38 FIGURE 5-4. SHOVEL-TEST SAMPLE AREA 1...... 47 FIGURE 5-5. SHOVEL-TEST SAMPLE AREAS 2 THROUGH 4...... 48 FIGURE 5-6. SHOVEL-TEST SAMPLE AREAS 5 AND 6...... 50 FIGURE 6-1. MAP OF GARRETT‘S CHANCE #1 (18PR702)...... 52 FIGURE 6-2. PROJECTILE POINTS FROM GARRETT‘S CHANCE #1 (18PR702)...... 54 FIGURE 7-1. LOCATION OF EARLY COLONIAL SURFACE SCATTER...... 56 FIGURE 7-2. CONTOURS OF PLOWZONE DEPTHS AT GARRETT‘S CHANCE #2 (18PR703)...... 61 5

FIGURE 7-3. SURFACE MODEL OF PLOWZONE DEPTHS AT GARRETT‘S CHANCE #2 (18PR703).61 FIGURE 7-4. COMPOSITE MAP OF GARRETT‘S CHANCE #2 (18PR703)...... 62 FIGURE 7-5. FEATURE MAP OF GARRETT‘S CHANCE #2 (18PR703)...... 63 FIGURE 7-6. PLAN AND PROFILE OF FEATURE 18/19...... 63 FIGURE 7-7. PLAN OF FEATURE 2 AND PROFILE A—B...... 65 FIGURE 7-8. PROFILE G—H OF FEATURE 2...... 65 FIGURE 7-9. BURNED DAUB FROM FEATURE 2...... 67 FIGURE 7-10. WINE BOTTLE AND BRITISH STONEWARE MUG FROM FEATURE 2...... 67 FIGURE 7-11. RHENISH STONEWARE FROM FEATURE 2 AND OTHER CONTEXTS...... 68 FIGURE 7-12. PLAN AND PROFILES OF FEATURE 2...... 69 FIGURE 7-13. FEATURE 2 EXCAVATED, LOOKING EAST...... 70 FIGURE 7-14. BORROW PITS AT GARRETT‘S CHANCE #2 (18PR703)...... 71 FIGURE 7-15. PLANS AND PROFILES OF FEATURES 1, 3, AND 4...... 72 FIGURE 7-16. METAL ARTIFACTS RECOVERED FROM FEATURES 1 AND 21...... 74 FIGURE 7-17. KNIFE OR RAZOR FROM FEATURES 1 AND 21...... 74 FIGURE 7-18. LATEEN SPOON FRAGMENTS FROM FEATURES 1 AND 21...... 75 FIGURE 7-19. PROFILES OF FEATURE 21, SOUTHWEST QUADRANT...... 75 FIGURE 7-20. PIN, THIMBLE, INCISED BONE COMB, TACKS AND LEAD BALL...... 76 FIGURE 7-21. PROFILE OF FEATURE 22...... 77 FIGURE 7-22. PROFILES OF FEATURE 23...... 80 FIGURE 7-23. TOBACCO PIPE BOWLS FROM GARRETT‘S CHANCE #2 (18PR703)...... 81 FIGURE 8-1. GARRETT‘S CHANCE #3 (18PR704) VIEWED FROM THE TOBACCO BARN...... 82 FIGURE 8-2. MAP OF GARRETT‘S CHANCE #3 (18PR704)...... 83 FIGURE 8-3. DIAGNOSTIC STONE TOOLS FROM GARRETT‘S CHANCE #3 (18PR704)...... 85 FIGURE 8-4. TESTING GRID AT GARRETT‘S CHANCE #3 (18PR704)...... 86 FIGURE 8-5. PLOWZONE DEPTH INTERVALS DETERMINED FROM SHOVEL TESTS...... 87 FIGURE 8-6. PLOWZONE DEPTHS UNDER 0.5 FT...... 87 FIGURE 9-1. LOCATION AND EXTENT OF GARRETT‘S CHANCE #4 (18PR709)...... 91 FIGURE 10-1. COMPTON FARM BUILDING LOCATIONS...... 94 FIGURE 10-2. COMPTON FARM HOUSELOT...... 95 FIGURE 10-3. PRINCIPAL (EAST) FAÇADE OF THE COMPTON HOUSE...... 95 FIGURE 10-4. SOUTH FAÇADE OF THE COMPTON HOUSE...... 96 FIGURE 10-5. ENTRY GATE TO COMPTON FARM HOUSELOT...... 96 FIGURE 10-6. CLASSICAL REVIVAL MANTLE, SOUTH PARLOR...... 97 FIGURE 10-7. CLASSICAL REVIVAL MANTLE, NORTH PARLOR...... 97 FIGURE 10-8. STANDARD WINDOW AND DOOR TREATMENT THROUGHOUT INTERIOR...... 98 FIGURE 10-9. TYPICAL DOOR, SECOND FLOOR...... 98 FIGURE 10-10. TYPICAL LOCK SET...... 99 FIGURE 10-11. OUTBUILDINGS ASSOCIATED WITH THE HOUSE...... 100 FIGURE 10-12. WORKSHOP...... 100 FIGURE 10-13. POULTRY HOUSE, SOUTH FAÇADE...... 101 FIGURE 10-14. STABLE (BARN 2) AND CORN CRIB (LEFT)...... 102 FIGURE 10-15. CORN CRIB...... 103

6

List of Tables TABLE 2-1 SOIL TYPES, SLOPES, AND PERCENTAGES...... 14 TABLE 3-1 SEQUENCE OF PREHISTORIC CULTURAL PERIODS ...... 16 TABLE 3-2 METRICAL ANALYSIS OF MARYLAND CLOVIS POINTS...... 17 TABLE 3-3 COMPONENTS IDENTIFIED BY WANSER (1982) ...... 23 TABLE 3-4 MATERIAL FROM HOMELAND PREHISTORIC SITE (18CH665)...... 25 TABLE 3-5 CHAIN OF TITLE, STONE FARM (AKA SUNNYSIDE)...... 28 TABLE 3-6 CHAIN OF TITLE, COMPTON FARM...... 30 TABLE 5-1 ARTIFACTS RECOVERED FROM FIELD A...... 39 TABLE 5-2 ARTIFACTS RECOVERED FROM FIELD B...... 39 TABLE 5-3 ARTIFACTS RECOVERED FROM FIELD C...... 40 TABLE 5-4 ARTIFACTS RECOVERED FROM FIELD E...... 41 TABLE 5-5 ARTIFACTS RECOVERED FROM FIELD F...... 41 TABLE 5-6 ARTIFACTS RECOVERED FROM FIELD G...... 42 TABLE 5-7 ARTIFACTS RECOVERED FROM FIELD H...... 43 TABLE 5-8 ARTIFACTS RECOVERED FROM FIELD I...... 43 TABLE 5-9 ARTIFACTS RECOVERED FROM FIELD K...... 44 TABLE 5-10 ARTIFACTS RECOVERED FROM FIELD L...... 45 TABLE 5-11 ARTIFACTS RECOVERED FROM FIELD M...... 45 TABLE 5-12 ARTIFACTS COLLECTED FROM FIELD N...... 46 TABLE 5-13 ARTIFACTS COLLECTED FROM SAMPLE AREA 1 (SURFACE)...... 47 TABLE 5-14 ARTIFACTS COLLECTED FROM SAMPLE AREA 2...... 48 TABLE 5-15 SHOVEL TEST DATA FOR SAMPLE AREA 3...... 49 TABLE 5-16 SHOVEL TEST DATA FOR SAMPLE AREA 4...... 49 TABLE 5-17 SHOVEL TEST DATA FOR SAMPLE AREA 5...... 49 TABLE 5-18 SHOVEL TEST DATA FOR SAMPLE AREA 6...... 50 TABLE 5-19. PREHISTORIC AND HISTORIC RECOVERY RATES...... 51 TABLE 6-1 ARTIFACTS COLLECTED FROM GARRETT‘S CHANCE #1 (18PR702)...... 53 TABLE 7-1 SELECT CATEGORIES OF OBJECTS SURFACE COLLECTED...... 56 TABLE 7-2 ARTIFACTS RECOVERED THROUGH SHOVEL TESTING...... 60 TABLE 7-3 ARTIFACTS RECOVERED FROM FEATURE 19, POSTHOLE...... 64 TABLE 7-4 ARTIFACTS RECOVERED FROM FEATURE 2, BY STRATUM...... 66 TABLE 7-5 PIPESTEM BORES FROM FEATURE 1...... 70 TABLE 7-6 ARTIFACTS RECOVERED FROM FEATURE 3...... 71 TABLE 7-7 ARTIFACTS RECOVERED FROM FEATURE 4 ...... 71 TABLE 7-8 ARTIFACTS RECOVERED FROM FEATURE 1...... 73 TABLE 7-9 PIPESTEM BORES FROM FEATURE 1...... 73 TABLE 7-10 ARTIFACTS RECOVERED FROM FEATURE 21...... 78 TABLE 7-11 ARTIFACTS RECOVERED FROM FEATURE 22...... 79 TABLE 7-12 ARTIFACTS RECOVERED FROM FEATURE 24...... 79 TABLE 7-13 ARTIFACTS RECOVERED FROM FEATURE 23...... 80 TABLE 7-14 PIPESTEM BORE DIAMETERS FOR COMBINED PRINCIPAL SOURCES...... 81 TABLE 8-1 ARTIFACTS SURFACE COLLECTED FROM GARRETT‘S CHANCE #3 (18PR704)...... 84 TABLE 8-2 SHOVEL TESTING ARTIFACTS FROM GARRETT‘S CHANCE #3 (18PR704)...... 86 TABLE 8-3 ARTIFACTS FROM EXCAVATION UNITS, GARRETT‘S CHANCE #3 (18PR704)...... 88 TABLE 8-4 COMBINED ASSEMBLAGES FOR GARRETT‘S CHANCE #3 (18PR704)...... 89 TABLE 8-5 LITHIC MATERIALS SUMMARY FOR ALL PROVENIENCES...... 90 TABLE 9-1 CERAMICS RECOVERED FROM GARRETT‘S CHANCE #4 (18PR709)...... 92 TABLE 9-2 VESSEL GLASS RECOVERED FROM GARRETT‘S CHANCE #4 (18PR709)...... 92

7

Chapter 1. Introduction LANDESIGN, Inc., proposes residential development of 113 acres of the 134-acre Stanwick Farm at Aquasco, Prince George‘s County, Maryland (Maryland Archeological Research Unit No. 9), the project to be called Garrett‟s Chance. In anticipation of concerns of the Prince George‘s County Department of Planning, and out of concern for the increased design costs that might be incurred to avoid potentially significant archaeological resources, LANDESIGN, Inc., commissioned this Phase I archaeological and architectural survey of the area of potential effects, Phase II site examinations of two of the inventoried sites, and a Phase III data recovery at the early 18th century Garrett‘s Chance #2 (18PR703) site. All of this work is described in this report. The survey involved archival research (largely title research), architectural recording and assessment, controlled surface collecting of the plowed and disked fields, and shovel testing in some forested portions of the project area. Phase II investigations involved repeat surface collections, shovel testing and limited test unit excavation and, at. the early 18th-century site, mechanical stripping to determine the extent of subplowzone features. Phase III work at the early historic site entailed feature mapping and complete excavation of all borrow pits, one root cellar, and one of six structural postholes. This report documents the methods and results of the Phase I intensive archeological survey. It consists of eight sections: 1) Introduction 2) Project Location and Environment 3) Culture History 4) Research Design and Methods 5) Field and Laboratory Results 6) Summary and Interpretations 7) Recommendations 8) Supporting Documentation All of the work described herein was conducted in accordance with the Standards and Guidelines for Archeological Investigations in Maryland (Shaffer and Cole 1994), the Specifications for Consulting Engineers Services Manual–Section IV (Maryland Department of Transportation 1986), and the Consultant Specifications for Archeological Procedures (Maryland State Highway Administration 1992).

8

Chapter 2. Project Location and Environment Location The study area is in southeastern Prince George‘s County, Aquasco District, on the Western Coastal Plain in Maryland Archeological Research Unit 9 (Figure 2–1), Benedict USGS 7.5 minute quad (Figures 2–2 through 2-6). It is comprised of cultivated fields and interspersed woodlots occupying a dissected Pliocene plateau three miles west of the Patuxent River. Swanson Creek (the boundary between Charles and Prince George‘s counties) defines the west edge of the project area, Dr. Bowen Road the north, the 21-acre ―Sunnyside‖ houselot the east, several private holdings and an unnamed branch of Swanson Creek the south (Figure 2-4). Environment Elevations range from 60 ft above mean sea level (amsl) along the Creek to 160 ft amsl toward MD 381/Aquasco Road to the east. (Aquasco Road was known in the 19th- century as Woodville Road for the village it passed through, subsequently named Aquasco.) Figure 2-3 illustrates the topography of the area, exaggerating the vertical scale by a factor of 12.6. Swanson Creek is the deep, broad V to the right of the figure. The smaller, less deeply cut drainages toward the center represent the upper reaches of an unnamed tributary of Swanson Creek. The entire project area has suffered extensive and severe erosion, a condition common throughout the interior of Southern Maryland. Those portions of the project area that remain in cultivation reveal deposits of Pliocene sand and gravel, the soil mantle having long eroded away. Soils (Figure 2-6; Table 2-1) in the project area consist primarily of Marr sandy loam (44%) and Westphalia fine sandy loam (27%) with slopes of 2% to 12%. The Westphalia fine sandy loams tend to be steeper and more severely eroded than the Marr sandy loams. Marr series soils tend to be deep. They developed from old deposits of fine sandy to very fine sandy materials containing considerable amounts of silt and clay, lending them greater resistance to erosion than the Westphalia series soils. Pockets of Fallsington and Woodstown sandy loams occur in the northwest and southwest corners of the parcel. The sandy nature of all of these deposits and their steep slopes render them excessively well drained and highly susceptible to drought damage. Under current farming conditions, these soils produce high yields of high to very high quality tobacco, historically the principal cash crop of the area (Kirby et al., 1967). They also produce high yields of maize and wheat; important crops, but of secondary importance to tobacco. Prince George‘s County produced about one–third of all tobacco grown in Maryland in 1860 (Lawrence 1878:8), and was the premier tobacco–producing county in Southern Maryland throughout the latter half of the 18th century and well into the 19th century. In recent years local farmers have abandoned tobacco cultivation, focusing instead on maize and soybeans. Forest growth has encroached on the fields over the past 30 years. These woodlots consist of mixed deciduous growth and moderately dense understories. The irregular topography and gravelly soils beneath the leaf mold suggest that currently forested areas experienced more severe erosion than those portions of the tract that remain in cultivation. Three 1867 deeds refer to a

9

portion of this parcel as part of Doves Rest, ―also known as Gravel Walk‖ (Land Records of Prince George‘s County FS#5/ 214 and 215, and HB1/151, December 1867). Clearly, the erosion visible today represents a long-term process of soil degradation. Both parcels appear to lie on the south side of the unnamed tributary that defines the southern edge of the project area.

Figure 2–1. Maryland Archeological Research Unit map. (Source: Shaffer and Cole 1994)

10

Garrett‘s Chance

Figure 2–2. USGS 7.5‘ Topographic Map, Benedict, MD (1953; photorevised 1974).

11

Figure 2-3. Elevation profile of project area. N.B. Vertical scale exaggerated by a factor of 12.6.

12

Figure 2–4. Project area location map.

Figure 2–5. Project area topographic map.

13

Figure 2–6. Soils map. Table 2-1 Soil types, slopes, and percentages.

Fs: 4% - Fallsington sandy loam. M1C2: 20% - Marr sandy loam, 2 to 12% slopes, moderately eroded. M1C3: 4% - Marr sandy loam, 6 to 12% slopes, severely eroded. M1D2: 15% - Marr sandy loam, 12 to 20% slopes, moderately eroded. M1D3: 5% - Marr sandy loam, 12 to 20% slopes, moderately eroded. Mw: 7% - Mixed alluvial land. SaE: 8% - Sandy land, steep. WaC2: 9% - Westphalia fine sandy loam, 6-12% slopes, mod. eroded. WaD2: 10% - Westphalia fine sandy loam, 12-20% slopes, mod. eroded. WaB2: 3% - Westphalia fine sandy loam, 2 to 6% slopes, mod. eroded. WeB2: 5% - Westphalia-Evesboro complex, 2-6% slopes, mod. eroded. WoB2: 10% - Woodstown sandy loam, 2 to 5% slopes, mod. eroded.

14

Chapter 3. Culture History Regional Prehistory The prehistory of the Middle Atlantic Coastal Plain province has been extensively researched by Custer (1984), Dent (1995), Steponaitis (1978, 1983), Wanser (1982), Wright (1973), and many other scholars. The principal prehistoric and historic periods are summarized below with regard to their representation in the immediate vicinity of the study area (Table 3–1). A subsequent section details available information on the prehistory of the immediate area.

PALEOINDIAN STAGE During the latter part of the last glacial period, known as the Wisconsin, ending about 14,000 BC, most of northern North America was deeply buried beneath thick sheets of ice. The vast amounts of water contained in these continental glaciers lowered ocean levels by as much as 130m. Large expanses of the now submerged continental shelf were exposed with dry land extending for many kilometers beyond the present shorelines. The glaciers did not flow as far south as present day Maryland, and the Chesapeake Bay of today existed only as the valley through which flowed the ancestral Susquehanna River. Glacial recession 11,000 years ago (c. 9,000 BC) raised the sea level and inundated the ancestral Susquehanna valley. By 9,000 years ago (c. 7,000 BC) the rising waters flooded the lower portion of the valley. By 3,000 BC, the valley was flooded as far north as Annapolis, Maryland. By 1,000 BC, the Chesapeake Bay and the inundated portion of the Potomac River reached their present limits and modern climactic and biotic regimes developed to their present state. Oysters and a variety of benthic and pelagic fishes occupied newly created niches in what is now one of the richest estuarine environments in the world. Oak and hickory boreal forests covered the region, and swamps, marshes, and streams formed in the hinterland and along the coasts (Carbone 1976, Lippson 1973, Schubel 1981). Glacial recession and the concomitant shift in weather patterns promoted the growth of mixed conifers and oaks, which replaced the boreal forest that was comprised of jack pine and lesser numbers of spruce, fir, birch, hemlock, and alder (Delcourt and Delcourt 1981). Native Americans were attracted to the coastal environment by rich aquatic and terrestrial resources. Prior to the formation of the Chesapeake Bay (c. 3,000 BC), people occupied a broad range of upland and lowland settings, invariably close to a water source. Paleoindian tools—of which the lanceolate Clovis point, often with basal and edge grinding, is the hallmark in the Eastern Woodlands—date between 13,000 and 7,500 BC. They are rare in the region. Generally, avocational collectors and professional archaeologists find them in redeposited contexts, often associated with multi–component sites in floodplains (Brown 1979). Brown (1979) identified 44 then-known Clovis points in coastal (28), piedmont (10), and Appalachian (6) Maryland. The metrical summaries of those points appear in Table 3-2. The Maryland State Highway Administration has excavated a Paleoindian component at the deeply stratified Higgins site in Anne Arundel County (Ebright 1992). The site is located along a small drainage that appears to have shifted its course and overflowed its banks many times, colluvial and alluvial silts covering the Paleoindian and later components. The Higgins site is exceptional in its preservation of Paleoindian and Early Archaic components. Such sites have not been reported in the vicinity of the study area.

15

Table 3-1

Sequence of prehistoric cultural periods Paleo-Indian Early Woodland Date Range: 13,000-7,500 BC Date Range: 1,000-300 BC Diagnostic Points: Clovis, Hardaway-Dalton Diagnostic Points: Rossville, Calvert Diagnostic Vessels: None Diagnostic Vessels: Accokeek, Marcey Creek, Dames Climate: Gradual post-glacial warming Quarter, Selden Island Sea level: 70-110 ft below present Climate: Mild and damp Vegetation: Succession of spruce, then pine Sea level: 7-13 ft below present Fauna: Megafauna, replacement by Vegetation: Modern, stable modern fauna Fauna: Modern, stable Early Archaic Middle Woodland Date Range: 7,500-6,000 BC Date Range: 300 BC-AD 900 Diagnostic Points: Kirk-Palmer, Warren Diagnostic Points: Selby Bay, Jack's Reef Diagnostic Vessels: None Diagnostic Vessels: Popes Creek, Mockley, Wolfe Climate: Warming and increased rainfall Neck, Hell Island Sea level: 58-70 ft below present Climate: Modern, stable Vegetation: Pine replaces spruce, oak Sea level: 3-7 ft below present increases; expansion of swamps Vegetation: Modern, stable Fauna: Modern species; swamp species Fauna: Modern, stable Middle Archaic Late Woodland Date Range: 6,000-4,000 BC Date Range: AD 900-Contact Diagnostic Points: LeCroy, Stanly, Morrow Diagnostic Points: Jack's Reef, Triangles Mountain, Guilford Diagnostic Vessels: Page, Keyser, Shepard, Potomac Diagnostic Vessels: None Creek, Moyoane, Riggins Climate: Warm and wet, drying Climate: Modern, stable Sea level: 43-58 ft below present Sea level: 1-3 ft below present Vegetation: Oak-hickory association Vegetation: Modern, stable dominates Fauna: Modern, stable Fauna: Modern interior wetland species Contact established Date Range: 16th-mid 18thC Late Archaic Diagnostic Points: Triangles, some European Date Range: 4,000-1,000 BC materials Diagnostic Points: Broadspear, Savannah River, Diagnostic Vessels: Potomac Creek, iron Brewerton Climate: Modern, stable Diagnostic Vessels: Steatite Sea level: 1-2 ft below present Climate: Warm & dry, cooling after 2,300 Vegetation: Modern, stable BC Fauna: Modern, stable Sea level: 13-43 ft below present Vegetation: Climax oak-hickory; mature estuarine/wetlands communities Fauna: Modern terrestrial and marine

16

ARCHAIC STAGE Archaeologists generally defined the Archaic Stage as a period of cultural diversification, represented by more varied projectile point styles and more varied adaptations to the environment than characterize the preceding stage. Early/Middle Archaic There are no Early or Middle Archaic period sites (7,500 to 6,000 BC and 6,000 to 4,000 BC) recorded within the immediate vicinity of the project area, although there are sites of this period in Maryland. Some researchers feel that the coastal locations favored by Early and Middle Archaic peoples were abandoned in favor of Piedmont locations (Kavanagh 1982:50), but this may be based on the lack of study of sites submerged by rising sea levels. Table 3-2 Metrical analysis of Maryland Clovis points. Statistic Length Width Thickness L/W W/T Mean 60.0 25.7 7.1 2.3 3.7 Median 58.5 24.8 7.1 2.3 3.5 Min 33.0 16.0 4.6 0.0 2.5 Max 102.5 39.2 10.3 3.7 5.5 Std 15.0 5.0 1.3 0.6 0.7 n=44 Data from Lois E. Brown (1979)

Late Archaic By the Late Archaic period (4,000 to 1,000 BC), the forests around the Chesapeake Bay were primarily deciduous. The rich plant and animal life provided a wide array of foods and raw materials. Expanding Late Archaic communities took advantage of this great abundance, as evidenced by increases in both the number and size of Late Archaic sites over those of previous periods. Late Archaic peoples could have exploited the freshes of the Susquehanna, Potomac, and Patuxent rivers, as well as the shallow waters and spreading estuaries of the bay, for crabs, oysters, and anadromous fishes. At the end of the period the deciduous forests were widespread and less diverse, thereby decreasing the heterogeneity and richness of terrestrial resources. With the encroachment of brackish water into inland bays and waterways, and the stabilization of sea level during this period, the estuarine species such as shellfish became better established, and more importantly, accessible to human occupants of the area. The dominance of deciduous forests and the stabilization of sea level may have caused a shift from interior wetlands to riverine and estuarine environments. Estuaries provided numerous locations for habitation where resources were close, plentiful, and diverse. It was during the Late Archaic that local Native American groups developed more complex technologies (e.g., canoes, fish weirs, and nets), and adopted more sedentary lifestyles in large, more or less permanent, base camps along the Bay and its major tributaries, with associated seasonal camps and resource collecting sites in the interior. The expanding waters of the Chesapeake Bay and its tributary rivers, creeks, marshes, and swamps provided an extensive network for travel and communication.

17

Overland travel became more difficult as the shoreline became deeply etched by down– cutting interior streams and inundated tidal creeks. The waterways served as both transportation corridor and as a source of food. Exotic materials on Late Archaic period sites, such as rhyolite from the Blue Ridge Province of Maryland, Pennsylvania, and Virginia, argillite from the lower Hudson Valley and southeastern Pennsylvania, and steatite from Maryland‘s piedmont, indicate extensive trade networks and/or travel.

WOODLAND STAGE Archaeologists divide the Woodland Stage (c. 200 BC to AD 1600) into three periods: Early, Middle, and Late. Each period is characterized by distinctive settlement and subsistence patterns and ceramic styles. While Late Archaic peoples may have experimented with pottery making, it is the widespread appearance of ceramics that marks the onset of the Woodland Stage. Early Woodland The Early Woodland period in the Middle Atlantic Region, between 1,000 BC and 400 BC, is characterized by a continuation of many of the cultural traditions and subsistence and settlement patterns established in the Late Archaic. There was a pronounced decline in trade and exchange networks with fewer exotic materials being found on sites of this period relative to those of earlier periods, although Ohio cherts appear on Early and Middle Woodland sites in the region. Shellfish, migratory waterfowl, anadromous fish, and other marine and estuarine species were procured from the waters of the Bay, and faunal remains found at sites indicate a high reliance on woodland animals. The present vegetation patterns of the region, with tulip poplar and sweet gum in the lowlands, and oak, hickory, chestnut, and pine found in the uplands, were established by this time. Early Woodland peoples made extensive use of these resources. Underground storage facilities, grinding tools, and faunal remains often are found on Early Woodland sites (Gardner 1982). The Early Woodland period is divided in the Maryland Coastal Plain into two phases: Marcey Creek (1,000–750 BC) and Accokeek (750–400 BC). They are defined largely on the basis of pottery styles. Marcey Creek ceramics are molded (as opposed to coiled) and they are tempered with crushed steatite. Pot forms imitate steatite vessel forms of the terminal Late Archaic. They are undecorated and usually lack lug handles. Examples of Marcey Creek ceramics are found on sites throughout the Delaware and Susquehanna River valleys and in the Coastal Plain and Piedmont provinces of Maryland and Virginia, with some occurring in New York State. Selden Island wares also are found in association with Marcey Creek ceramics. They have thinner walls, steatite tempering, and cord marking on exterior surfaces. Projectile points of this phase are the Holmes/Bare Island, Claggett, Dry Brook, and Orient Fishtail points, all of which made their first appearance in the terminal Late Archaic. The Accokeek phase is named for a pottery type identified at the Accokeek site in Prince George County (Stephenson, et al. 1963), about 15 miles (9.3 km) northwest of Hughesville. Accokeek vessels are small conical vessels, tempered with sand or crushed quartz, with cord marked exterior surfaces and, often, smoothed rims. Accokeek ceramics are found in association with Calvert projectile points.

18

Wright (1973) and Custer (1984) postulate a continuation of Late Archaic settlement and subsistence patterns into the Early Woodland. Local populations formed macrobands and occupied semi–sedentary base camps during certain seasons. At other times of the year, they split into microbands and occupied short–term task specific and seasonal camps. With the development of food preservation techniques, such as underground storage, larger populations could be supported in smaller areas. Food storage reduced the need for seasonal migration. It also required a degree of sedentism in order to maintain access to, and control over, stored foods. Population growth probably occurred at this time. Base camps appear in the Chesapeake Bay along the major river drainages, and several extensive surveys, conducted along the Wicomico, Severn, South, and Patuxent rivers, have identified numerous Early Woodland sites. In his survey of the Severn River, Wright (1968, 1969) identified eight sites with Marcey Creek components. Steponaitis (1978) found three Marcey Creek components along the South River, and ten within the Patuxent River drainage (1980, 1983). Both Wright and Steponaitis found the majority of the Marcey Creek sites in the upper reaches of the rivers, with a few sites next to estuaries. All of these sites are shell middens. Wanser (1982) documented 28 aqssemblages from Early Woodland components along the Wicomico–Allen‘s Fresh– Zekiah Swamp drainage, 21 one which are situated in interior wetlands settings. This pattern indicates a riverine orientation for Early Woodland sites, especially those of the Marcey Creek phase. The Accokeek phase sites represent a shift from the established Late Archaic– Marcey Creek period sites. Steponaitis identifies three trends: a greater number of Accokeek sites than Marcey Creek, suggesting population growth; an increase in the amount of artifacts found on Accokeek sites, indicating longer occupations, and; an increase in oyster use, and exploitation of a broad range of terrestrial and aquatic resources. Intensive gathering in rich ecozones supported a shift toward increased sedentism and population growth. A shift in trade networks also is seen with the acquisition of exotic materials and tools: chert from New York, Canada, Indiana, Ohio, and Tennessee; copper from the Great Lakes region; and Adena or Adena–like goods similar to those found in Ohio. The latter examples are found almost exclusively at mortuary sites, indicating a complex Adena–like mortuary practice. The West River site in southern Anne Arundel County is the closest identified manifestation of Adena to the study area (Ford 1976). Middle Woodland Subsistence and settlement pattern changes distinguish the Middle Woodland period in the Middle Atlantic region from earlier periods. The Middle Woodland is divided into two phases: Popes Creek (400 BC–A.D.200) and Selby Bay (A.D.200–800), each characterized by distinctive ceramics and projectile point types. Popes Creek Net Impressed ceramics have a medium to coarse sand temper comprising 50% to 70% of the paste. The vessels are coil constructed, in the form of wide–mouthed jars, with conical or semi–conical bases. Interiors are scraped and exterior finishes are net impressed. Rims are decorated with incised horizontal lines, often with

19

finger smoothed and incised chevron patterns. Popes Creek ceramics rarely are cord marked. Wright (1973) identified a local variant that he has named Smallwood ware, but the only significant difference is the presence of some shell and quartz tempering in a sandy paste. Rossville projectile points occur in deposits with Popes Creek ceramics. They occur on sites from southern New England to the Chesapeake Bay. The Popes Creek tool assemblage also includes bone awls, knives, grinding stones, mortars, axes, choppers, and hammer stones of local lithic material. The Selby Bay phase follows the Popes Creek phase, and is represented by Mockley Cord–marked and Net Impressed pottery, and exotic lithic materials. Mockley ceramics are tempered with coarse crushed shell, comprising about 20% to 30% of the paste. The vessels are coil constructed, medium to large in size, with rounded or semi– conical bases. Vessels from the beginning of the period are predominantly cord–marked. Cord marking appears to have been gradually supplanted by net impressed treatments, both plain and crumpled. Vessel rims often are undecorated with some vessels having their exterior surfaces smoothed just below the rim. The smoothed necks commonly are decorated with incised cross–hatching, diamonds, chevrons, or parallel lines, with occasional punctates. Mockley pottery is found on sites from the western coastal plain of Virginia to the Delaware River. On Maryland‘s Western Shore they occur in association with Selby Bay bifaces—made from non–local rhyolite, argillite, and jasper—and elliptical two–holed gorgets, hematite squares, grinding stones, bifacially retouched flakes, and worked bone. Gardner, et al. (1989), also recovered several Piscataway points from a pit at 18CV272 in association with Mockley sherds. The chronological placement of Piscataway points, however, is still a point of contention among scholars in the region (e.g., Ebright 1992:38). The Popes Creek phase may represent local development, with an intensification of the subsistence patterns established during the Accokeek phase of the Early Woodland. Large semi–permanent macroband sites were located along the upper portions of major river drainages, with associated satellite procurement stations located in strategic spots near the base campsites. There is some discontinuity between the lithic assemblages of the Popes Creek and Selby Bay phases. Popes Creek tools generally were made from locally available quartz and quartzite. Selby Bay phase lithic assemblages are entirely different, dominated as they are by exotic materials: rhyolite from the Blue Ridge Province of Maryland and Pennsylvania, argillite from the northeast, and cherts from New York and Ohio. Luckenbach, et al. (1987), suggest that there was a greater affinity of Selby Bay phase peoples with populations to the north, if not migration into the Maryland Coastal Plain Province from the north. Custer (1986) hypothesized that this settlement pattern reorganization may have culminated in the establishment of small chiefdoms by the Late Woodland period. Gibb and Hines (1997) suggest intensive use of particular aquatic resources, specifically oysters, to the near exclusion of other aquatic and terrestrial resources at the Smithsonian Pier site (18AN284) on the Rhode River. Because of the seasonal nature of their use of this resource, and the relative lack of competing species (e.g., drumfish, boring sponges), Middle Woodland visitors to the Smithsonian Pier site appear not to have affected the local oyster population‘s ability to reproduce. Neither the Smithsonian Pier site nor the Luce Creek site (18AN143) on the Severn River yielded

20

definitive evidence of horticulture, although Ballweber (1994a) found ample evidence of hickory nut processing at Luce Creek. Late Woodland The first true signs of horticulture in the Middle Atlantic region mark the beginning of the Late Woodland Period (c. AD 800). The period ends with sustained European contact in the 17th century (after A.D.1600). Horticulture was widely and rapidly adopted throughout the northeastern United States at this time and may have been introduced by cultures to the west of the Chesapeake Bay region. The environment remained essentially the same and local peoples continued gathering plants, hunting, fishing, and oystering. At the time of European contact, aborigines relied less on estuarine resources than did their immediate precursors. Horticultural villages on floodplains were the primary occupation sites of the native inhabitants. Archaeologists divide the Late Woodland into two phases: Little Round Bay (AD 800–1250) and Sullivans Cove (AD 1250–c.1600). Little Round Bay Phase ceramics include incised and fabric impressed wares of the Rappahannock series. Both are shell–tempered. The vessels are coil constructed, with smooth interiors and rough exteriors. They tend to be more thinly potted, and the temper is smaller, than the earlier Selby Bay vessels. Rappahannock ceramics are wide–mouthed jars with rounded or semi–conoidal bases. Griffith (1980) defined eight varieties of Rappahannock Incised pottery, based on decorative treatment. Motifs include horizontal bands, zigzags, and squares or triangles, occasionally filled in with incised lines. Generally, the more complex geometric forms occurred during the period between AD 900 and AD 1300. Fabric impressions on Rappahannock wares typically are clear and not over–stamped. Some vessels have pseudo–cord impression patterns at the rim. Projectile points associated with the Rappahannock ceramic types include Jacks Reef points—found throughout Maryland, Delaware, Virginia, Pennsylvania, New York, Ohio, Michigan, and Ontario—and Levanna points—found throughout Maryland, Virginia, Delaware, Pennsylvania, New Jersey, New York, Ontario, and into New England. Other Late Woodland artifacts include bone awls, obtuse angle pipes, grinding stones, and pitted stones. Sullivans Cove pottery is thinly potted with light crushed shell tempering. Vessels have conical bases and constricted necks. Body sherds are partially cord–marked and smoothed. Rim exteriors are decorated with cord wrapped stick impressions, and horizontal lines and herringbone patterns. Rappahannock Incised ceramics with less complex motifs also are found with Sullivans Cove pottery, as is the Rappahannock Herringbone motif. The small triangular Madison projectile point, found throughout the northeastern United States, typically is the only projectile point found on Sullivans Cove phase sites. The small size of the Madison point indicates that Late Woodland peoples replaced the throwing spear, which required a larger and heavier point, with the bow and arrow.1 Sullivans Cove assemblages also include: grinding stones, convex–edged end scrapers, knives, and other stone tools. It was during the Sullivans Cove period that

1 See Nassaney and Pyle (1999) on the morphological distinction between dart and arrow points.

21

horticulture seems to have led to a shift to village life in locations away from the shores of the Chesapeake. Custer (1984) suggests that vast changes occurred in the settlement and subsistence patterns of the Late Woodland. Prior to A.D.1000, settlement and subsistence patterns centered around intensive gathering and hunting with some use of cultigens. Groups followed seasonal rounds, moving from base camp to base camp, with occasional forays to task specific sites to procure shellfish, waterfowl, and other resources. Wright (1973) suggests that the Little Round Bay Phase occupations centered on base camps at the estuarine/transition zones, with frequent use of numerous nearby procurement camps. Wright interpreted the Obrecht site, near the head of the Severn River, as a base camp for the Purcell site on the Magothy River and the Oakridge site on the Patapsco River. The two smaller sites served as resource procurement sites. Obrecht, a large oyster shell midden measuring 180m in length, produced materials from the Middle Woodland and Late Woodland periods. Wright interprets the broad array of faunal remains and cooking features at the Obrecht site as evidence of a large macroband base camp. The Purcell site is an oyster shell midden site, measuring at least 25m in length, with a similar broad array of faunal remains. Wright suggests that it is a microband base camp, probably occupied in the fall. The Elkridge site is a very large site on the estuarine portion of the Patapsco River, at the confluence of three major tributaries. It is well placed for the exploitation of spring runs of spawning fish. Development has destroyed a number of smaller shell sites near Elkridge that could have served as microband procurement sites. Procurement sites were selected for their ease of access to seasonally available oyster, waterfowl, or fish, or nuts. Increased reliance on cultigens lessened the need for satellite camps, and this shift is reflected in the archaeological record. The functions of base camps changed as they became village sites devoted to the production, storage, and protection of food. The need for cropland also required a shift away from coastal areas to fertile floodplains. Horticulture in the Bay region became important some time around AD 1000, during the Sullivans Cove Phase. Smaller villages and isolated household sites, or clusters, surrounded larger settlements. Sullivans Cove phase peoples still used sites previously used for oystering, waterfowling, fishing, and hunting, but not as intensively. Local Prehistory The project area is remarkable for the paucity of archaeological studies undertaken within a two–mile (3.2 km) radius in Prince George‘s and Charles counties. Richard Stearns recorded numerous prehistoric sites along the Patuxent River in the 1950s, including the Kennedy Run lithic scatter (18PR101) at the confluence of that drainage and the river. Several highway surveys, very limited in scope and intensity, were conducted in the area, none of which identified sites (Wesler et al., 1981). Ingersoll and Kenney (1978), and Hopkins, Collier, and Fischler (1992) surveyed portions of the power transmission line that passes through Aquasco en röute to Chalk Point on the Patuxent River. Gibb (2000), investigating portions of a proposed road realignment in connection with the construction of the Hughesville Bypass, found only scattered evidence of late historic occupation and the railbed of the Southern Maryland Railroad (18CH697), built in the 1870s between Brandywine, Prince George‘s County, and Mechanicsville, St. Mary‘s County.

22

The lack of finds in this portion of upland Prince George‘s and Charles Counties can be attributed to the lack of archeological investigation. Numerous sites have been located by avocational archeologists and by the Potomac River Survey along the margins of Zekiah Swamp, Hickey (1970) reporting 57 in his master‘s thesis, Late Archaic through Early Woodland components predominating. His study and Wanser‘s (1982) analysis of collections from Zekiah Swamp indicate an abrupt shift in settlement patterning in the Early Woodland (Table 3–3). Woodland sites reveal a preference for coastal lowland sites, undoubtedly representing their increasing reliance on the stable aquatic communities that developed in the wake of glacial recession and rising sea levels. Table 3-3 Components identified by Wanser (1982) Period Interior % Interior % Coastal % Total Wetlands Uplands Lowlands s Early Archaic 78 85 11 12 3 3 92 Middle Archaic 54 77 8 11 8 11 70 Late Archaic 313 82 39 10 30 8 382 Early Woodland 21 57 4 11 12 32 37 Middle Woodland 39 55 6 8 26 37 71 Late Woodland 46 56 7 9 29 35 82 Totals 551 N/A 75 N/A 108 N/A 734 Despite this shift, and even before it, indigenous peoples settled––at least seasonally and temporarily––upland interior sites, presumably to take advantage of resources that were absent or less accessible along the coast and interior wetlands. Gibb‘s (1991) investigation of the Allen‘s Fresh No. 1 site in the transition area between the open water of the Wicomico River and the wetlands of Zekiah Swamp demonstrates that these peoples did not go into the uplands to procure stone for tool making: the gravel bars of the lower Zekiah provided an inexhaustible supply of quartz pebbles. Plentiful gravels for stone tool making may have existed in the project area, especially along the stream banks, in the distant past, even before commercial agriculture degraded the soils. Barse (1982) conducted a survey of a proposed 11+ mile (6.8 km) transmission line corridor in northern St. Mary‘s County in a setting very similar to the setting of the western portion of the project area and to upland areas throughout Southern Maryland. His field crew excavated 41 shovel tests and surface collected plowed fields and other exposed surfaces. They identified 36 lithic scatters (one with an 18th–century domestic component in addition to prehistoric material––18ST407) at the upper reaches of several tributaries of the Potomac River at elevations of 100 to 180 ft (30.5 to 54.9 m) above mean sea level. Few produced temporal diagnostics, and those components that could be identified dated to the Early Archaic (2), Late Archaic (5), and Early–Middle Woodland (6) periods, with the latter two predominant. All of the sites occupied bluffs overlooking stream valleys or remnant terraces. Pebbles and cobbles native to the soils appeared throughout. Cheek et al. (1983) tested five of those sites, recovering from four of the five a total of 31 pieces of flaked stone and one fire–cracked rock. They found only one temporally diagnostic artifact: a Vernon side–notched point from 18ST418. Barse (1982), drawing from earlier work conducted by The Catholic University along the Piscataway drainage in southern Prince George‘s County, offered a predictive model applicable to his St. Mary‘s survey and appropriate to generate expectations for the

23

interior uplands of Prince George‘s and Charles counties. Beginning in the Late Archaic period, indigenous groups established large base camps at the heads of embayed streams. From these sites they harvested anadromous fishes during the annual spawning runs. These sites should produce comprehensive stone tool kits and late stages of lithic reduction. Smaller sites on low order streams and in interfluvial upland areas should represent hunting camps and ―primary quarry activities‖ with relatively few tool types represented and a preponderance of primary flakes (Barse 1982: 13). Pending an analysis of the kinds of flakes and tools recovered from these sites, it is difficult to evaluate the model. In general, it is accurate in that lithic scatters, devoid of discernible subsurface features, characterize upland sites throughout the region. That groups sought these upland areas for usable stone seems improbable given the gravel bars that exist along the lower reaches of the larger streams. That is not to say that they did not use upland deposits of pebbles, but they may have used them on an as–needed basis, creating expedient tools and subsequently discarding them before moving on. In other words, the differences in early and late stage flake frequencies between large fluvial and coastal sites and upland sites may have little or nothing to do with quarrying and stone availability: those differences may have more to do with the range of activities undertaken at the respective sites and their concomitant needs for re-sharpening and repairing stone, bone, wood, and textile implements, and creating new tools, perhaps as expedients. Gathering of upland plants and hunting and trapping seem to be the most likely uses of upland areas. Until substantial resources are invested in the study of these artifact and feature poor sites, however, a more definitive identification of activities and variability among these sites will not be achieved. In December 1998 and January 1999, Barse et al. (1999) conducted a Phase I archeological survey of the Hughesville Transportation Improvement Project area, testing the proposed widened ROW of MD 5 and the proposed eastern bypass. Two of their shovel tests––A34 and B34 fall within the proposed Foster Lane relocation ROW. The report documents typical soil profiles, not those for all of the units, but the A34 and B34 apparently revealed only a typical Beltsville silt loam profile and yielded no artifacts. Further south, just north of Homeland Drive and overlooking the upper reaches of Indian Creek, they identified a probable brick clamp (18CH664) of uncertain date and an Archaic lithic scatter (18CH665). Their subsequent investigation, not yet reported, revealed two intact clamp bases. The prehistoric lithic scatter lies within a cultivated field on a bench above an unnamed tributary of Indian Creek at 175 ft (53.3 m) above mean sea level, a setting entirely consistent with the kinds of upland sites discussed above. Through surface collecting (approximately 750m2, about 8,000 ft2) and 17 shovel tests, the field crew collected 72 prehistoric artifacts (Table 3–4).2 Barse et al. (1999: 4.20) report that the one Lamoka projectile point recovered appeared to have been resharpened and most of the flakes were ―biface thinning specimens, though several larger flakes reflect[ed] primary reduction activities.‖ The larger flakes, two flakes with cortex, and six cores clearly

2 The Halifax projectile point recovered from surface collection Block 5 does not appear in the catalogue, hence the report understates total prehistoric finds by one item.

24

represent early stages of lithic reduction, but the remainder of the material represents late stages of lithic reduction, suggesting tool repair and replacement rather than quarrying. The two projectile points indicate hunting, but the fire–cracked rock is difficult to interpret. Barse et al. (1999: 4.13) suggest some structure to the site based on the distribution of fire–cracked rock: Field counts [of controlled surface collected material] showed that the site could be divided into two separate and low density artifact concentrations. These two concentrations were defined primarily on the basis of higher counts of fire–cracked rock and debitage. These two concentrations probably mark the loci of two or more hearths that have been disturbed by plowing. With raw counts ranging between 0 and 9 for the six 100m2 collection units, the clustering observed by the investigators is not statistically supportable. Table 3-4 Material from Homeland Prehistoric Site (18CH665). Object type Count Cores 6 Flakes with cortex 2 Flakes without cortex 24 Bifaces 3 Projectile points 2 Fire–cracked rock 35 Total 72 (Source: Barse et al., 1999) Balicki et al. (2004) conducted a surface collection of 18CH665 in connection with a Phase II study of the Homeland Brick clamp Site (18CH664). They recovered 21 pieces of flaked stone, including: two bifaces, a quartz core, a tested jasper pebble, and three quartz flakes with cortex. Although they did not analyze the material and its distribution, Balicki et alia‘s project map (2004: Figure 5) depicts a broad arc some 80m in breadth open to the north. The catalogue lacks sufficient detail to distinguish clearly between early and late stages of lithic reduction, although the small collection appears to embrace both: biface thinning flakes and retouch flakes are unlikely to be recovered through surface collection. The question remains whether the fire–cracked rock identified in 1999 represents roasting or boiling of meat, plants, or both. Moreover, is the site a ―palimset (sic) representing a number of these small short–term occupations dating to the Archaic‖ (Barse et al. 1999:4.22), or is it a single component site with structure representing the segregation of activities, groups, or both? The degree of faith one places in the regional projectile point typology will influence one‘s interpretation since the supporting evidence for multicomponent occupation is the distinction drawn between a Halifax point and a Lamoka point. Clearly, however, much remains to be learned from these small upland scatters, plowed or unplowed, and that will require resources seemingly out of scale with the return of artifacts and preserved features, but not necessarily out of scale with the potential return of information. Minimally, more intensive close interval sampling and

25

more detailed core and flake typologies (e.g., Gibb 1991: Appendix B) are required. The concept of successive stages of lithic reduction––with its teleological assumption of a formal tool as end product––also requires reexamination. The data from both the 1999 and 2000 studies suggest opportunistic collecting and working of lithic resources, of which traces can be expected at Garrett‘s Chance. Regional and County History Historic settlement patterning in the Chesapeake Tidewater region has been examined by Pogue (1984), Smolek (1984), Lukezic (1990), and Gibb (1996). Concerned with 17th and 18th–century Euro–American settlement along the bay and its tributaries, these studies all note a preference for sites along major navigable rivers near potable water and soils suited to tobacco and wheat cultivation, with little aggregation and avoidance of upland areas. Gibb‘s analysis aimed at documenting and interpreting variability and offered a statistical technique for identifying sites that may have functioned differently than those tobacco plantations along the navigable waterways. Settlement patterning in Maryland‘s Tidewater region for the 19th and 20th centuries has not been studied and the comments below pertaining to these later settlements are based on preliminary research.

COLONIAL PERIOD Land grants from the Lords Baltimore, proprietors of the Maryland colony, varied greatly in size. Tracts listed in the various rent rolls range from a few acres to thousands of acres, with around seventy percent of the patents granted for parcels between 50 and 249 acres (123.5 to 615.3 ha) (Wykoff 1937; Gibb 1996). Most 17th–century archeological sites occur within a few hundred feet of navigable water and near soils suitable for producing tobacco in large quantities, if not high quality; but a few have been found a mile or more inland, surrounded by soils ill–suited to tobacco culture. Tenants occupied all tracts, only the Lord Proprietor actually owning the land. Tobacco was the principal cash crop, except where wheat dominated in portions of St. Mary‘s County and the lower Eastern Shore, with maize, cattle, and swine raised for home consumption, ship provisioning, and limited coastal trade.

POST–COLONIAL PATTERNS As the colonists patented all of the prime lands along the coast, they began to move inland. By the middle of the 18th century, the interior of Southern Maryland was thoroughly colonized and a nascent road system developed. The Lords Baltimore had begun to alienate land, selling it in fee simple and abolishing quit rents. Farm tenancy increasingly became the means by which rural families gained access to farmland from large, wealthy landowners. Stiverson (1977) and Marks (1979) have examined patterns of farm tenancy for the 18th and early 19th centuries, respectively, but I am unaware of any historical studies of late 19th and early 20th–century tenancy in Southern Maryland. Gibb‘s (1991) analysis of late 19th–century agricultural schedules from the federal censuses for the Allen‘s Fresh District of Charles County, based on a 30 percent (n=61) systematic sample of the 1880 entries, indicates a 33 percent farm tenancy rate with 16.5 percent renting and 16.5 percent farming on shares. Farm size in that district decreased dramatically during the last half of the century with median values for improved acreage of 178 (440 ha) and 60 (148 ha) for 1850 and 1880, respectively. Yields of the principal

26

cash crops––tobacco, maize, and wheat––plummeted by 17, 60, and 40 percent, respectively. Tobacco, clearly, remained the most important crop. Whether or not agriculture followed a similar course in the Aquasco District of Prince George‘s County remains uncertain: the intensive sampling and analysis of late 19th–century agricultural census data simply hasn‘t been undertaken. While not especially industrial, the counties of Southern Maryland had craftsmen, sedentary and itinerant, providing some goods and services for businesses and residents. For example, the Trustees of Charlotte Hall Academy, just south of Hughesville, St. Mary‘s County, contracted Richard Carnes in 1783 to make 250,000 bricks and lime for mortar with which to build their first school building (Gibb 1990; 1989: 5). They contracted with a man named Kirkley or Kirkby in 1857 to burn 150,000 to 250,000 bricks for a new classroom building and the Building Committee reported at the end of July of that year a kiln of 65,000 bricks ready for firing. They discovered the following year that the bricks were of inferior quality and the contractors were required to effect repairs (Gibb 1989:6–7). Barse et al (1999) uncovered the remains of two brick clamps in their Phase I investigation of the Hughesville bypass corridor, the Homeland Brick Clamp site (18CH664), tested more extensively by Balicki et al (2004). Balicki et al (2004) did recover a molded ogee brick, suggesting bricks made for a building (possibly as replacements for existing fabric) and mapped two large and one medium sized borrow pit, four spoil piles, and six small borrow or mixing pits. Brickyards are notoriously difficult to date, lacking much of the domestic refuse that archaeologists rely on for dating sites, and the Homeland Brick Clamp, apparently unassociated with any domestic or non–domestic buildings, remains an enigma. Other trades and industries in the area included smithing, wheelwrighting, and milling, particularly of timber. Such sites appear on atlas maps of the late 19th century, but have little visibility in the industrial schedules of the decennial censuses. Tract History The project area is a recent amalgamation of two modest-sized farms: the approximately 50-acre farm of Dr. Michael J. Stone (now called Sunnyside), and approximately 80 acres of the Compton farm, both of which appear on the Martenet map of 1861 and the Hopkins atlas map of 1878, but—for lack of detail—were omitted from the Martenet map of 1885 (Tables 3-5 and 3-6; Figures 3-1 through 3-3). The metes and bounds of Dr. Stone‘s purchase of two small tracts in December of 1844 (Land Records JBB1/21) permit an accurate reconstruction of the farm‘s boundaries (Figure 3-4). Unfortunately, descriptions of the Compton farm rely on references to neighboring properties and stream banks, precluding accurate measures of its size and placement on the landscape. Tracing ownership of parcels back into the 17th century is a difficult undertaking. An undivided parcel, say 100 acres, can be traced back to the original patent with a fair degree of confidence; even so, the bounds of that tract were not necessarily immutable or defined with any exactitude. The problem in identifying occupants, as opposed to owners, further complicates matters since tenants infrequently left paper trails. They may appear among the proprietary rent rolls (1651-1772), but in later records appear, if at all, only in census and judicial records. Added to these factors, the researcher must consider that many tracts were subdivided or entirely redefined. These restraints manifest themselves in this study in the form of question marks that appear throughout Tables 3-5 and 3-6. More intensive research

27

that addresses adjoining tracts may clarify some of the issues discussed below. Only limited plat reconstruction for the neighborhood was possible for this project (Figure 3-5). The 1901 will of Mary S. Compton clearly notes that the Compton farm was part of a tract called Doves Nest. The 1679 patent for Doves Perch references the northeast corner of Bernard Johnson‘s Doves Nest; hence the former bordered the east side of the latter and the two tracts, totaling 350 acres, encompassed both the Compton and Stone farms. Johnson had acquired Doves Nest from Nathaniel Truman in 1686. The patents and Bernard Johnson‘s will strongly indicate that Johnson occupied the two tracts, although no document notes the location of his houselot. Moreover, a combined parcel of 350 acres likely was occupied by one or more tenant households in addition to the Johnson household. Johnson left four heirs, all daughters, discounting the possibility that one or more sons set up household on Doves Nest or Doves Perch. When Johnson died around 1702, however, he bequeathed both tracts to his daughters who, by virtue of an indenture, partitioned each tract equally, each taking 37½ acres of Doves Nest and 50 acres of Doves Perch. At the time of this agreement (1702), all were married and the properties conveyed, by law, to their husbands. (There is no evidence of the establishment of trusts, the common means by which women protected family property prior to marriage in the Colonial period.) Mary and Philip Willery and Elizabeth and Hugh Williams conveyed their divided and unconsolidated portions to William Wilkerson in 1711. Catharine and Paul Rawlins and Martha and Thomas Nelson sold their interests in Doves Nest as well, although when and precisely to whom I have not determined. William Wilkerson, a large landowner in Patuxent Hundred (later Aquasco District), appears to have acquired most if not all of Doves Nest and Doves Perch during the first quarter of the 18th century and his family retained ownership until just before the Revolutionary War, conveying the land to Samuel Crane. By means undetermined (possibly from the Morton family), John TW Compton acquired 90 acres of Doves Nest prior to the Civil War. The Compton family retained ownership throughout the second half of the 19th century and through most of the 20th century. Similarly, Dr. Michael J. Stone purchased two parcels to the east in 1844, totaling 50 acres, most of which remained in the hands of his descendents well into the 20th century. Table 3-5 Chain of title, Stone farm (aka Sunnyside). Grantor Grantee Instrument Date Tract Acres Betty Carney Taussig Covington P. & Monique G. Deed 6810/700 10/29/1987 43.29 Stanwick Betty Carney Taussig Betty Carney Taussig Deed 5456/820 8/22/1981 43.29 Jos Kneffler Taussig, pers Betty Carney Taussig Deed 5339/370 11/13/1980 43.29 rep for Anna Lee Craycroft Turner Covington P. Stanwick Covington P. & Monique G. Deed 6810/711 10/27/1987 21.25 Stanwick Covington P. Stanwick Deed 5580/322 9/16/1982 21.25

Anna Lee Craycroft Betty Carney Taussig Will 28/498 Turner Paul L & Naisby W Sam‘l L & Shirley A Canter Deed 3181/388 7/27/1965 2.71 Herring Paul L & Naisby W Anna Lee Craycroft Turner Deed 2981/624 5/25/1964 46 Herring Anna Lee Craycroft Paul L & Naisby W Herring Deed 2981/621 5/20/1964 46 Turner Nellie B. Stone Anna Lee Craycroft Turner Will GSO1/388 4/5/1948 Eleanor ―Nellie‖ B. Stone Nannie Lawrence Deed 696/307 6/15/1943 4

28

Grace H. Stone Eleanor Stone Will WTD7/210 ½ interest Frederick D. Stone Grace & Eleanor Stone Will WTD6/577 ⅓ interest Margaret T. Stone Grace, Eleanor & Frederick Will WTD1/285 ¼ interest Stone Thomas Somerville Stone Stone heirs 1919 Susan Ann Stone Thomas Somerville Stone Will JB1/602 1899 Michael Jenifer Stone Susan Ann Stone Will WAJ1/122 John W Compton Michael Jenifer Stone Deed HB1/151 or 1870 HB3/301? George & Ellen H Morton Michael Jenifer Stone JBB4/21 12/17/1844 Pt of 51.44 William & Mary Lord Baltimore William Holmes/Holms Pat Cert 2347 1723 William & 100 Mary

29

Table 3-6 Chain of title, Compton farm. Grantor Grantee Instrument Date Tract Acres Ann C. Fey (aka Lines) & Andrew F. & Patricia L. Deed 15801/184 4/8/2002 82 Katherine C. Long Metroka Daniel F. Fey, rep for est Ann C. Fey & Katherine C. Deed 7157/397 11/14/1988 82 (½ interest) of Peter Wood Compton Long Daniel F. Fey, rep for est Peter Wood Compton Deed 5624/738 11/18/1982 82 (½ interest of Henry Compton Henry Compton Jr & Peter Wood Compton Deed 389/187 8/16/1932 82 (½ interest Rebecca Compton Henry Compton Jr & Henry Compton Jr Deed 389/187 8/16/1932 82 (½ interest Rebecca Compton Henry Compton Peter Wood Compton Wills 39/142 ? Life estate Mary S. Compton Henry Compton & Rebecca Wills WAM1/201 1/21/1901 Doves Nest 80 Compton John TW Compton Christiana M. Key & Deed HB@/45 1/15/1868 90 Margaretta M. Key, trustees for Mary S. Compton ? John TW Compton < 1860

Wm Wilkerson Samuel Crane 2/17/1773 Doves Nest ? ? Wm Read ? ? Doves Nest 40 or 48 John George Josias Wilson ? ? Doves Nest 73½ Musselwhite? Martha & Thomas Nelson John George Musselwhite ? ? Doves Nest 37½ Catharine & Paul Wm Wilkerson ? ? Doves Nest Rawlins Mary & Philip Willery Wm Wilkerson ? 10/10/1711 Doves Nest 87½ & Doves Perch Elizabeth & Hugh Wm Wilkerson ? 3/21/1711 Doves Nest 87½ Williams & Doves Perch Heirs Mary Willery, Heirs Mary Willery, Indenture 8/10/1702 Doves Nest 87½ each from Elizabeth Williams, Elizabeth Williams, & Doves unconsolidated Catharine Rawlins, & Catharine Rawlins, & Perch parcels Martha Nelson Martha Nelson Bernard Johnson Heirs Mary Willery, Wills < 1702 Doves Nest 150a & 200a Elizabeth Williams, & Doves Catharine Rawlins, & Perch Martha Nelson Nathaniel Truman Bernard Johnson <1679 1686 Doves Nest 150 Thomas Truman Nathaniel Truman ? 1680 Doves Nest 150 Robt Dove & John Thomas Truman assigned 1664 Doves Nest 150 Anderson Lord Baltimore Robt Dove & John Patent 12/235 1664 Doves Nest 150 Anderson Lord Baltimore Bernard Johnson Patent 20/303 6/21/1679 Doves Perch 200

30

Project area

0.5 miles

N

Figure 3–1. Martenet map of Prince George‘s County, Maryland, detail (1861).

31

~1 mile

Figure 3–2. Martenet map of Maryland (1885). Aquasco/Woodville circled.

32

Project area

2650 ft

N

Figure 3–3. Hopkins map of Prince George‘s County, Maryland, detail (1878). N.B. Stone and Compton properties circled.

Woodville to Aquasco Road

15

Woodville Road 14 17 18 George & Ellen Morton to 21 Michael Jenifer Stone JBB4/21 17 Dec 1844 1 51.7 acres 2 Barn 9 6 Branch John B. 13 3 11 5 Thomas 12 4 W. Bowling Branch SCALE 0 1000 ft

Figure 3-4. Plat reconstruction of part of project area (1844).

33

Brooke Court Manor

Black Well Aquasco Manor

19 Hatchell 2 5 of Kenaday's 18 Branch Head16 15 14 13 12 1 17 10 9 Doves 8 7 Beall Compton Pearch 6 21 Bigger Head Farm Digges Lodge 5 1 Stone 665 a 1760 4 to 4 3 Kenadays Branch Farm 1 3 4 2 34 Robt Bradly 2 Hatchett 4 25 1737 200 a 27

33 Patuxent Doves Creycroft's 24 1 River Purchase 28 Geo Biggs Jonathan Ellis 32 to Nest pt of Creycroft's Jas Thomas 22 Arthur Purchase 74 a 1741 S W 3 C Ludford Thos Pagett's C P 2 Land Beale 3

Figure 3-5. Reconstructed plats and Stone and Compton farms. Summary Archival research suggests a moderate to high probability of encountering Archaic and Early Woodland lithic scatters at drainage heads and larger, richer prehistoric sites on the creek terraces. Mid–18th through early 20th century farmsteads, quarters, and tenantries are likely throughout. I cannot identify any but the most recent occupants of the two farms with a high degree of confidence.

34

Chapter 4. Research Design and Methods Research Design Background research indicates that the study area has moderate to high potential for yielding Archaic and Early Woodland lithic scatters––particularly on the higher ground above the drainages––and late Colonial through early 20th–century farmsteads. Both types of resources are frequently encountered through shovel test surveys with sampling intervals of 20 m/65 ft or less. Surface collecting of cultivated fields, under the right conditions, is more effective than shovel testing, especially in deflationary environments where the likelihood of finding buried deposits is remote. Most of the land slated for development either is in open, cultivated fields, or is on steeply sloped forested lands recently disturbed by earthmoving machinery in connection with intensive soil testing. The initial survey aimed at the identification of all cultural remains within those areas slated for development and an initial evaluation of those resources. This phase resulted in the identification of scattered cultural materials—prehistoric and historic—at several loci and the defining of four archaeological sites. One of those sites—a prehistoric scatter of lithics and fire-cracked rock—likely to date to the Late Archaic period lies outside of the area of potential effects. No further investigation was undertaken at Garrett‘s Chance #1 (18PR702). An extensive scatter of late 19th through 20th century ceramics and glass on a severely eroded slope was inventoried as archaeological site 18PR709 (Garrett‘s Chance #4 [18PR709]). Because of its heavily eroded condition downslope from an extant 20th century farmhouse, no further work was undertaken at this site. Examination of a relatively cultivated field with dark soils revealed a thin scatter of Colonial domestic artifacts and burned daub. Close interval shovel testing and a single excavation unit were excavated to determine the extent and vertical integrity of this site, Garrett‘s Chance #2 (18PR703). The site was subsequently stripped with a front end loader to expose the features for mapping and subsequent excavation. Although this work was not driven by any explicit research questions, it employed standard methods designed to recover data on spatial organization of a Chesapeake region houselot other than intensive plowzone sampling: the Phase 2 work had demonstrated that, while sub-plowzone features existed, plowzone artifact densities were too low to admit statistically significant spatial analysis. An after-the-fact research design—actually, the same implicit research design that governs most work on Colonial Chesapeake homelots—appears in the chapter on Garrett‘s Chance #2 (18PR703). Surface collecting on a relatively level, but heavily eroded knoll near the north edge of the property resulted in the identification of a prehistoric site, Garrett‘s Chance #3 (18PR704). While most of the material collected included fire-cracked rock and lithic debitage representing all stages of lithic reduction, the field crew found only one temporally diagnostic artifact: a re-sharpened black chert Clovis point. Close interval shovel testing (25 ft) and twelve 2½ ft by 2½ excavation units produced small quantities of fire-cracked rock and lithic debitage, but only two additional tools: a quartzite uniface and a Bare Island biface of limonitic sandstone. Given the severity of erosion (plowzone as little as 0.2 ft in depth) and low yield of cultural material, no further work was undertaken. In addition to standard, phased archaeological investigations, I examined and recorded six buildings: three 20th-century tobacco barns (one a collapsed ruin, an extant

35

structure in fair condition, and a third in good condition) and a c. 1930 farmhouse with two associated 20th-century outbuildings. Methods Archaeological fieldwork at Garrett‘s Chance involved three phases of work. Phase 1 consisted of surface collecting of the cultivated fields and machine-disturbed forested areas, and very limited shovel testing of some of the more level areas in the forest and pasture. With one assistant I walked all of the fields on transects 15 ft to 20 ft apart, flagging all observed cultural materials. If the material appeared to be clustered, I designated the cluster an archaeological site and drew the approximate extents on a 1:1200 scale topographic map supplied by the consulting engineers and collected all of the flagged objects. Scattered refuse was simply collected by field (designated A through N) for inventorying. In neither case were artifacts piece-plotted. The wooded parcels slated for development (numbered 1 through 4) were all steep. They were investigated using two methods: reconnaissance of the surfaces exposed by heavy equipment used in testing percolation rates, and judgmental shovel testing of a few locales that were level or nearly so. Two grassland lots with moderate slopes were shovel tested at 65 ft intervals (#s 5 and 6). All shovel tests, each measuring approximately 40 cm in diameter, were excavated stratigraphically, the soils described (Appendix A) and screened through ¼–inch hardware mesh. All shovel tests in the wooded and grassland lots proved negative. Shovel tests used for site examination were excavated at 25 ft intervals and in the same manner of the judgmentally placed woodland units. Those at Garrett‘s Chance #2 (18PR703) were supplemented with a 3 ft by 4 ft excavation unit over the top of a possible feature and in the part of the site with the highest apparent density of surface material. Only plowzone was excavated, the underlying feature left untouched. The steep and clearly eroded slopes to the west and south were tested to demonstrate soil loss. Shovel tests at Garrett‘s Chance #3 (18PR704) were supplemented by nine 2½ ft by 2½ ft excavation units that sampled, in systematic aligned fashion, the thin plowzone for additional prehistoric materials. Artifact yields from shovel testing and the one 3 ft by 4 ft excavation unit at Garrett‘s Chance #2 (18PR703) proved to low to indicate intensive plowzone sampling to investigate possible activity patterning, nor were the shovel-test recovery rates large enough to permit meaningful computer-aided simulations. I used machine stripping of the plowzone to examine the extent of the site and condition of sub-plowzone features. Stripping involved use of a front-end loader that first removed the bulk of the plowzone, then—operating in reverse and dragging its tripartite bucket—scraping the subsoil. I flagged and subsequently defined and mapped the features comprising the footprint of an earthfast house with a burned daub filled root cellar and a series of intersecting borrow pits containing greater and lesser amounts of household refuse seemingly dating to the early 18th century. All of the larger pits and the root cellar, as well as one structural posthole and mold, were excavated stratigraphically and in entirety. All artifacts recovered from the field surfaces, shovel tests, excavation units, and features were collected, cleaned, and inventoried. The property owner will donate the assemblages to the State of Maryland and they will be curated at the Maryland Archaeological Conservation Laboratory.

36

Chapter 5. Results of Survey Introduction This chapter reports the findings of the first phase of field and laboratory work. It describes the initial survey, breaking down the project area (Figure 5-1) into fields (A-N) and non-cultivated study areas (1-6). None of the study areas—cultivated or non- cultivated—has been rigorously defined: irregularities of topography and diffuse scatters of cultural material made such definitions unnecessary. Site boundaries, however, were defined more topographically and in terms of the apparent distribution of cultural materials. Chapters 6 through 9 address all phases of work at each of the archaeological sites and Chapter 10 provides details on of above-ground architecture. Figures 5-2 and 5- 3 illustrate the typical settings for surface collecting at Garrett‘s Chance.

Figure 5–1. Survey locations A-N and 1-6.

37

Figure 5–2. Cultivated field at Garrett‘s Chance. N.B. Tracks from percolation testing machinery.

Figure 5–3. Area K, looking southeast.

Garrett’s Chance Field A Field A ranges in elevation from 130 to 159 ft amsl with an 11% slope, and it includes approximately 0.9 acres. It is 200 ft and more from the head of a small drainage that empties into a branch of Swanson Creek. The soils are gravelly sand loams, severely eroded and, in places, gullied. Surface inspection yielded a few prehistoric or historic artifacts, widely dispersed (Table 5-1). Collecting conditions were excellent. The

38

recovery rates for prehistoric and historic artifacts were 7.8 and 16.7 artifacts per acre, respectively. Table 5-1 Artifacts recovered from Field A. Class Sub-Class Variety Type Form Quantity Lithic Fire-cracked rock quartzite 2 Flake Decortication quartz 1 Primary quartz 2 Shatter quartz 1 Tertiary quartz 1 Flake Total 5 Lithic Total 7 Ceramic Earthenware White granite transfer-printed bowl 1 Whiteware indeterminate indeterminate 6 Stoneware American gray indeterminate hollowware 4 Ceramic Total 11 Vessel Glass Vessel glass Beverage bottle indeterminate hollowware 1 machine-made flask 1 indeterminate indeterminate hollowware 1 machine-made hollowware 1 Vessel glass Total 4 The field crew recovered 15 sherds of ceramic and glass vessels dating to the late 19th and 20th centuries. The seven lithic artifacts represent early and late stages of lithic reduction and possibly food processing or preparation (the fire-cracked rock). Lacking temporal diagnostics, I cannot say with confidence that the flakes are associated with the fire- cracked rock: they may represent more than one occupation; therefore, these dispersed objects cannot be designated a lithic scatter, short-term resource procurement site, or any other category of site type. Garrett’s Chance Field B Field B ranges from 130 to 159 ft in elevation with a slope of 11%, and it includes approximately 1.3 acres. It includes a cultivated slope between two drainages and lies just about 100 ft above the flood plain of the branch of Swanson Creek. Soils are sandy loams with considerable gravel. The soils are eroded and gullying continues with each substantial rain. Collecting conditions were excellent (Table 5-2). The recovery rates for prehistoric and historic artifacts were 3 and 6.7 artifacts per acre, respectively.

Table 5-2 Artifacts recovered from Field B. Class Sub-Class Variety Type Form Quantity Architecture Brick Common red 1 Ceramic Earthenware White granite indeterminate cup 1 Whiteware indeterminate flatware 4 indeterminate 1 sponged hollowware 1 Porcelain indeterminate indeterminate indeterminate 1 Ceramic Total 8 Lithic Shatter cortical 4

39

Field B yielded surprisingly few prehistoric artifacts, and their identities as byproducts of lithic reduction are problematic. The setting appears consistent with the kinds of landforms commonly occupied in Southern Maryland by indigenous peoples, especially during the Late Archaic through Early Woodland periods. It is possible that there were more suitable locations along the main branch of the creek. Alternatively, extensive erosion has obliterated most of the evidence for prehistoric occupation. The one brick fragment and eight ceramic sherds were scattered throughout the field and likely represent late 19th-century manuring of the field; i.e., spreading of manure from sties and stalls where household refuse, including stray pieces of inedible trash, were fed to livestock (see Gibb 1996: 149ff for a discussion of this largely overlooked transformational process). Garrett’s Chance Field C Field C ranges between 132 and 146 ft in elevation with slopes between 9% and 13%, and it includes approximately 2 acres. A portion of this area has been designated Garrett‘s Chance # 4 (18PR709) and I discuss that portion with the other formally designated sites. The material recovered from Area C and treated in this section consists of only seven ceramic and glass vessel sherds on a rise just east of the main site (Table 5- 3). The surface was heavily eroded and gravelly sand. The artifacts, though few, point to a mid to late 19th century origin. They likely are related to the main site. The recovery rates for prehistoric and historic artifacts were 0 and 3.5 artifacts per acre, respectively.

Table 5-3 Artifacts recovered from Field C. Class Sub-Class Variety Quantity Ceramic Earthenware Plain redware 1 Whiteware 3 Stoneware American gray 1 Vessel Glass Vessel glass indeterminate 1 Table glass 1 Garrett’s Chance Field D Field D is a small area due north of a ruined tobacco barn. It ranges in elevation between 110 and 130 ft, with a slope of 8%, and it includes approximately 1.6 acres. Few artifacts were collected from this location, despite good to excellent collecting conditions on the sandy loam surface. The few materials collected have been combined with those from Garrett‘s Chance #4 (18PR709, discussed below): they appeared in the field to have been transported farther downslope from the site by plowing and erosion. Garrett’s Chance Field E Field E occupies a slope above the main channel of Swanson Creek, between 82 and 110 ft above mean sea level, with a 10% slope, and it includes approximately 1.94 acres. Its setting is similar to that of Field K (see below), but more elevated and distant from the creek than Garrett‘s Chance #1 (18PR702, see below). The soil is sandy loam with some gravel and considerable gullying. Artifacts were few and diffuse (Table 5-4). A prehistoric presence is uncertain; the four pieces of cortical shatter may be byproducts of plowing. The few historic materials might be classified as late 19th/early 20th century

40

domestic refuse, and their presence in the field may be attributed to manure spreading. The recovery rate for both prehistoric and historic artifacts was two per acre. Table 5-4 Artifacts recovered from Field E. Class Sub-Class Variety Type Form Quantity Lithic Shatter quartz indeterminate 4 Ceramic Earthenware Whiteware indeterminate indeterminate 3 Vessel Glass Vessel glass Case bottle paneled pharmaceutical 1 Garrett’s Chance Field F Field F (about 2.3 acres) was arbitrarily defined during fieldwork to more accurately and precisely note the distribution of surface materials, distinguishing them from those in the northern half of the field (Field E; see above). In retrospect, we could have saved ourselves the bother: only a few prehistoric and historic artifacts were recovered (Table 5-5). Conditions were nearly identical with those observed for Field E, although the elevation range and slope were somewhat less (82 to 100 ft, 9% slope). Table 5-5 Artifacts recovered from Field F. Class Sub-Class Variety Type Form Quantity Ceramic Earthenware White granite indeterminate hollowware 2 Whiteware indeterminate indeterminate 2 transfer-printed bowl 1 Stoneware American gray painted hollowware 1 Ceramic Total 6 Lithic Biface quartz 1 Flake Decortication quartzite 1 Shatter quartz 8 Lithic Total 11 Again, most if not all of the shatter may have been caused by plowing. Only one likely decortication flake and a biface fragment were recovered from Field F. Six ceramic sherds were also recovered, all attributable to the late 19th and early 20th centuries and likely products of manure spreading. The recovery rates for prehistoric and historic artifacts were 4.8 and 2.6 artifacts per acre, respectively. Garrett’s Chance Field G Field G is a small plowed surface north of the Compton farmstead and Field N, ranging from 138 to 148 ft in elevation with a slope of 11%, and it includes approximately 1.2 acres. The surface is very gravelly and gullied, exposing deeply buried fine sand deposits in places. It might best be described as the base of a knoll, just above an eastward trending drainage. Much of the material likely eroded from the hill above, especially the historic artifacts (Table 5-6). These include 15 ceramic sherds, all but one of which (a painted body sherd from a large Chinese bowl) are attributable to the late 19th and early 20th centuries. A porcelain door knob and a porcelain electrical insulator certainly date to the late historic period. The locations of these finds bear no direct behavioral relationship to the behaviors that led to their deposition; viz., plowing and erosion displaced the material.

41

The field crew recovered 11 lithic artifacts from the surface of Field G, including a quartz Piscataway point and a black chert side-notched point that might have been a Lamoka point. It lacks a tip and is spalled. I found it in a small gully. A quartz biface base and flakes representing all stages of lithic reduction were recovered, albeit in small numbers and without any spatial clustering. Materials represented include quartz, fine- grained quartzite, rhyolite, and chert. Only one fire-cracked rock was recovered. Given the highly eroded setting, I will not hazard an interpretation of the prehistoric finds, but will point out they are consistent with the general pattern in the region of Late Archaic through Early Woodland occupation of the interior uplands around drainage heads. The finds represent at least two occupations and the recovered materials were diffuse to the point that many of the non-diagnostic prehistoric artifacts may be unrelated to both. Given the slope and very severe erosion, these artifacts bear no direct spatial relationship to the behaviors that led to their original deposition. The recovery rates for prehistoric and historic artifacts were 9 and 16.4 artifacts per acre, respectively. Table 5-6 Artifacts recovered from Field G. Class Sub-Class Variety Type Form Quantity Lithic Biface quartz 1 Fire-cracked rock quartz 1 Flake Decortication quartz 1 quartzite 1 rhyolite 1 Primary quartz 2 Shatter quartz 1 Tertiary quartz 1 Flake Total 7 Projectile point Piscataway quartz 1 Lamoka side-notched? chert 1 Lithic Total 11 Architecture Misc. modern Misc. modern machine-made porcelain insulator 1 Ceramic Demi-Porcelain indeterminate indeterminate flatware 2 Earthenware indeterminate burned flatware 1 indeterminate 1 White granite indeterminate flatware 1 indeterminate 1 Whiteware indeterminate hollowware 1 indeterminate 2 Porcelain Chinese painted large bowl 1 indeterminate painted saucer 1 Stoneware Alkaline glazed indeterminate hollowware 1 American gray indeterminate hollowware 3 Ceramic Total 15 Furnishing Misc. modern Misc. modern machine-made porcelain door knob 1 Vessel Glass Vessel glass indeterminate indeterminate hollowware 1 machine-made hollowware 2 Vessel Glass Total 3

42

Garrett’s Chance Field H Field H is a large (about 3.1 acres), steeply (15%) sloped expanse of cultivated field ranging between 100 and 130 ft above mean sea level, just above the confluence of a tributary of Swanson Creek and one of the largest drainages in the project area. This south-trending drainage probably served as a boundary between the Stone and Compton farms through the second half of the 19th century and the whole of the 20th. The sandy loam soil is very gravelly, gullied in places, and slopes down to the east and south. Collecting conditions were excellent, but the finds few and uninformative (Table 5-7). The recovery rates for prehistoric and historic artifacts were 0.3 and 1.6 artifacts per acre, respectively. Table 5-7 Artifacts recovered from Field H. Class Sub-Class Variety Type Form Surface Lithic Core quartzite 1 Ceramic Earthenware Whiteware indeterminate indeterminate 1 Vessel Glass Vessel glass indeterminate burned indeterminate 1 indeterminate indeterminate 2 jug handle 1 Vessel Glass Total 4 Garrett’s Chance Field I Field I consists of a knoll, the top of which is occupied by a tobacco barn. The elevations range from 146 to 168 ft above mean sea level with an overall slope of 10%, and as much as 15% in many places. The field is about 2.6 acres. The field crew identified a prehistoric site (Garrett‘s Chance #3 [18PR704]) on top of the knoll and that area is considered separately below. The knoll-top topography and soils are strikingly different from those of the slope and base of the knoll. The latter are steep, very gravelly sandy loams with considerable gullying. Table 5-8 enumerates finds from the slopes and more level area near the access road: it does not include material from 18PR704. The recovery rates for prehistoric and historic artifacts were 1.9 and 2.7 artifacts per acre, respectively. Table 5-8 Artifacts recovered from Field I. Class Sub-Class Variety Type Form Quantity Lithic Core quartz 1 Core Total 1 Flake Decortication quartz 1 Primary quartz 2 Flake Total 3 Shatter quartz 1 Lithic Total 5 Ceramic Earthenware White granite (blank) indeterminate 1 Whiteware indeterminate indeterminate 2 Stoneware American gray (blank) pan 1 Ceramic Total 4 Vessel Glass Vessel glass indeterminate machine-made indeterminate 3

43

Field I yielded a smattering of late 19th/early 20th century ceramic and glass sherds, and a few prehistoric lithic flakes, scattered on a surface of whole and plow-cracked quartz cobbles and gravel. Given the steep slopes and highly eroded condition of the soil, these finds probably are spurious and insignificant. Garrett’s Chance Field J Field J is a small (about 0.6 acres) narrow field above the farm road and below the knoll designated Field I. Like Field I, Field J is heavily eroded, its 15% slope ranging between 140 and 150 ft above mean sea level. The field crew recovered a single piece of cortical shatter, likely a byproduct of plowing this gravelly slope. Garrett’s Chance Field K Field K is a broad field, between 4¼ acres and 4½ acres. It slopes westward to Swanson Creek, from 140 to 80 ft above mean sea level. Its western and eastern thirds are steep (17% and 10%, respectively), and the middle third is relatively level at 5%. Soils are sandy loams, the steeper slopes comprised of very gravelly sandy loams. Recovered artifacts include a tumbler sherd and a sherd of undecorated pearlware plate, as well as three shattered flakes, four pieces of cortical shatter, and a very well shaped, finely flaked jasper end-scraper (Table 5-9). The isolated finds were dispersed and few (n=8 prehistoric artifacts over more than four acres, or two per acre, and four of those are quartz shatter that may be products of plowing). The historic materials probably are byproducts of manuring and the prehistoric materials indicate short-term occupations of unknown vintage. The recovery rates for prehistoric and historic artifacts were 1.8 and 0.4 artifacts per acre, respectively. Table 5-9 Artifacts recovered from Field K. Class Sub-Class Variety Type Form Quantity Lithic Flake Shatter quartz fine-grained 1 quartzite fine-grained 1 veined jasper 1 Scraper veined jasper heeled 1 Shatter quartz 4 Lithic Total 8 Ceramic Earthenware Pearlware indeterminate plate 1 Vessel Glass Vessel glass indeterminate tumbler 1 Garrett’s Chance Field L Field L, just below and east of the Compton farmstead, was distinguished from the lower Field H (see above) on the basis of a steep intervening slope. Elevations range from around 135 ft above mean sea level at the access road to 170 ft near the front entrance to the houselot. Field L is about 3.4 acres. The soils are heavily eroded and gullied, and they are comprised of gravelly sandy loams. The few prehistoric finds may be plow-cracked pebbles and the half-dozen ceramic and vessel glass finds are all from the mid 19th through 20th centuries, probably products of manuring. The recovery rates for prehistoric and historic artifacts were 1.5 and 2.4 artifacts per acre, respectively.

44

Table 5-10 Artifacts recovered from Field L. Class Sub-Class Variety Type Form Quantity Lithic Flake Shatter quartz 5 Ceramic Earthenware Whiteware dipped hollowware 2 indeterminate indeterminate 1 Stoneware American gray indeterminate hollowware 1 Ceramic Total 4 Vessel Glass Vessel glass indeterminate indeterminate hollowware 1 machine-made pharmaceutical 1 Vessel Glass Total 2 Garrett’s Chance Field M Field M is a broad (2.5 acres), relatively level field in the easternmost part of what, until recently, was the Compton farm. It borders the south edge of the access road and the western margin of the large drainage that once marked the border between the Compton and Stone farms. Elevations range from 114 to 142 ft above nean sea level, with slopes of 5% to 8%. Soils range from fine sandy loam silt loams and clay loams in the western half to gravelly sandy loams in the east. The soils in the western half also appeared to be very dark and surface reconnaissance established that a level, relatively uneroded portion was a Colonial site (Garrett‘s Chance #2 [18PR703]) which is described below. For present purposes, Filed M refers to those portions of the field that do not include the Colonial site, and Table 5-11 does not include materials collected from within the site boundaries as described in the next chapter. The field crew recovered 16 pieces of flaked stone, including a quartz scraper. Six of the flakes were decortication flakes and eight were shatter. All of the prehistoric material was dispersed throughout the field. Of the six ceramic and glass vessel sherds recovered, the Rhenish gray and wine bottle sherds probably derive from Garrett‘s Chance #2 (18PR703). The remaining sherds are likely products of manuring. As analysis of the Garrett‘s Chance #2 (18PR703) materials will demonstrate, the refined white paste earthenware and machine-made/hand-tooled lip bottle sherds are unrelated to that site. The recovery rates for prehistoric and historic artifacts were 6.4 and 2.4 artifacts per acre, respectively. Table 5-11 Artifacts recovered from Field M. Class Sub-Class Variety Type Form Quantity Lithic Flake Decortication chert 1 quartz 5 Primary rhyolite 1 Secondary quartz 1 Shatter quartz cortical 6 quartzite cortical 1 Scraper quartz scraper 1 Ceramic Earthenware White/Pearl ware plate 1 Stoneware Rhenish gray mug 1 Vessel Glass Vessel glass Beverage bottle machine-made tooled lip 1 indeterminate indeterminate indeterminate 1

45

Wine bottle indeterminate bottle 2 Garrett’s Chance Field N Field N borders the north side of the Compton farmstead homelot. The field slopes (9%) down toward Field G and ranges in elevation from 158 to 170 ft above mean sea level, and it includes approximately 0.8 acres. Only six flakes, four of which were cortical shatter, were found (Table 5-12). All of the historic materials, including a ―sad‖ iron (a stove-warmed iron for pressing fabrics) and several sherds of demi-porcelain and one of alkaline-glazed stoneware, date sometime between the late 19th and mid-20th centuries. The dates of these objects are consistent with expectations for the occupation of the early 20th century Compton House (see below). The recovery rates for prehistoric and historic artifacts were 7.8 and 31.2 artifacts per acre, respectively. Table 5-12 Artifacts collected from Field N. Class Sub-Class Variety Quantity Lithic Flake Primary 2 Shatter 4 Ceramic Demi-Porcelain indeterminate 2 Demi-Porcelain Total 2 Earthenware Whiteware 11 Porcelain Misc. modern 1 Stoneware Alkaline glazed 1 American gray 2 Furnishing Sad Iron 1 Vessel Glass Vessel glass indeterminate 3 Misc. modern 1 Sample Area 1 Sample Area 1 corresponds to Proposed Lot 9, which borders the tributary of Swanson Creek. It is entirely wooded and very steep, except for a narrow ridge (125 ft by 300 ft on which a tobacco barn is situated (see below). This relatively level area has been cleared in recent years—or kept clear while the surrounding area reforested—for use as a paint gun range, and percolation testing equipment has roiled the surface, thereby facilitating surface collecting. Elevations range between 140 and 160 ft above mean sea level, with a 6 % slope down to the south. The barn is at the south end of the ridge (Figure 5-4). Soil scientists cleared a track across the ridge to admit a drilling rig. The field crew minutely examined this track for cultural material, but found only two ceramic sherds, despite better than 80% surface visibility throughout (Table 5-13). The undecorated White Granite rim and black transfer-printed white earthenware probably date to the second half of the 19th century. We did find a small pile of handmade common red bricks with frogs (faceted depressions designed to promote drying and firing of the green brick and to provide greater surface area for bonding) about 75 ft northeast of the northeast corner of the barn on the edge of the clearing. The bricks lack mortar and appear to have been stockpiled. I suspect that they were placed here by one of the Compton family. The family has demonstrated its interest in Colonial Revivalism in its treatment of the Compton House interior.

46

Figure 5-4. Shovel-test sample area 1.

Table 5-13 Artifacts collected from Sample Area 1 (surface). Ceramic ware Type Form Element Decoration Color Quantity White granite molded flatware rim indeterminate 1 transfer- Whiteware printed flatware base landscape black 1 A single shovel test near the brick pile revealed a brown (10YR4/3) gravelly sandy loam topsoil, 0.65 ft thick, on a dark yellowish brown (10YR4/4) gravelly clay loam subsoil, but no cultural material. The shovel test confirmed the field crew‘s observation of the graded track: the surface had been severely eroded. Sample Areas 2 through 6 all lie within the confines of the Stone farm. The eastern 21-acres of the Stone farm, now called Sunnyside, is not slated for development. It is occupied by the c.1844 house of Dr. Michael Jenifer Stone and associated outbuildings. Sample Area 2 Sample Area 2 is entirely wooded and lies just east of Field H. Most of the parcel (Lot 5) is steep (>20%) and severely eroded, with elevations ranging between 100 and 130 ft above mean sea level (Figure 5-5). The field crew examined a freshly graded track that led to proposed septic field locations and collected five ceramic sherds scattered over approximately 150 linear feet of graded track (Table 5-14). We found no other historic material, architectural or domestic, and no prehistoric artifacts.

47

Figure 5-5. Shovel-test sample areas 2 through 4. Table 5-14 Artifacts collected from Sample Area 2. Ceramic ware Type Form Element Decoration Color Quantity Pearlware shell-edged plate rim edged blue 1 Whiteware Old Blue plate marley indeterminate dark blue 1 transfer- Whiteware printed plate marley floral green 1 White granite molded flatware rim indeterminate 1 White granite indeterminate flatware rim undecorated 1 A single shovel test pit on the most level portion of the sample area revealed a thin (0.2 ft) dark grayish brown (10YR3/2) humus on 0.45 ft of brown (10YR4/3) fine sandy loam, and a dark yellowish brown (10YR4/4) fine sandy loam subsoil. No artifacts were recovered. The soils suggest a recently stabilized surface that, prior to reforestation, had deflated. The artifacts from the graded track are inconsistent with those recovered from Fields H and M, to the west and north, respectively, and intensive surface reconnaissance failed to yield any evidence of historic occupation of Sample Area 1. Sample Area 3 Sample Area 3 corresponds to proposed Lot 4. It is entirely wooded and steep (13%), with an 80 ft diameter knoll top at 168 ft above mean sea level. Surface reconnaissance revealed no evidence of human occupation. Two shovel-tests, excavated approximately 65 ft apart, revealed gravelly loamy sands, but no cultural material (see Figure 5-5; Table 5-15).

48

Table 5-15 Shovel test data for Sample Area 3. Unit Stratum Feet below grade Munsell value Texture STP1 1 0.45 10YR4/3 Gravelly loamy sand STP1 2 1.00 10YR5/3 Gravelly sand STP2 1 0.85 10YR2/2 Gravelly loamy sand STP2 2 1.10 10YR4/3 Gravelly sand Sample Area 4 Sample Area 4 is entirely wooded and corresponds to proposed Lot 3. The land is very steeply sloped (<25%) with an elongated knoll near the center of the lot, 220 by 60 ft, at an elevation of 175 ft above mean sea level. Surface reconnaissance revealed no evidence of human occupation. Three shovel tests, excavated approximately 65 ft apart, produced no cultural materials, revealing only dark gravelly loamy sands and clay loams (see Figure 5-5; Table 5-16). Table 5-16 Shovel test data for Sample Area 4. Unit Stratum Feet below grade Munsell value Texture STP1 1 0.40 10YR3/2 Gravelly loamy sand STP1 2 0.90 10YR4/4 Gravelly sand STP2 1 0.70 10YR2/1 Gravelly loamy sand STP2 2 1.00 10YR4/6 Gravelly clay loam STP3 1 0.15 10YR2/1 Sandy loam STP3 2 0.50 10YR4/4 Gravelly loamy sand STP3 3 1.10 10YR4/6 Gravelly clay loam Sample Area 5 Sample Area 5 is mixed meadow and woodland and is in proposed Lot 2. The wooded portions are steep, with slopes up to 13%, while the southward sloping meadow is only about 5%. Six shovel tests—excavated approximately 65 ft apart on two transects in the meadow—exposed very dry, gravelly sands and loamy sands, but no cultural material (Figure 5-6; Table 5-17). Elevations range between 132 and 140 ft above mean sea level. The area probably experienced considerable erosion prior to stabilization. The meadow is planted in hay, and probably has been for many years. Table 5-17 Shovel test data for Sample Area 5. Unit Stratum Feet below grade Munsell value Texture STP1 1 1.30 10YR4/3 Gravelly loamy sand STP1 2 1.70 10YR4/4 Gravelly sand STP2 1 0.90 10YR4/4 Gravelly sand STP3 1 0.70 10YR4/4 Gravelly loamy sand STP3 2 1.00 10YR4/6 Gravelly sand STP4 1 0.70 10YR4/4 Gravelly loamy sand STP4 2 1.10 10YR4/4 Gravelly sand STP5 1 0.90 10YR4/3 Gravelly loamy sand STP5 2 1.30 10YR4/4 Gravelly sand STP6 1 0.70 10YR4/3 Gravelly loamy sand STP6 2 0.90 10YR4/4 Gravelly sand

49

Figure 5-6. Shovel-test sample areas 5 and 6. Sample Area 6 Sample Area 6 is mostly moderately sloped (7%) meadow land, with some steeper woodland (<33%). Elevations range between 168 and 176 ft above mean sea level. Six shovel tests—excavated approximately 65 ft apart on two transects in the meadow— exposed very dry, clay and silt loams devoid of cultural material (see Figure 5-6; Table 5- 18). The area probably experienced considerable erosion prior to stabilization. I attribute differences in soil texture between this area and nearby Sample Area 5 to differences in elevation (about 36 ft), units in the two areas sampling different geological beds. Table 5-18 Shovel test data for Sample Area 6. Unit Stratum Feet below grade Munsell value Texture STP1 1 0.60 10YR4/4 Silt loam STP1 2 1.00 10YR4/6 Clay loam STP2 1 0.50 10YR4/4 Silt loam STP2 2 0.70 10YR4/6 Clay STP3 1 0.25 10YR4/4 Silt loam STP3 2 0.40 10YR4/6 Clay STP4 1 0.50 10YR4/3 Silt loam STP4 2 0.70 10YR4/6 Clay STP5 1 0.60 10YR4/3 Silt loam STP5 2 0.80 10YR4/4 Clay loam STP6 1 0.50 10YR4/3 Silt loam STP6 2 0.80 10YR4/4 Clay loam

50

Summary of Results Shovel testing revealed no cultural resources. Surface reconnaissance, however, proved very effective in identifying light scatters of both prehistoric and historic materials (Table 5-19). These scatters result, in part, from initial cultural transformation processes (short-term, low-intensity prehistoric occupations and manure spreading during the 19th and early 20th centuries), and from severe, long-term erosion of the steep slopes and unconsolidated sandy soils. Table 5-19. Prehistoric and historic recovery rates. Area/Site Acres Prehistoric/acre Historic/acre A 0.9 7.8 16.7 B 1.3 3.0 6.7 C 2.0 0 3.5 D 1.6 N/A N/A E 1.9 2.0 2.0 F 2.3 4.8 2.6 G 1.2 9.0 16.4 H 3.1 0.3 1.6 I 2.6 1.9 2.7 J 0.6 0 0 K 4.5 1.8 0.4 L 3.4 1.5 2.4 M 2.5 6.4 2.4 N 0.8 7.8 31.2 18PR702 .74 90.5 0.7 18PR703 1.43 0 78.3 18PR704 .23 330.4 13.0 18PR709 .43 0 435.0 Surface reconnaissance and collecting also resulted in the identification of four archaeological sites. These sites, and their eligibility for listing on the National Register of Historic Places, are the subjects of the next four chapters.

51

Chapter 6. Garrett’s Chance #1 (18PR702) Introduction Garrett‘s Chance #1 (18PR702) was identified in the lower portion of Field K on a low bench paralleling Swanson Creek (Figure 6-1). While part of proposed Lot 17, this narrow (420 by 80 ft) field is not slated for development. It slopes moderately (<8%) to the west and to the creek. The soils are gravelly sands, classified as Fallsington sandy loam. No subsurface testing was undertaken, only surface collecting. Artifacts were flagged and their distributions noted before they were collected. Results The field crew identified and collected three loci at Garrett‘s Chance #1 (18PR702). These loci, separated by small gullies, may result from erosion rather than represent distinct occupational loci. Only lithic materials were in evidence, totaling 62 pieces, of which 26 were fire-cracked rocks and three were projectile points (Table 6-1). Locus 1, the northernmost of the three, yielded 29 artifacts, including possible (neither is well formed) Bare Island and Piscataway projectile points (Figure 6-2). Twelve fire-cracked rocks comprise 41% of the assemblage. Decortication, primary and secondary flakes, as well as one indeterminate piece of shatter, complete the list. Only quartz and quartzite, likely of local origin, are represented. Locus 2 was identified on the basis of six pieces of fire-cracked rock, four large decortication flakes, and a piece of cortical shatter. Its small size, low yield, and proximity to Locus 1 suggest that it is part of the larger locus. Locus 3 is the largest and southernmost of the three loci. It yielded 27 objects, including a quartz Piscataway point (see Figure 6-2) and eight fire-cracked rocks. Like Locus 1, it also produced flakes representing the principal stages of lithic reduction, including four tertiary flakes suggestive of tool maintenance.

Figure 6-1. Map of Garrett‘s Chance #1 (18PR702).

52

Table 6-1 Artifacts collected from Garrett‘s Chance #1 (18PR702). Class Locus Sub-Class Variety Type Form Quantity Lithic 1 Fire-cracked rock (blank) (blank) quartz 3 quartzite 9 Fire-cracked rock Total 12 Flake Decortication (blank) quartz 5 quartzite 5 Decortication Total 10 Primary (blank) quartz 1 Primary Total 1 Secondary (blank) quartz 3 Secondary Total 3 Shatter (blank) quartz 1 Shatter Total 1 Flake Total 15 Projectile point (blank) poss Bare Island quartz 1 poss Piscataway quartz 1 Projectile point Total 2 1 Total 29 2 Fire-cracked rock (blank) (blank) quartz 3 quartzite 3 Fire-cracked rock Total 6 Flake Decortication (blank) quartz 1 quartzite 3 Decortication Total 4 Shatter (blank) quartzite 1 Flake Total 5 2 Total 11 3 Fire-cracked rock (blank) (blank) quartzite 8 Fire-cracked rock Total 8 Flake Decortication (blank) chert 1 quartz 5 quartzite 1 Decortication Total 7 Primary (blank) quartz 2 Primary Total 2 Secondary (blank) quartz 2 Secondary Total 2 Shatter (blank) quartz 3 Shatter Total 3 Tertiary (blank) quartz 4 Tertiary Total 4 Flake Total 18 Projectile point (blank) Piscataway quartz 1 3 Total 27 Lithic Total 67

53

Analysis and Interpretation Together, the three loci—covering about three-quarters of an acre— evidence stone tool making, probably from locally gathered or quarried quartz and quartzite pebbles, and from at least one chert pebble, probably acquired locally as well. Although collecting conditions were excellent, the relatively few tertiary and secondary flakes—representing late stages of lithic reduction—were found. Given the small sizes of these types of flakes, their under-representation is not surprising. The three loci together yielded a recovery rate of 90.5 artifacts per acre. The three bifaces recovered all appear to be projectile points, although the designation of Bare Island bifaces as projectile points is problematic. Their characteristic asymmetry and uncertain vintage suggests that they were hafted bifaces used as hand tools, rather than as projectiles, over many centuries. The temporal placement of Piscataway points is also problematic (e.g., Ebright 1992: 38; Reeve 1992). Based on these two enigmatic point types, and keeping in mind that none of the three points can be definitively classified, and lacking evidence of steatite or clay vessels, I have tentatively identified 18PR702 as a Late Archaic period site.

Figure 6-2. Projectile points from Garrett‘s Chance #1 (18PR702). Fire-cracked rock points to activities that go beyond stone procurement and tool making. The site‘s location near the bank of Swanson Creek suggests efforts to harvest aquatic resources (fish and plants), as well as hunting of game

54

attracted by the riparian environment. Presumably, when the site was occupied, the level of Swanson Creek was higher and the water flowed faster and more clearly, unencumbered by agricultural sediments. Garrett‘s Chance #1 (18PR702) may be eligible for listing on the National Register of Historic Places, per Criterion D, if it proves to have potential to contribute to our knowledge of Late Archaic through Early Woodland settlement and subsistence patterns in the interior upland of the Patuxent drainage. Clear evidence of erosion and deep plowing into the sand and gravel substrate, however, does not bode well for the preservation of deposits. No further testing is recommended as long as the site will be preserved in the wetland buffer along Swanson Creek.

55

Chapter 7. Garrett’s Chance #2 (18PR703) Introduction The southern end of Field M, viewed from the farm road, had darker soils then those from anywhere else in the project area (Figure 7-1). Surface collecting of approximately 1.4 acres yielded a small, but unmistakable, collection of Colonial period materials. Recollection of the field after intense rains yielded more of the same, for a combined inventory of over one hundred objects (Table 7-1), including Rhenish blue and gray stoneware, British brown and white salt-glazed stonewares, and tin-glazed and lead- glazed earthenwares (approximately 78.3 artifacts per acre, based on repeated passes at 5- ft intervals and on perpendicular transects). Conspicuous by their absence were Queensware and Pearlware, suggesting a late 17th through mid-18th century date for the site occupation.

Figure 7-1. Location of Early Colonial surface scatter. Table 7-1 Select categories of objects surface collected. Category Quantity Rhenish blue/gray stoneware 13 British brown stoneware 2 White salt glazed stoneware 3 Tin-glazed Earthenware 2 Plain red earthenware 1 Wine bottle glass 14 Table glass 7 Vial 1 Tobacco pipe 63 Handwrought Nail 6 Total 112

56

This section documents the results of successive phases of shovel testing, mechanical stripping and mapping, feature excavation, and analysis. Archival Record The historical record pertaining to the project area, and specifically to the tracts called Dove‘s Nest and Dove‘s Perch, has been explored in Chapter 3. A few salient points, as they pertain to the archaeological findings at this historic site—Garrett‘s Chance #2 (18PR703)—warrant restatement. Dove‘s Nest, the parcel that appears to have included the eastern portion of the Compton farm, was patented in 1664 and remained in the hands of the Truman family until Bernard Johnson purchased it in 1686. His heirs—four daughters and their spouses—divided all of his lands equally in 1702, two of the daughters and their husbands selling their divided interests to a major local landowner, William Wilkerson, in 1711. The remaining heirs eventually sold their shares as well, although to whom and when I have not determined. Bernard Johnson owned 350 acres between the two tracts, Dove‘s Nest and Dove‘s Perch. Surviving documents located during this research do not specify where Johnson, his heirs, or Wilkerson lived. Johnson appears to have used Dove‘s Nest as his dwelling plantation, but the Compton farm comprises one-half or less of that tract. Wilkerson almost certainly lived elsewhere. As will become evident from the ensuing analysis, no artifacts were recovered that might identify the occupants of the site (e.g., wine bottle seals), and there were at least two successive occupations of the site. Bernard Johnson may have been the builder and short-term occupant of the farmstead, but his successors likely were tenants, his or those of his heirs and William Wilkerson. Research Design and Methods Garrett‘s Chance #2 (18PR703) was identified on the basis of a light scatter of early Colonial artifacts (n=13) and non-diagnostic prehistoric artifacts (n=4) on the surface of the recently plowed field. Soils at the highest point of the field also were darker than those in the immediate vicinity, suggesting at least one subsurface feature with ashy fill that had been truncated by the plow. Repeated surface collecting narrowed the area of interest to approximately 1.5 acres, the southern third of which was severely eroded. Shovel-testing at close intervals (25-ft) further narrowed the area in which artifacts and burned daub could be recovered, approximately 50-ft by 50-ft. The shovel tests also demonstrated a very low density of material, an observation supported by the excavation of a unit measuring 4-ft by 3-ft where shovel-testing indicated the highest concentration might be found. The yield was low: 86 artifacts, of which 63 were small fragments of daub. The field crew also recovered 23 tobacco pipe, vessel glass, and nail fragments. The unit did uncover part of a feature of indeterminate size and function. The number and distribution of surviving features remained unknown. The data, therefore, suggested a low-density housesite with wattle and daub construction. Anticipated finds included posthole and mold complexes defining one or more buildings, borrow pits (possibly filled with refuse), and fenceline postholes. Artifacts suggested occupation in the late 17th and early 18th centuries. Archival research and preliminary archaeological investigation indicated that Garrett‘s Chance #2 (18PR703) did not meet eligibility Criteria A or B for listing on the

57

National Register; its occupants were not particularly prominent in the history of the area and no important historical events occurred at the site. The site had the potential to contribute to our knowledge of architectural patterns in the Chesapeake region, especially in the interior uplands, during a period of increasing social, economic, and political stability; and as such, met Criterion C. Severe erosion of the site along its southeastern and southwestern margins, and the relatively low-yield of artifacts from the plowzone, indicated that the site likely had little potential for revealing patterns of homelot organization, which also would have made the site eligible under Criterion C. The likely survival of intact subsurface features held the promise of yielding information on patterns in the use of wealth and food acquisition, making the site eligible for National Register listing under Criterion D. The applicant, in preparing the application for subdivision under the new policies of the County, determined that subdivision and development would not be feasible because Garrett‘s Chance #2 (18PR703) was in the midst of one of the parcel‘s two premier lots. Having exhausted redesign options, the applicant requested mitigation of impacts to the site and authorized an intensive study of Garrett‘s Chance #2 (18PR703). Andrew Garte & Associates presented a scope of work with several implied research questions, which the applicant accepted, that involved stripping of the site to expose all surviving features, mapping of the features, and sampling of each to recover those data for which the site appeared to be eligible for National Register listing. The site, situated more than three miles from the Patuxent River and dating to the late 17th and early 18th centuries, represented a third generation settlement; a period of natural population increase, decline of indentured servitude, increase in slavery, and a stable staple economy of tobacco agriculture supplemented with maize and livestock rearing. Little is know of the period in the interior and the King‘s Reach site (18CV83), on the east side of the Patuxent River, is the only extensively excavated domestic site of the period in the region (Pogue 1990). Pogue (1990) uncovered two earthfast structures, the larger of which comprised a hall and parlor arrangement with a pantry ell and a stud- in-trench rear shed addition. The smaller—a two-bay structure—was tied the larger dwelling by a succession of two foreyard fences. Both dwellings had earthen cellars, the larger dwelling exhibiting a succession of collapsed cellars. Pogue interpreted the complex as a planter house with adjacent quarter. Its cellars produced an array of household goods. Plowzone sampling produced distributions that Pogue (1990) interpreted as sheet middens, but much of the material likely was plowed out of the many artifact-rich cellars. Principal questions posed for Garrett‘s Chance #2 (18PR703) revolved around the architecture and houselot organization. 1. Assuming architectural complexity represents social complexity (e.g., household size, division of labor along lines of age and sex, cohabitation of planters and bonded labor): a. How many buildings were present? b. Were all the buildings heated; i.e., did they all have hearths? c. How large were those buildings? Do they exceed 40 ft by 20 ft, a substantial dwelling size for earlier periods? d. Do they exhibit common dimensions (e.g., multiples of 4 ft or a perch of 16.5 ft)?

58

e. Were those buildings internally differentiated; i.e., is there evidence of functionally specialized spaces, permanent divisions (e.g., walls), or ells? 2. In their seminal article on earthfast architecture, Carson et al. (1981) described techniques for building construction and inferred motivations of those who had those structures built. a. Were the structures built on independent posts or through the raising of bents or sidewalls? b. Is there evidence of building repair and maintenance, evidence running counter to the concept of impermanence and temporary residence? 3. Plowzone sampling and artifact distribution simulations have been successfully used on Colonial sites in the Chesapeake region to identify middens and activity areas (e.g., Keeler 1978; King and Miller 1987; King 1988); however, low artifact density and severe erosion of the southern portion of the site precluded effective use of that method. Distribution of subplowzone features provides the best approach in such cases for investigation of houselot organization. a. Borrow pits are an integral component of earthfast construction and investigators typically regard these features as indicators of a rear, or work, yard. Are borrow pits present and do they cluster on one side of the buildings? b. Is there evidence of privy pits and where are they located relative to the buildings? c. Are there fencelines and, if so, how do the fences define space around the buildings? 4. Historian Gloria Main (1982) and others, and a number of archaeologists (e.g., Pogue 1990; Gibb 1996), have examined the role of material wealth and its expressions in the Colonial Chesapeake region. How did the colonists invest wealth in foods and goods, and what were they trying to express through the acquisition and use of material culture? a. Is the material culture from the site heterogeneous and rich? (Surface collecting suggested that it was neither.) b. Did they acquire personalized objects such as wine bottles with custom seals or slipwares with names, initials, dates, or other commemorative texts? Full exposure of the features and of a sufficiently large area around them would provide information on architecture and, within the bounds imposed by extensive erosion, some data on homelot organization. Feature sampling would permit more accurate and precise dating then was possible from the small combined collections of the Phase 1 and Phase 2 studies. The recovered materials also had the potential for revealing patterns in the use of wealth; specifically, the choices in vessel acquisition, personal items, and foods. Ultimately most of the exposed pit features were excavated in their entirety, flotation samples were recovered, and—with funding provided by the principal investigator—a sample of those materials were identified by Justine Woodward McKnight, whose report appears as an appendix to this document. Field methods for the Phase 3 included: extensive mechanical stripping to expose buildings, pits and fencelines; gridding with a theodolite; cleaning and mapping of all features at a scale of 1:12; stratigraphic excavation of features; screening through ¼-inch hardware mesh; and recovery of several liters of feature soils with high charcoal content for flotation extraction of plant remains. Mechanical stripping was guided by topographic

59

considerations and clear evidence of erosion, not by surface or subsurface artifact distributions, both of which were too sparse to define the site extent. Recovered materials were processed in accord with the standards and guidelines promulgated by the Maryland Historical Trust. Drawings were digitized and a site map constructed of the individual feature drawings. The digitized site map was then placed on the digital property map using geo-referenced stakes placed around the site by the applicant‘s engineers. Descriptions of the features, their soils and inclusions, appear below. Field and Laboratory Results Following intensive surface collecting, the Andrew Garte & Associates field crew shovel tested the entire area in which artifacts were recovered, including some areas that were clearly eroded. Three dozen shovel tests yielded 51 artifacts, although 16 of those were small bits of daub that collectively weighed less than an ounce (Table 7-2). The daub offered strong evidence for at least one earthfast dwelling, the daub emanating from a wattle and daub chimney. Examining the distribution of plowzone depths, it was obvious why the recovery rate of less than one object per unit (daub excluded) was so low: the site had been severely eroded. The simulated contour map of plowzone depths in Figure 7-2 depicts the deepest deposits between grid coordinates N160 E160 and N180 E180. The heavier lines delineate the contours of plowzone 0.5 ft and more in depth, while the broken lines indicate exceptionally thin plowzone (<0.5 ft). Figure 7-3 represents the same data in three dimensions, the peaks representing the deepest plowzone deposits. A single excavation unit measuring 4 ft by 3 ft was excavated at N160 E 160 to test the deeper deposits and to examine a possible feature encountered by a shovel test of the same designation. A feature was partially uncovered that, subsequently, proved to be a structural posthole, one of six that eventually were found. The unit, however, produced only some daub and a handful of tobacco pipe bowl fragments, ceramic and glass vessel sherds, and nails; a disappointing return that indicated little value in intensive plowzone sampling. Table 7-2 Artifacts recovered through shovel testing. Sub-Class Variety Quantity Stoneware Rhenish gray 1 Stoneware British brown 1 Stoneware White salt glazed 1 Earthenware Slipped redware 1 Vessel glass indeterminate 3 Vessel glass Wine bottle 2 Tobacco pipe Ball clay 12 Bead indeterminate 1 Button 1 Brick Common red 1 Daub 16 Nail indeterminate 8 Window glass indeterminate 3 Total 51

60

200

190

180

170

160

h t

r 150

o N

140

130

120

110

100 100 110 120 130 140 150 160 170 180 190 200 East Figure 7-2. Contours of plowzone depths at Garrett‘s Chance #2 (18PR703).

Figure 7-3. Surface model of plowzone depths at Garrett‘s Chance #2 (18PR703). N.B. Depth exaggerated by factor of X10. Given the low artifact yield, the clear evidence of severe erosion, and the positive evidence of sub-plowzone features, we employed a front-end loader to first strip most of the plowzone and then, by dragging backward with its tripartite bucket, scrape the subsoil clean. The field crew located and mapped six original postholes and six replacement postholes and associated molds, representing the footprint of an earthfast building measuring 20 ft by 16.5 ft (a perch, a standard unit of measure in surveying from the Colonial period through the early 20th century). That the building was a dwelling seemed evident by a large root cellar at one end, filled with burned daub and charcoal. The crew

61

also identified at least eight borrow pits to the south of the dwelling (Figures 7-4 and 7- 5), most of which intersected one another in a manner typical of early Colonial sites (e.g., Gibb 1996).

Figure 7-4. Composite map of Garrett‘s Chance #2 (18PR703). N.B. Filled circles indicate positive shovel tests.

62

Figure 7-5. Feature map of Garrett‘s Chance #2 (18PR703). Only one posthole complex was excavated: Features 18 and 19. It was less than 2 ft deep (typical of postholes truncated by plowing and erosion). The postmold depicted in Figure 7-6 is not that of the replacement post but of the original, exposed by cutting the section profile back from the A—B line. Although little survived, it appeared to be rectangular in section, rather than the more common circular sections of undressed timbers typically employed in this kind of construction. Feature 18—the original posthole—produced only one undecorated ball clay tobacco pipe bowl fragment. Since that hole, and the other five original postholes, probably represents the initial historic period occupation of the site, the paucity of artifacts should not surprise the reader.

Figure 7-6. Plan and profile of Feature 18/19.

63

Feature 19 produced a surprising variety of materials, although in low numbers (Table 7-3).These materials offer several insights into the history of the occupation. They include one of only two porcelain sherds recovered from the site and one of only two tobacco pipestems with 4/64‖ diameter bores. The recovery of a small quantity of daub, window glass, an unmarked lead came, and five nails is consistent with the fact that Feature 19 is a replacement posthole. Clearly, at one point in the life of the building— probably in the early 18th century—it was cut free and lifted off of its original earthfast posts and new holes were dug and fresh posts inserted. Since windows likely were first removed to prevent breakage, the window glass and came may have resulted from window repair prior to post replacement. Table 7-3 Artifacts recovered from Feature 19, posthole. Weight Artifact type (ounces) Quantity British brown stoneware 1 Porcelain burned 1 Tin-glazed earthenware painted blue 1 Tobacco pipe undecorated 1 Tobacco pipe 4/64 1 Wine bottle dark green 1 Vessel glass aqua 1 Window glass lt olive green 1 Daub 0.05 Nail 5 Bone 2 Nut/Pit burned 1 Lead came 0.1 1 Within the building, at its south end, the field crew delineated and excavated, stratigraphically and in three sections, an apparent root cellar filled with a large quantity of burned daub and charcoal.3 Section 1 extends from Section Line A—B eastward and comprises an apparent stepped entry to the cellar (Figure 7-7). Section 2 lies between Section Lines A—B and G—H, while Section 3 is the western end of the pit. The strata from all three correspond to one another and the deposits and their artifacts are analyzed together. The cellar Measured approximately 26 ft2, 20.5 ft2 without the apparent entrance on the east side. The volume of the main part of the cellar measured approximately 34 ft3. Strata 1 and 2/2A are comprised of large fragment of well-preserved burned daub and charcoal, respectively (Figure 7-8). They yielded numerous sherds of melted colorless and greenish glass, some incorporating melted lead (Table 7-4). Over 300 handwrought nails from Stratum 1 and 76 handwrought nails from Stratum 2, as well as more than 50 indeterminate, but likely handwrought, nails from the combined strata were recovered. Also recovered from this deposit was the better part of a shattered British brown stoneware mug, a nearly completely mendable wine bottle, and about 100 sherds of tin- glazed earthenware, probably from a single decorate plate or platter. Staffordshire combed slip ware, Rhenish blue and gray stoneware, and some sherds of lead-glazed

3 All cardinal directions refer to the grid, which is approximately N60˚W; i.e., the house is oriented roughly east-west.

64

earthenware of uncertain origin and date were also recovered in small numbers. Bones were few in the upper strata, but more common in the underlying strata, and the field crew collected several pins and furniture tacks while excavating the deposit. Large samples of charcoal and daub have been retained as part of the site collection.

Figure 7-7. Plan of Feature 2 and profile A—B.

Figure 7-8. Profile G—H of Feature 2.

65

There can be little doubt that the dwelling burned and the hearth and chimney collapsed into the root cellar after the floor had burned. Fire-reddening of the north wall of the cellar indicates that the fire continued to burn after the collapse. The burned daub, charcoal, numerous nails, melted window glass and cames, three burned and/or broken—but largely restorable—vessels, and very small quantity of food refuse, support this hypothesis (Figures 7- 9 and 7-10). Given that the replacement postholes had very few small daub inclusions, no large ones, and little charcoal, it seems certain that the building burned sometime after it had been repaired. The surviving fragments of three vessels in the rubble point to accidental burning. But there are small fragments of other vessels as well (Figure 7-11; see Table 7-4) Table 7-4 Artifacts recovered from Feature 2, by stratum. Sub-Class Variety 1 1A 2 2A 3 4 4a 6 Grand Total Porcelain Chinese 1 1 Stoneware British brown 36 1 1 40 Stoneware indeterminate 1 3 4 Stoneware Rhenish gray 3 2 2 7 Stoneware White salt glazed 2 1 3 Earthenware Plain redware 1 1 Earthenware Slipped redware 1 1 Earthenware Tin-glazed 88 2 11 12 1 118 Vessel glass Table glass 2 2 Vessel glass Vial 13 1 2 1 17 Vessel glass Wine bottle 37 1 3 3 45 Tobacco pipe Ball clay 20 1 8 37 4 1 69 Tobacco pipe (5/64ths‖) (6) (6) (13) (1) (1) (27) Tobacco pipe (6/64ths‖) (1) (1) Tobacco pipe Terra cotta 1 2 3 Buckle Brass/Tinned 2 2 Thimble Brass/Tinned 1 1 Pin Brass/Tinned 1 1 2 1 2 7 Tack Furniture 3 2 5 Came Lead 8 1 1 10 Nail Handwrought 287 27 21 55 28 11 1 1 431 Nail indeterminate 27 13 14 1 56 Daub 1 1 Window glass Window glass 6 1 5 1 13 Vessel glass indeterminate 103 1 106 Indet. Metal indeterminate 1 2 3 Indet. Metal Wire 1 1 2 Gunflint English flint 1 1 Bone Various 13 3 1 17 Egg shell Bird 1 1 Oyster Oyster valve 6 1 2 9 Vegetal Nut/Pit 10 5 1 16 Fire-cracked rock 1 1 Flake Decortication 2 1 3 Flake Secondary 1 1 Flake Shatter 1 1 1 3 Flake Tertiary 1 1

66

Figure 7-9. Burned daub from Feature 2.

Figure 7-10. Wine bottle and British stoneware mug from Feature 2.

67

Figure 7-11. Rhenish stoneware from Feature 2 and other contexts. N.B. Drawing at right reconstructs a similar fragmentary Rhenish stoneware bulbous mug from the Patuxent Point site (18CV271). Source: Gibb 1996. Three terra cotta tobacco pipe fragments were found in the upper strata, but they may be fortuitous inclusions, possibly even components of the daub. The ball clay tobacco pipestems, with 28 measurable bores from all of the strata combined, indicate an 18th century date, all but one (6/64‖) measuring 5/64ths of an inch. Section drawings of the root cellar document two interesting features: the probable stepped entry noted above, and a deeper, possibly smaller and original pit along the south side (Figures 7-12 and 7-13).

68

Figure 7-12. Plan and profiles of Feature 2.

69

Figure 7-13. Feature 2 excavated, looking east. N.B. stepped entry (top) and deeper pit (right). Forming an arc to the south of the dwelling were a series of eight borrow pits, six of which intruded upon one another in two sets (Figure 7-14). Intersecting borrow pits typically occur on early Colonial sites with earthfast construction, likely pointing to the original construction and subsequent repair of wattle and daub chimneys. Feature 1 proved to be the earliest dateable feature on the site (only Feature 21 yielded enough measurable pipestems [n=30+] for a reliable date and no other artifacts that could be dated absolutely were recovered from the site). Binford‘s simple regression formula and Hanson segmental regression for 1680-1750 produced dates of 1703 and 1698, respectively. Given these early dates, Feature 1 may have supplied the daub for the initial construction and subsequent maintenance of the dwelling. The pit remained open long enough to accumulate lenses of domestic debris as well as thick deposits of daub that ranged from heavily sintered to nearly unaltered by fire. Table 7-5 Pipestem bores from Feature 1. Diameter (64ths) Feature1 Percentage Sums 7 5 15.1515152 35 6 22 66.6666667 132 5 6 18.1818182 30 33 1 197 5.96969697 Binford 1703.449394 Hanson 1697.999394

70

The roughly circular pit measures about 8 ft in diameter and 15 inches deep (Figure 7-15). It is basin-shaped and was intruded by two smaller pits, Features 3 and 4, both of which yielded very small quantities of domestic and architectural material (Tables 7-6 and 7-7).

Figure 7-14. Borrow pits at Garrett‘s Chance #2 (18PR703). Table 7-6 Artifacts recovered from Feature 3. Sub-Class Variety Stratum 1 Vessel glass Wine bottle 1 Tobacco pipe Ball clay 3 Nail Handwrought 2 Daub 2 Bone Fish 3 Mammal 2 Flake indeterminate 1 Table 7-7 Artifacts recovered from Feature 4 Sub-Class Variety Stratum 1 Stratum 2 Total Earthenware Tin-glazed 1 1 Stoneware indeterminate 1 1 Tobacco pipe Ball clay 8 1 9 (7/64ths‖) (2) (2) (6/64ths‖) (1) (1) Nail Handwrought 2 1 3 Nail indeterminate 1 1 Daub 2 ou. Bone Fish .05 ou. 1 1 Oyster Oyster valve .05 ou. .05 ou.

71

Figure 7-15. Plans and profiles of Features 1, 3, and 4. The field crew recovered very few ceramic or glass vessel sherds from Feature 1, but other forms of trash included a large quantity of bone representing various fish (bone and scales) and birds (bone and egg shell), cattle, sheep, swine, and deer. Only 62 nails and nail fragments and under one pound of daub pale in comparison to the recovery of those artifact classes from Feature 1. Several interesting utensils were recovered, including: a latten spoon with trifid terminal, a 15‖ octagonal steel tube that may be a gun barrel, a possible pocket knife, a smoker‘s companion, and what appears to be a small mill pick, possibly used to sharpen millstones (Figures 7-16 through 7-18). All await conservation. A triangular projectile point, stone biface, and a handful of stone flakes probably indicate a diffuse prehistoric occupation of the site, the surface materials incidentally incorporated into the feature fill. No aboriginal pottery was recovered from Feature 1, or from any other part of the site.

72

Table 7-8 Artifacts recovered from Feature 1. Sub-Class Variety Stratum 1 Stratum2 Surface Total Stoneware Rhenish gray 2 2 Earthenware Tin-glazed 1 3 4 Vessel glass Wine bottle 4 1 1 6 Tobacco pipe Ball clay 34 15 16 65 (7/64ths‖) (4) (1) (5) (6/64ths‖) (12) (4) (6) (22) (5/64ths‖) (1) (5) (6) Smoker's companion 1 1 Spoon Latten 1 1 Pocket knife 1 1 Mill pick 1 1 Nail Handwrought 28 3 5 36 Nail indeterminate 13 5 2 20 Tack Tack 1 1 Indet. Metal Wire 1 1 Indet. Metal 4 2 6 Key Handwrought 1 1 Buckle 1 1 Egg shell Bird Bone Fish Oyster Oyster valve 1 1 Projectile point Madison 1 1 Biface 1 1 Flake Decortication 3 3 indeterminate 1 1 Secondary 1 1 Shatter 1 1 Tertiary 2 1 3 Feature 21 forms the core of a second cluster of borrow pits, products of excavation, deposition, and overlapping reëxcavation (Figure 7-19). Features 22 and 24 are the intrusive pits, larger and less peripheral to the core pit than Features 3 and 4 were to Feature 1. Pipestem analysis puts the mean date for the formation of the deposits to between 1713 and 1720 (Table 7-9). Hanson‘s formula, which deals with shorter time periods, probably returned the more accurate of the two dates. It suggests a significant episode of chimney rebuilding at that time. Table 7-9 Pipestem bores from Feature 1. F21 Total Percentage Sums 2 5.882353 14 14 41.176471 84 18 52.941176 90 34 1 188 5.529411765 Binford 1720.29471 Hanson 1712.52

73

Figure 7-16. Metal artifacts recovered from Features 1 and 21.

Figure 7-17. Knife or razor from Features 1 and 21.

74

Figure 7-18. Lateen spoon fragments from Features 1 and 21.

Figure 7-19. Profiles of Feature 21, southwest quadrant. Although Feature 21 lacked the large deposit of variously sintered daub in its bottom, contra Feature 1, it produced a similar array of artifacts (Table 7-10). It was rich in faunal material—including fish, bird, cattle, swine, sheep, and deer—had few ceramic and glass vessel sherds (other than the remains of one or more tin-glazed earthenware vessels), and produced a few unusual utensils: a single bit felling axe (see Figure 7-16), a key, a dozen pins and a thimble, and an incised piece of bone that likely was part of a decorative hair comb (Figure 7-20). Like Feature 1 and its satellites, Feature 21 was dug for daub and then was used for discard of architectural and domestic refuse, particularly kitchen refuse.

75

Figure 7-20. Pin, thimble, incised bone comb, tacks and lead ball. N.B. Lincoln one-cent piece for scale. It was not recovered from the site. Intrusive Feature 22 (Figure 7-21; Table 7-11), small as it was, yielded a number of ceramic and glass sherds, the handle of another latten spoon, and only nine measurable pipestems. (A tenth produced different measurements from opposite ends, one of five for the site.) Ethnobotanist Justine W. McKnight extracted a quantity of organic material from 10 liters of soil from the south half of Feature 22. Among the carbonized remains of local trees (oak, poplar, beech, hickory, walnut, chestnut), grasses and sedges, she recovered 25 fragments of maize. Maize was a staple of the Colonial diet in the Chesapeake region. Feature 24 (Table 7-12) artifacts were few and unexceptional, replicating what had been found elsewhere on site, including another smoker‘s companion and a badly deteriorated table knife, as well as a Rhenish stoneware, a dozen tin-glazed earthenware sherds, a lead-glazed red earthenware, a flake, and a fire-cracked rock. Feature 23 (Figure 7-22; Table 7-13) lies apart from the two borrow pit clusters, as does Feature 5 (unexcavated). Roughly ovoid in plan, it measured approximately 4½ ft by 3½ ft, was steep-sided, and yielded only two buckles, traces of daub, 17 handwrought nails, a wine bottle sherd, and a remarkable quantity of large, fire-cracked pebbles. I suspect, but cannot demonstrate, that this silt filled pit was used to mix soil with water for use as daub.

76

Figure 7-21. Profile of Feature 22. Analysis and Interpretation The various lines of evidence—archival, spatial, stratigraphic, and artifactual— admit a fairly straightforward interpretation. The dwelling, on land occupied by aborigines sometime in the Late Woodland, if not earlier, was built by Bernard Johnson for himself or by and for tenants, in the last decade of the 17th century. The pipestem dates, scarce white salt-glazed stoneware sherds, and relatively abundant Rhenish gray and British brown stoneware sherds support that initial date. The lack of case bottle sherds and the recovery of only one lead-backed tin-glaze (i.e., majolica) from Feature 2, Stratum 4A, argues against an earlier date, as does the rare occurrence of tobacco pipestems with bore diameters of 7/64ths of an inch (n=4, two each from Features 4 and 21). Sometime in the early part of the second decade of the 18th century, possibly with the conveyance of this part of Dove‘s Nest to William Wilkerson, a tenant household rehabilitated the dwelling, replacing the original posts and reconstructing the wattle and daub chimney. While some of their trash may have ended up in the original borrow pit (Feature 1), the tenants probably deposited most of their kitchen refuse, and some architectural debris, into their principal borrow pit (Feature 21).

77

Table 7-10 Artifacts recovered from Feature 21. Sub-Class Variety Str. 1 Str. 2 Str. 3 Str. 5 Str. 6 Surface Total Stoneware Rhenish gray 7 1 8 Earthenware Plain redware 1 1 Earthenware Tin-glazed 40 5 45 Vessel glass Wine bottle 1 2 3 Tobacco pipe Ball clay 56 9 2 6 73 (7/64ths‖) (2) (2) (6/64ths‖) (12) (2) (14) (5/64ths‖) (16) (1) (1) (18) (5/64 & 6/64ths‖) (2) (2) (6/64 & 7/64ths‖) (1) (1) Pocket knife 1 1 Felling axe Single bit 1 1 Comb Incised 1 1 Pin Brass/Tinned 10 2 12 Key Handwrought 1 1 Nail Handwrought 43 13 2 1 59 Nail indeterminate 3 3 Daub Indet. Metal indeterminate 5 5 2 1 1 14 Window glass Window glass 1 1 Bone Fish 1 1 Bone Mammal 1 2 3 Bone various 2 3 7 12 Egg shell Bird 1 1 Oyster Oyster valve 1 1 Fire-cracked rock 1 1 Flake Decortication 2 2 Flake Shatter 1 1 Flake Tertiary 3 3 Flake English flint 3 1 4

78

Table 7-11 Artifacts recovered from Feature 22. Sub-Class Variety Str. 1 Surface Total Stoneware Rhenish gray 1 1 Earthenware Tin-glazed 9 9 Earthenware Plain redware 2 2 Earthenware indeterminate 1 1 Vessel glass Wine bottle 9 9 Vessel glass Table glass 1 1 Vessel glass indeterminate 1 1 Spoon Latten 2 2 Tobacco pipe Ball clay 19 1 20 Tobacco pipe (6/64ths‖) (3) (3) Tobacco pipe (5/64ths‖) (5) (1) (6) Tobacco pipe (5/64 & 6/64ths‖) (1) (1) Pin Brass/Tinned 1 1 Brick Common red 1 1 Daub Nail Handwrought 14 2 16 Nail indeterminate 36 36 Window glass Window glass 1 1 Indet. Metal indeterminate 3 3 Bone Fish Bone Mammal 2 2 Bone indeterminate 2 2 Oyster Oyster valve 2 4 6 Biface 1 1 Fire-cracked rock 2 2 Flake 1 1 Table 7-12 Artifacts recovered from Feature 24. Sub-Class Variety Quantity Stoneware Rhenish gray 1 Earthenware Tin-glazed 12 Earthenware Plain redware 1 Vessel glass Wine bottle 1 Table knife 2 Tobacco pipe Ball clay 8 Smoker's companion 1 Pin Brass/Tinned 1 Indet. Metal indeterminate 4 Indeterminate indeterminate 1 Nail Handwrought 5 Nail indeterminate 15 Daub 5.5 ou. Bone various 12.6 ou. Oyster Oyster valve Fire-cracked rock 1 Flake 1

79

Figure 7-22. Profiles of Feature 23.

Table 7-13 Artifacts recovered from Feature 23. Sub-Class Variety Str. 3 Total Vessel glass Wine bottle 1 1 Buckle Brass/Tinned 2 2 Daub .1 ou. .1 ou. Nail Handwrought 17 17 Fire-cracked rock 47.8 ou. 47.8 ou. Precisely when the site was abandoned remains uncertain. The dwelling clearly burned in its entirety, the leaded windows destroyed along with some furnishings (furniture tacks) and utensils (at least one British Brown stoneware mug, one tin-glazed earthenware plate or platter, and a wine bottle). The chimney collapsed through the burned wooden floor and into the re-dug root cellar. (The lack of a separate chimney bay may have made this building more susceptible than most to catastrophic fire.) The burning of the structure was hinted at when the field crew first examined the surface, the soils considerably darker than anywhere else in the project area. There is no evidence of subsequent occupation, to this day, of Garrett‘s Chance #2 (18PR703). Table 7-14 suggests a mean date of 1724, and the several tobacco pipe bowls recovered from various

80

contexts support that date (Figure 7-23). The recovery of only seven sherds of white salt- glazed stoneware—three from Feature 2 (the latest of the features to be filled) and four from the surface or shovel tests—suggests abandonment during the second quarter of the 18th century. One can speculate about why this portion of Dove‘s Nest was abandoned and left to agriculture. The dwelling, of course, could have been rebuilt; but clearly there were reasons for not having done so, at least at the location of Garrett‘s Chance. The reason may have been systemic: an increasingly eroded, exhausted soil that had to be left fallow, a new farmstead built closer to more productive lands. Equally likely, the owners may have opted to till the land themselves or with enslaved labor, and not have renewed leases on the land. Housing for tenants may have proved inconveniently sited for enslaved labor. Answers may be waiting in the yet untapped portions of the archival record. Table 7-14 Pipestem bore diameters for combined principal sources.

Diameter (64ths) F1 Total F21 Total Surface Site Total 7 5 2 1 11 6 22 14 14 67 5 6 18 40 103 4 0 0 1 3 Totals 33 34 56 184 Binford 1703 1720 1733 1723 Hanson 1698 1713 1724 1714

Figure 7-23. Tobacco pipe bowls from Garrett‘s Chance #2 (18PR703).

81

Chapter 8. Garrett’s Chance #3 (18PR704) Introduction Garrett‘s Chance #3 (18PR704) occupies the top of an eroded knoll, just northeast of an abandoned tobacco barn (Figure s 8-1 and 8-2). An initial intensive surface collection yielded two dozen prehistoric artifacts, mostly fire-cracked rock, but including one black chert Clovis point and a quartzite scraper from about one-quarter of an acre. After a fierce rain three days later (21 May 2004), I re-collected the field, recovering 14 objects including four decortication flakes, a secondary flake, and nine more fire-cracked rocks. On June 18, after yet another heavy rain, I collected 38 objects, including ten decortication flakes, ten secondary and tertiary flakes, and 18 fire-cracked rocks. The succession of surface collections amidst the growing corn crop made piece plotting infeasible.

Figure 8-1. Garrett‘s Chance #3 (18PR704) viewed from the tobacco barn. Based on the surface collection results, Andrew Garte & Associates established a 20 ft shovel testing grid, the corners of which were mapped by Landesign‘s survey crew and located on the electronic map for the entire project area. This grid provided the sampling frame for our Phase 2 effort to identify the horizontal and vertical extents and integrity of the site, and to collect additional materials to determine whether or not the site contained more than a Paleoindian component. Shovel testing (n=45) yielded consistent results: small numbers of flakes and fire-cracked rock from a thin plowzone that virtually disappeared around the margins of the site. Twelve test units measuring 2½

82

ft by 2½ ft, regularly placed at 20 ft intervals in the apparent center of the site, produced few artifacts (30 fire-cracked rocks, 42 flakes, and one Bare Island biface, or an average of eight objects per unit).

Figure 8-2. Map of Garrett‘s Chance #3 (18PR704). The following sections examine the results of each phase of study, in turn, offering interpretations of those findings and of the study as a whole. Results

SURFACE The plowed surface of the highest and most level portion of Field I was on the north side of Barn 4. While the rest of the field was comprised of gravel and eroded sandy loam, the summit was virtually devoid of gravel, the soil fine sandy loam. The field crew, walking parallel transects less than 15 ft apart collected material that, while dispersed, appeared to lie immediately north and northeast of the barn. The sloping edges of the field revealed plowzone deposits that changed to the same kind of gravelly sandy loam observable throughout the rest of Field I and in Field J below. Erosion clearly would play a significant role in site definition. As defined, the site comprised less than 10,000 ft2 at an elevation of 174 ft above mean sea level. It lies at the heads of drainages leading westward to the main branch of Swanson Creek and southward to the branch of the creek that defines the southern boundary of the Stone and Compton farms. Both drainage heads are 400 to 500 ft distant and there is no permanent body of water in the immediate vicinity of the site. Surface streams that might have existed during the late Pleistocene and early Holocene are no longer evident, or survive only as the eroded drainages that now constitute a portion of the Swanson Creek watershed. The Patuxent River is more than three miles to the east and, in any case, did not exist as anything more than a stream during the Paleoindian period.

83

The results of the three surface collections appear in Table 8-1. Of the 49 fire- cracked rocks recovered, 27 (55%) are quartzite. Quartzite is not so prominent in the flaked stone assemblage, which includes decortication, secondary, and tertiary (but no primary) flakes of quartz (21 of 25, or 84%). A single Clovis point and one quartzite scraper were recovered from the surface (Figure 8-3). The only other artifacts noted on the surface and collected were three sherds of late 19th/early 20th century vessel glass. Table 8-1 Artifacts surface collected from Garrett‘s Chance #3 (18PR704). Class Sub-Class Variety Type Form Quantity Lithic Fire-cracked rock cherty limestone cherty limestone 3 quartz quartz 19 quartzite quartzite 27 Fire-cracked rock Total 49 Flake Decortication quartz quartz 12 quartzite quartzite 2 Decortication Total 14 Secondary quartz quartz 3 Secondary Total 3 Tertiary chert chert 1 quartz quartz 6 quartzite quartzite 1 Tertiary Total 8 Flake Total 25 Projectile point Clovis Clovis chert 1 Uniface Scraper quartzite quartzite 1 Lithic Total 76 Vessel Glass Vessel glass indeterminate indeterminate hollowware 1 machine-made hollowware 2

84

Figure 8-3. Diagnostic stone tools from Garrett‘s Chance #3 (18PR704). N.B. Clovis point on left and quartzite scraper on right from surface. Bare Island biface in middle from excavation unit.

SHOVEL TESTING The field crew undertook shovel testing at Garrett‘s Chance #3 (18PR704) to determine the horizontal and vertical limits of the site (Table 8-2; Figures 8-4 through 8- 6). The western, eastern, and southeastern boundaries reflect the effects of severe erosion, while the remaining boundary demarcates the distribution of artifacts as revealed by shovel testing. Shovel testing (n=45) revealed a roughly oblong distribution, with the east-west axis of the ellipse measuring 125 and that of the north-south axis 100 ft. None of the shovel tests produced more than two prehistoric artifacts (historic materials were virtually absent), hence I have plotted presence-absence data rather than attempt computer simulations of distributions. The resulting assemblage is equally divided between fire-cracked rock and flakes, quartzite again dominating the former, quartz the latter. Flakes also represent the full range of lithic reduction, although primary flakes—the middle of the range—are absent, mirroring the surface collection findings.

85

Table 8-2 Shovel testing artifacts from Garrett‘s Chance #3 (18PR704). Class Sub-Class Variety Type Form Quantity Lithic Fire-cracked rock chert chert 3 limonitic sandstone limonitic sandstone 1 quartz quartz 3 quartzite quartzite 9 Flake Decortication quartz quartz 2 quartzite quartzite 1 Secondary chert chert 1 quartz quartz 3 Shatter quartz quartz 2 Tertiary quartz quartz 7 Flake Total 16 Lithic Total 32 Nail Wire common common 1 Nail Total 1

Figure 8-4. Testing grid at Garrett‘s Chance #3 (18PR704). Circles represent shovel tests; those that are filled are positive (1-2 artifacts). Squares are excavation units, all yielding some artifacts. Ellipse is approximate site boundary.

86

10

8

6

4 Frequency 2

0 0.3 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 0.75 0.8 Plowzone Depth (ft)

Figure 8-5. Plowzone depth intervals determined from shovel tests.

Figure 8-6. Plowzone depths under 0.5 ft. N.B. Filled circles represent plowzone depths under 0.5 ft.

87

EXCAVATION UNITS While the quantity of material recovered proved small and no features were encountered, the second phase of study was expanded to include 12 small test units to more accurately identify artifact densities and to expand the search for subsurface features (see Figure 8-4; Table 8-3). Three nails (one handwrought), 30 fire-cracked rocks, 42 flakes, and one Bare Island biface (see Figure 8-3) were recovered. Again, quartzite dominates the fire-cracked rock sub-assemblage, quartz the flaked stone. The Bare Island piece—a stemmed biface with asymmetrical triangular blade—is made of limonite- cemented sandstone. All stages of lithic reduction are represented, with even four primary flakes noted. Table 8-3 Artifacts from excavation units, Garrett‘s Chance #3 (18PR704). Class Sub-Class Variety Type Form Quantity Architecture Nail Handwrought common common 1 indeterminate indeterminate indeterminate 2 Nail Total 3 Lithic Fire-cracked rock (blank) quartz quartz 10 quartzite quartzite 20 Fire-cracked rock Total 30 Flake Decortication chert chert 1 quartz quartz 5 quartzite quartzite 3 Decortication Total 9 Primary quartz quartz 3 quartzite quartzite 1 Primary Total 4 Secondary quartz quartz 5 Secondary Total 5 Shatter indeterminate indeterminate 1 quartz quartz 8 quartzite quartzite 1 Shatter Total 10 Tertiary chert chert 1 limonite limonite 1 quartz quartz 11 rhyolite rhyolite 1 Tertiary Total 14 Flake Total 42 Projectile point Bare Island limonitic sandstone limonitic sandstone 1 Lithic Total 73 Summary and Interpretation Three surface collections, 45 shovel tests, and twelve 2½ ft by 2½ ft excavation units failed to identify any features, either sub-plowzone or plowzone concentrations. On the contrary, all identified evidence of a severely eroded landform, surrounded and underlain by marine deposits of sands and gravels.

88

Table 8-4 Combined assemblages for Garrett‘s Chance #3 (18PR704). Class Sub-Class Variety Type Form Totals Vessel Glass Vessel glass indeterminate indeterminate hollowware 1 machine-made hollowware 2 Architecture Nail Handwrought common common 1 Nail Wire common common 1 Nail indeterminate indeterminate indeterminate 1 Lithic Fire-cracked rock (blank) chert chert 3 cherty limestone cherty limestone 3 limonitic sandstone limonitic sandstone 1 quartz quartz 32 quartzite quartzite 56 Fire-cracked rock Total 95 Flake Decortication chert chert 1 quartz quartz 19 quartzite quartzite 6 Decortication Total 26 Primary quartz quartz 3 quartzite quartzite 1 Primary Total 4 Secondary chert chert 1 quartz quartz 11 Secondary Total 12 Shatter indeterminate indeterminate 1 quartz quartz 10 quartzite quartzite 1 Shatter Total 12 Tertiary chert chert 2 limonite limonite 1 quartz quartz 24 quartzite quartzite 1 rhyolite rhyolite 1 Tertiary Total 29 Flake Total 83 Projectile point Bare Island limonitic sandstone limonitic sandstone 1 Clovis Clovis chert 1 Projectile point Total 2 Uniface Scraper quartzite quartzite 1 Lithic Total 182 The combined assemblages (Table 8-4) include just a few historic artifacts likely related to the construction and use of the tobacco barn. They are few and dispersed and have no insights to offer into the occupation of the Compton farm. The value of this site is its prehistoric assemblage, comprised of 95 fire-cracked rocks (34% quartz, 59% quartzite with traces of cherty limestone and sandstone) and 83 flakes. The flakes represent all facets of lithic reduction (Table 8-5) and quartz, with few exceptions, was the principal material. The three formal tools recovered (and not included in the totals below), however, were of black chert, quartzite, and a limonite-cemented sandstone. Both bifaces were broken and may have been replaced on-site with locally available materials.

89

It would be unwise, however, to classify Garrett‘s Chance #3 (18PR704) as a lithic resources procurement and workshop site because fire-cracked rocks are present, and even outnumber the flaked stone artifacts. Table 8-5 Lithic materials summary for all proveniences.

cherty limonitic Flake Type quartz quartzite chert limestone rhyolite limonite sandstone indeterminate Total Decortication 19 6 1 26 Primary 3 1 4 Secondary 11 1 12 Shatter 10 1 1 12 Tertiary 24 1 2 1 1 29 Total 67 9 4 0 1 1 0 1 83 Fire-cracked rock 32 56 3 3 1 95 Fire-cracked rocks usually indicate pavements on which plants or animals—but probably roots and nuts—were roasted, or ‗pot-boiling‘—the placement of heated stones in vessels unsuited for direct placement on a fire. The people who occupied Garrett‘s Chance #3 (18PR704) appear to have selected quartzite for these ‗firestones,‘ accepting also quartz. These patterns in materials choice, and the suggested activities of stone tool replacement and food-processing or preparation, raise a critical question: who lived here? The Clovis point clearly marks a Paleoindian presence. The site lies atop a knoll; therefore these materials are not redeposited. A Bare Island biface is the only other formal tool recovered (the scraper lacks a distinguishable style). The reader will note that I have referred to it throughout this document as a biface and not as a point. General wisdom places this style in the Late Archaic and Early Woodland (and by extension, the Transitional period between the two), but its characteristic asymmetry has led some to regard it as a hafted biface—a knife or scraper, if you will—rather than as a projectile point (Hranicky 1994: 28; Christopher Polglase, pers. comm.., 2004). I suggest that it is a generalized hafted tool made of tenacious lithic materials from which a thin sharp edge is not required, and it is not characteristic of any one period. In short, I think this biface is no more suitable as a temporal diagnostic than the scraper noted above. If the reader accepts this reasoning, then we can conclude together—or at least hypothesize—that Garrett‘s Chance #3 (18PR704) is a single component Paleoindian site.

90

Chapter 9. Garrett’s Chance #4 (18PR709) Introduction Garrett‘s Chance #4 (18PR709) manifested itself as a dense concentration of ceramic and glass vessel fragments on a moderately steep plowed slope, adjacent to a steep woodlot. The soils were severely eroded and gravelly. A single shovel test pit at the approximate center of the concentration revealed a thin layer of brown (10YR4/3) gravelly sandy loam on top of dark yellowish brown (10YR4/4) clay loam. Six earthenwares, including two slipped buff-bodied sherds, two Pearlware/Whiteware, and two Whiteware, were recovered, along with one wine bottle and two beverage bottle sherds. The contrasting textures suggest redeposition of sands and gravels on eroded clay subsoil or the complete loss of the original topsoil and much of the gravelly sandy B- horizon resting on a marine clay deposit. In any case, the site clearly lost much of its integrity to erosion. The near lack of architectural artifacts (window glass, nails, brick) also creates uncertainty as to whether Garrett‘s Chance #4 (18PR709) is a house site or some sort of trash dump. Results The field crew collected all ceramic and glass refuse from the surface of the plowed field, defining a roughly elliptical area measuring 250 ft by 100 ft (Figure 9-1). Brick was not collected, but only a half dozen or so pieces were noted, all apparently handmade or early machine made.

Figure 9-1. Location and extent of Garrett‘s Chance #4 (18PR709). Only two sherds of window glass were recovered from the surface. No nails were found, but a few bits of oxidized iron were noted and may represent nails. Nearly all of

91

the collected materials include ceramic (n=143) and glass (n=44) vessel sherds (Tables 9- 1 and 9-2). The two Chinese porcelain and three lead-glazed red and buff-bodied earthenwares suggest a second quarter of the 19th century, or earlier, date for the deposit. The single sherd of Westerwald points to a much earlier date, perhaps the first through third quarters of the 18th century; but the majority of ceramics date from the middle of the 19th through early 20th centuries. The numerous pieces of hand-finished and fully machine-made glass bottle sherds and the white glass canning jar lid suggest a c.1900 date. Table 9-1 Ceramics recovered from Garrett‘s Chance #4 (18PR709). Class Sub-Class Variety Type Form Quantity Ceramic Earthenware Plain buff-bodied lead-glazed hollowware 1 Plain redware lead-glazed hollowware 2 White granite Total 15 Whiteware dipped Total 8 Whiteware indeterminate Total 56 Whiteware painted Total 16 Whiteware shell-edged plate 5 Whiteware transfer-printed Total 3 Whiteware Total 89 Yellowware Rockingham hollowware 1 Porcelain Chinese Total 2 Porcelain indeterminate Total 20 Stoneware Alkaline glazed indeterminate hollowware 1 Stoneware American gray Total 29 Stoneware Westerwald indeterminate hollowware 1 Ceramic Total 143

Table 9-2 Vessel glass recovered from Garrett‘s Chance #4 (18PR709). Class Sub-Class Variety Type Form Quantity Vessel Glass Vessel glass Beverage bottle hand-finished beverage bottle 1 machine-made beverage bottle 7 Case bottle hand-finished baking soda bottle 1 vial 2 machine-made vial 1 indeterminate indeterminate hollowware 10 machine-made hollowware 5 Jar glass seal lid 3 machine-made jar 4 Wine bottle handmade wine bottle 10 Vessel Glass Total 44

Summary and Interpretation Surface collecting of Garrett‘s Chance resulted in the recovery of nearly 200 artifacts—almost exclusively domestic in character (ass opposed to architectural)—from a severely eroded surface. The material may derive from the extant Compton House on

92

the hill above. The mixture of materials from c. 1700 to the early 20th century, combined with the lack of stratigraphic integrity and modest evidence of architecture limits the research potential of this site. While the assemblage may be statistically representative of the deposits from which it was drawn, the transformational processes—other than severe erosion—that led to its creation and modification remain uncertain and probably defy characterization because of the erosion.

93

Chapter 10.Standing Structures Introduction There were eight extant buildings and ruins on the Compton farm in 2004, but none on the portion of the Stone farm slated for development. Of the eight, three comprise the houselot: the dwelling, a workshop, and a poultry house. The remaining structures are all agricultural and include a corn crib and stable (Barn 3), two extant tobacco barns, and one tobacco barn ruin (Barn 1). Each is briefly described below.

Figure 10-1. Compton farm building locations. Compton House and Houselot The Compton house is a modest-sized farmhouse in good to excellent condition (Figure 10-2). It is a two-story, four-bay, L-plan, framed structure with three end-gables, a full front porch and a small side entry porch that leads into the kitchen (Figures 10-3 and 10-4). The foundation is mortared brick that has been parged recently. There is a one story framed addition with plain horizontal board siding on a poured concrete foundation on the northwest corner of the main block. The rest of the house has German siding. The kitchen, in the southwest corner of the house, has an interior stove chimney of brick, while the paired fireplaces in the two parlors share a brick stack with 2-course simple finish that emerges through the center of the ridgeline of the main block. Asbestos shingles were recently installed across the main, addition, and porch roofs. Both porches have been substantially rebuilt in recent years with particle board and plywood. The full-length front porch as a full-length half-hipped roof supported by unadorned 6‖ by 6‖ posts set on top of rusticated concrete block piers. The blocks are identical to those used to frame the entrance to the houselot (Figure 10-5).

94

Figure 10-2. Compton farm houselot. Most of the windows on the main block appear to be original 2/2 double-hung sashes with plain board moldings. The middle window on the second story of the main façade is a 1/1 double sash replacement. There is another replacement sash on the south façade.

Figure 10-3. Principal (east) façade of the Compton House.

95

Figure 10-4. South façade of the Compton House.

Figure 10-5. Entry gate to Compton farm houselot.

96

The interior of the dwelling and its additions has been modernized. The two fireplaces, located on opposite sides of a central wall in the two front parlors, have modest Classical Revival detailing. They appear to be identical, although one has been modified to accommodate the exhaust from a space heater.

Figure 10-6. Classical Revival mantle, south parlor.

Figure 10-7. Classical Revival mantle, north parlor.

97

Curiously, all of the windows and doors sport identical bulls-eye moldings, including those in the mid-20th century one-story addition (Figures 10-8 and 10-9).

Figure 10-8. Standard window and door treatment throughout interior.

Figure 10-9. Typical door, second floor.

98

Several doors have raised panels and working lock sets (see Figure 10-9; Figure 10-10).

Figure 10-10. Typical lock set. There are two outbuildings associated with the house: a workshop and a poultry house (Figure 10-11). The workshop (Figure 10-12) measures 8 ft by 10 ft. A poured concrete slab supports the framing structure of 6‖ by 6‖ sills, 2‖ by 4‖ studs, rafters, and plates, all milled. There are board purloins and no ridgepole. The board and batten siding is attached to horizontal board sheathing. The roof is corrugated metal. There are interior shelves. The poultry house (Figure 10-13) also has a poured concrete floor. The verticals, trusses, and ridgepole are all of milled dimensional lumber. The structure is sheathed in German siding and roofed with planks covered by corrugated metal roofing. A chain link and 2‖ by 4‖ fence extends off of the poultry house‘s north façade.

99

Figure 10-11. Outbuildings associated with the house.

Figure 10-12. Workshop.

100

Figure 10-13. Poultry house, south façade. Barn 1 Barn 1 is a ruin in the midst of cultivated fields (Study Fields C, D, and F). It lies flat on the ground, covered with vegetation. Of uncertain size, the barn appears to have had a gambrel roof with corrugated metal covering and partially dressed timber sills that rested on poured concrete piers. I suspect that there was an attached stripping shed, also built on poured concrete piers. Barn 2 Barn 2 is a gambrel roofed structure with a metal raised seam covering, a continuous poured concrete foundation, vertical board siding, and milled lumber and wire nails throughout. The first floor of this 15 ft by 28 ft building is divided into stalls, and the floor above is a hay loft. The stable is wracked and in poor to fair condition. A 10 ft by 12 ft corn crib lies a few feet to the northeast of the stable (Figure 10- 15). Its 6‖ by 6‖ milled sills rest atop poured concrete piers. The milled lath-like vertical slats support ¼-inch hardware mesh. Joists support a plank floor 6‖ above the sills. The crib is roofed with corrugated metal. Barn 3 Barn 3 is situated in the woods adjacent to a paintball range (Study Area 1). Its footprint is approximately 28 ft by 32½ ft. The gabled roof has a corrugated metal cover and lacks a ridgepole. There are 18 interior concrete piers along a center aisle, mortised framing, and milled vertical siding and trusses. A sheet metal collar along the exterior walls sets the lower 3 ft of board siding from the lengths above, probably making it easier and less costly to remove rotted and termite riddled boards. The barn is in good condition.

101

Figure 10-14. Stable (Barn 2) and corn crib (left). Barn 4 Barn 4 straddles the top of a knoll near the center of the project area‘s north boundary. It defines an edge of Garrett‘s Chance #3 (18PR704), the Paleoindian site discussed above. The barn is largely inaccessible because of enveloping vegetation. It also appears to be unsound. Milled boards attached with wire nails and machine-made hinges side the structure. Barn 4 is in poor to fair condition. Summary and Interpretations A brief survey of above-ground architecture within the project area (the reader will recall that the 20+ acres of the Sunnyside Farm are not part of the proposed subdivision) identified eight buildings, all of which are part of the Compton farm. They include the houselot with its 20th century vernacular dwelling house and two outbuildings, three tobacco barns, a corn crib, and stable. Together these buildings comprise the most important part of the farm‘s infrastructure, housing the owners, their draft animals and poultry, and feed for the animals. The three barns and probably stripping shed clearly demonstrate the Comptons‘ focus on tobacco culture throughout the 20th century. With the demise of tobacco as a principal crop in Maryland, and an emphasis on maize, soybeans, and wheat, the barns have become obsolete. Hay has not been a major crop on this farm in recent years, and, in any case, the technologies for harvesting and storing hay no longer require barns of any sort, much less those designed specifically for curing tobacco. The house remains habitable and the outbuildings are in good condition. The building stock is unexceptional and comparable examples can be found on farms throughout the region. The occupants, while respected in the community, were not prominent in the state or region, and there are no events of exceptional historical interest known to have occurred on the Compton farm.

102

Figure 10-15. Corn crib.

103

Chapter 11. Summary, Conclusions, and Recommendations This study of the Compton farm and western portion of the Stone farm resulted in the identification of four archaeological sites and eight above-ground structures. Of the buildings, all appear to be 20th century and only two remain in good condition: the main house and one tobacco barn. None of the buildings meet any of the four criteria for inclusion into the National Register of Historic Places. One of the archaeological sites— the probable Late Archaic site (Garrett‘s Chance #1, 18PR702) adjacent to Swanson Creek—will not be impinged upon by proposed construction. Garrett‘s Chance #4 (18PR709) is a heavily eroded deposit of domestic refuse, possibly related to the extant farmhouse above, and dates to the middle of the 19th through early 20th centuries. A shovel-test confirmed the lack of stratigraphic integrity. It meets none of the National Register criteria due to the length of time over which the material accumulated and the eroded nature of those deposits. Garrett‘s Chance #3 (18PR704) is a prehistoric site, possibly a single-component Paleoindian site yielding one Clovis point, a Bare Island biface, and a quantity of fire- cracked rock and some flakes. Testing failed to uncover any trace of subsurface features or densities of artifacts in the plowzone that might indicate some value in plowzone testing. The site is severely eroded and, as such, does not meet National Register criteria. Still, given its upland location that precludes the possibility of the Clovis point having been redeposited, the site has information value in terms of the kind of setting occupied and the prospect for finding additional Paleoindian sites in this area, possibly in better condition. Garrett‘s Chance #2 (18PR703) proved to be a relatively well-preserved late 17th/early 18th century plantation site, possibly occupied by tenants until the 1730s or 1740s. It is remarkable in that it lies on a small patch of relatively intact Ap-horizon in the midst of severely eroded, gravelly soils. A single dwelling—extensively repaired in the second quarter of the 18th century and burned some years later—and a series of trash- filled borrow pits constitute the site. Garrett‘s Chance #2 (18PR703) does not meet Criteria A or B; its occupants were not particularly prominent in the history of the area and no important historical events occurred at the site, so far as I can determine from intensive historical research on the community, the property, and its occupants. The site had the potential to contribute to our knowledge of architectural patterns in the Chesapeake region, especially in the interior uplands, during a period of increasing social, economic, and political stability; and as such, met Criterion C. Severe erosion of the site along its southeastern and southwestern margins, and the relatively low-yield of artifacts from the plowzone, indicated that the site likely had little potential for revealing patterns of homelot organization, which also would have made the site eligible under Criterion C. The likely survival of intact subsurface features held the promise of yielding information on patterns in the use of wealth and food acquisition, making the site eligible for National Register listing under Criterion D. The applicant, in preparing his application for subdivision under the new policies of the County, determined that subdivision and development would not be feasible because Garrett‘s Chance #2 (18PR703) was in the midst of one of the parcel‘s two premier lots. Having exhausted redesign options, the applicant requested mitigation of impacts to the site and authorized an intensive study that would ‗consume the resource.‘

104

Our investigations identified a small two-bay earthfast dwelling with a wattle and daub chimney, but lacking a chimney bay. The building was entirely renovated sometime in the early 18th century, possibly when the land was conveyed to William Wilkerson by the Johnson-Willery family in the second decade of the 18th century. The renovated building burned with architectural and some domestic materials collapsed into the root cellar by the hearth. Perhaps the use of a hood rather than a separate chimney bay contributed to the start and spread of the fire. Surrounding pits provided material for the original construction of the chimney and the extensive repair, if not total replacement of the chimney when the building was renovated. Smaller pits probably were dug for small repair jobs. The resulting open pits accumulated refuse, including burned and unburned daub from the chimney. The material culture recovered from the pits and, to a lesser extent, from the overlying plowzone, included kitchenwares in relatively small numbers, especially when compared to such contemporary and near contemporary rural sites as Patuxent Point (18CV271) and Kings Reach (18CV83). This relative paucity likely accounts for the site‘s low visibility on the plowed surface and in the shovel-tests and single excavation unit: plowing had truncated the features, but the artifact content of the features was small and comprised largely of ash and burned daub. The features also produced a couple of tools, including a single bit felling axe head and a possible mill pick. Ethnobotanist Justine W. McKnight extracted a quantity of organic material from 10 liters of soil from the south half of Feature 22. Among the carbonized remains of local trees (oak, poplar, beech, hickory, walnut, chestnut), grasses and sedges, she recovered 25 fragments of maize. Maize was a staple of the Colonial diet in the Chesapeake region. Feature 22 also was rich in faunal material—including fish, bird, cattle, swine, sheep, and deer—and produced a few unusual utensils: a key, a dozen pins, and an incised piece of bone that likely was part of a decorative hair comb. Overall, the site represents a succession of at least two households, neither of which used much wealth in the acquisition of manufactured goods beyond those that met the immediate needs of life and agriculture in the Colonial Tidewater region. The house was simple and perhaps, ultimately, unsafe; but it was not impermanent. The occupants maintained the building and it likely stood upwards of 20 years. The remaining fields of the project area yielded scatters of prehistoric and historic materials, all diffuse and all situated on severely eroded slopes. Materials recovered from the surfaces were moved to those locations by uniformly poor cultivation practices on steep, highly erodable slopes. The wooded portions of the project area, subjected to surface reconnaissance and limited shovel testing, and the two proposed lots currently in pasture and subjected to shovel testing, proved to be severely eroded; no doubt accounting for their having been taken out of cultivation more than 20 years ago. The combined Phase 1 through 3 investigations have exhausted the research potential of the area proposed for subdivision, although scholars undoubtedly will revisit the collections for Garrett‘s Chance #s 2 and 3. I recommend no further archaeological study in connection with the proposed subdivision of the Compton farm and the western portion of the Stone farm.

105

References Cited Balicki, Joseph, Bryan Corle, and Kristin Falk 2004 Data Recovery Investigations at the Homeland Brick Clamp (Site 18CH664), MD 5–Hughesville Bypass, Hughesville,Charles County, Maryland. Maryland State Highway Administration Archeological Report Number 296. Baltimore. Ballweber, Hettie L. 1994 Return to the Luce Creek Site (18AN143). Maryland Archeology 30(1): 1–16. Barse William P. 1982 Results of a Phase I Archeological Reconnaisance of the Proposed Ryceville- Loveville Transmission Line, Charles and St. Mary‘s County, Maryland. Thunderbird Archeological Associates. Submitted to the Southern Maryland Electric Cooperative. Barse William P., Daniel B. Eichinger, Marvin a. Brown, and E. Madeleine Scheerer 1999 Phase I Terrestrial Archeological Survey, Maryland Route 5–Hughesville, Charles County, Maryland. Maryland State Highway Administration Archeological Report Number 248. Baltimore. Berry, Brian J.L. 1967 Geography of Market Centers and Retail Distribution. Prentice–Hall, Princeton, New Jersey. Brown, Lois 1979 Fluted Projectile Points in Maryland. Paper distributed by the Council for Maryland Archeology. Carbone, Victor 1976 Environment and Prehistory in the Shenendoah Valley. Unpublished Ph.D. dissertation, Department of Anthropology, The Catholic University of America, Washington, D.C. Carson, Cary, Norman F. Barka, William M. Kelso, Garry W. Stone, and Dell Upton 1981 Impermanent Architecture in the Southern Colonies. Winterthur Portfolio 16: 135-196. Cheek, Charles H., Amy Friedlander, and Cheryl A. Holt 1983 A Phase II Investigation of the Prehistory and History of Five Sites in St. Mary‘s County, Maryland. Soil Systems, Alexandria, VA. Submitted to Southern Maryland Electric Cooperative, Hughesville, MD. Custer, Jay F. 1984 Delaware Prehistoric Archaeology: An Ecological Approach. University of Delaware Press, Newark. 1986 Late Woodland Cultures of the Middle Atlantic Region. University of Delaware Press, Newark.

106

Delcourt, Paul A., and Hazel R. Delcourt 1981 Vegetation Maps for Eastern North America: 40,000 YR BP to the Present. In Geobotany II, R. C. Romans, editor, pp. 125-165. Plenum Press, New York. Dent, Richard J., Jr. 1995 Chesapeake Prehistory: Old Traditions New Directions. Plenum Press, New York. Ebright, Carol A. 1992 Early Native American Prehistory of the Maryland Western Shore: Archeological Investigations at the Higgins Site. Maryland State Highway Administration, Archeological Report Number 1. 3 vols. Ford, T. L. 1976 Adena Sites on the Chesapeake Bay. Archaeology of Eastern North America 4: 63–89. Gardner, William M. 1982 Early Woodland in the Middle Atlantic: An Overview. Occasional Papers of the American Archaeological Institute 3:53–87. Gardner, William, Carole L. Nash, Joan M. Walker, and William P. Barse 1989 Excavations at 18CV272. Submitted to CRJ Associates, Camp Springs, Maryland. Gibb, James G. 2000 Phase I Intensive Archeological Survey of the Nubian Road and Foster Lane Realignments, MD 5 Hughesville Bypass, Charles County, Maryland. Maryland State Highway Administration Archeological Report Number 248. Baltimore. 1996 The Archaeology of Wealth: Consumer Behavior in English America. Plenum Press. 1991 An Intensive Systematic Surface Collection of the Allens Fresh Number 1 Site on Zekiah Swamp/Allens Fresh, Westwood Point, Charles County, Maryland. Gibb, James G., and Anson H. Hines 1997 Selby Bay Phase Subsistence Strategies at the Smithsonian Pier Site, Anne Arundel County, Maryland. Maryland Archeology 33 (1 & 2) 59-76. Griffith, Daniel R. 1980 Townsend Ceramics of the Late Woodland of Southern Delaware. Maryland Historical Magazine 75(1): 23–41. Hickey, Joseph Vincent 1970 The Prehistory of Southeastern Charles County, Maryland: An Archaeological Reconnaissance of the Zekiah Swamp. Unpublished M.A. thesis, Department of Anthropology, George Washington University.

107

Hopkins, Joseph W., Melanie D. Collier, and Benjamin R. Fischler 1992 Phase I Archeological Survey of the Proposed BG&E/PEPCO Calvert Cliffs to Chalk Point 500-kV Transmission Line, Calvert and Prince George‘s Counties, Maryland. Greenhorne & O‘Mara, Greenbelt, Maryland. Submitted to Baltimore Gas & Electric Company, Potomac Electric Power Company, and Black and Veatch. Hopkins, G. M. 1878 Atlas of Fifteen Miles Around Washington, Including the County of Prince George, Maryland. G. M. Hopkins, C.E., Philadelphia, PA. Hranicky, William Jack 1994 Middle Atlantic Projectile Point Typology and Nomenclature. Archeological Society of Virginia, Special Publication No. 3.. Ingersoll, Daniel W., and J.J. Kenney 1978 Archaeological Survey Report on the Transmission Line, Chalk Point to the Patuxent River via Aquasco. Submitted to the Southern Maryland Electric Cooperative. Kavanagh, Maureen 1982 Archeological Resources of the Monocacy River Region, Frederick and Carroll Counties, Maryland. Maryland Geological Survey, Division of Archeology File Report Number 164. Keeler, Robert W. 1978 The Homelot on the Seventeenth-Century Chesapeake Tidewater Frontier. University Microfilms, Ann Arbor. King, Julia A. 1988 A Comparative Midden Analysis of a Household and Inn in St. Mary‘s City, Maryland. Historical Archaeology 22(2): 17-39. King, Julia A., and Henry M. Miller 1987 A View from the Midden: An Analysis of Midden Distribution and Composition at the Van Sweringen Site, St. Mary‘s City, Maryland. Historical Archaeology 21(2): 37-59. Kirby, Robert M., Earl D. Matthews, and Moulton A. Bailey 1967 Soil Survey of Prince George‟s County, Maryland. U.S. Department of Agriculture, Soil Conservation Service. U.S. Government Printing Office, Washington, DC. Land Records of Prince George‘s County n.d. Land records, patents, warrants, rent rolls, wills, and guardian accounts available at the Maryland State Archives, Annapolis, and the Charles County Courthouse, Annapolis, Maryland.

108

Lawrence, Dawson 1878 Historical Sketch of Prince George County, MD. In Atlas of Fifteen Miles Around Washington, Including the County of Prince George, Maryland, pp. 7- 9. G. M. Hopkins, C.E., Philadelphia, PA. Lippson, Alice Jane 1973 The Chesapeake Bay in Maryland: An Atlas of Natural Resources. The Johns Hopkins University Press, Baltimore. Luckenbach, Alvin H., Wayne Clark, and Richard S. Levy 1987 Rethinking Cultural Stability in Eastern North American Prehistory: Linguistic Evidence from Eastern Algonqian. Journal of Middle Atlantic Archaeology 3:1–33. Lukezic, Craig 1990 Soils and Settlement Location in 18th Century Colonial Tidewater Virginia. Historical Archaeology 24 (1): 1–17. Main, Gloria L. 1982 Tobacco Colony: Life in Early Maryland, 1650-1720. Princeton University Press, Princeton. Marks, Bayly Ellen 1979 Economics and Society in a Staple Plantation System: St. Mary‘s. University Microfilms, Ann Arbor, MI. Martenet, Simon J. 1885 Martenet‟s Map of Maryland and District of Columbia including a sketch of Delaware and part of Pennsylvania, Virginia and West Virginia. Simon J. Martenet, Baltimore. Maryland Department of Transportation 1986 Specifications for Consulting Engineers‟ Services, Section IV. Maryland Department of Transportation, Baltimore, Maryland. Maryland State Highway Administration 1992 Specifications for Archeological Procedures. Maryland Department of Transportation, State Highway Administration, Baltimore, Maryland. Nassaney, Michael S., and Kendra Pyle 1999 The Adoption of the Bow and Arrow in Eastern North America: A View from Central Arkansas. American Antiquity 64 (3): 243–263. Pogue, Dennis J. 1990 King‘s Reach and 17th-Century Plantation Life. Jefferson Patterson Park & Museum Studies in Archaeology No. 1. St. Leonard, Maryland. 1984 Town Rearing on the Maryland Chesapeake Frontier, A Reinterpretation. Paper presented at the annual meeting of the Society for Historical Archaeology. Reeve, Stuart A.

109

1992 Changes in Time: A Seriation Chronology of Southern Maryland Projectile Points. Journal of Middle Atlantic Archaeology 8:107-137. Schubel, J.R. 1981 The Living Chesapeake. The Johns Hopkins University Press, Baltimore. Shaffer, Gary D., and Elizabeth J. Cole 1994 Standards and Guidelines for Archeological Investigations in Maryland. Maryland Historical Trust Technical Report Number 2. Smolek, Michael A. 1984 ―Soyle light, Well–Watered and On the River‖: Settlement Patterning of Maryland‘s Frontier Plantations. Paper presented at the Third Hall of Records Conference on Maryland History, St. Mary‘s City, MD. Steponaitis, Laurie C. 1983 An Historical Study of the Patuxent River Drainage, Maryland. Maryland Historical Trust Monograph Series 1. 1980 A Survey of Artifact Collections from the Patuxent River Drainage, Maryland. Submitted to the Maryland Historical Trust and the Coastal Zone Administration, Department of Natural Resources, Annapolis, Maryland. 1978 An Archeological Survey of the South River Drainage Basin, Anne Arundel County, Maryland, Maryland Geological Survey, Division of Archeology File Report Number 138. Stephenson, Robert L. and Alice L.L. Fergeson 1963 Accokeek Site: A Middle Atlantic Seaboard Cultural Sequence. Museum of Anthropology, University of Michigan Anthropological Papers Number 20. Ann Arbor, Michigan. Stiverson, Gregory A. 1977 Poverty in a Land of Plenty: Tenancy in Eighteenth Century Maryland. The Johns Hopkins University Press, Baltimore. United States Geological Survey 1953 U.S.G.S. 7.5–Minute Topographic map, Benedict, Maryland. Photorevised in 1974. U.S. Government Printing Office, Washington, D.C. Wanser, Jeffrey C. 1982 A Survey of Artifact Collections from Central Southern Maryland, Maryland Historical Trust Manuscript Series Number 23. Wesler, Kit W., Dennis J. Pogue, Al Luckenbach, Gordon J. Fine, Patricia A. Sternheimer, and Elisabeth Glyn Furgurson 1981 The Maryland Department of Transportation Archeological Resources Survey, Volume 2: Western Shore. Maryland Historical Trust Manuscript Series Number 6.

110

Wright, Henry T. 1968 A Report on the Archeological Survey of the Chesapeake Bay Center for Field Biology of the Smithsonian Institution. Report submitted to the Smithsonian Institution. 1969 A Report on the Archeological Survey of the Chesapeake Bay Center for Field Biology of the Smithsonian Institution. Report submitted to the Smithsonian Institution. 1973 An Archeological Sequence in the Middle Chesapeake Region, Maryland. Maryland Geological Survey, Archeological Studies Number 1. Wyckoff, V. J. 1937 The Sizes of Plantations in Seventeenth–Century Maryland. Maryland Historical Magazine 32: 331–339.

111

Appendix A: Shovel Test and Artifact Data

112

Appendix B: Artifact and Sample Catalogue

113

Appendix C: Archaeobotanical Report by Justine W. McKnight Archeobotanical Remains from Feature 22, the Garrett’s Chance Site (18PR703), Prince Georges County, Maryland

1. INTRODUCTION Ten liters of soil and one sample of waterscreen-recovered material from Feature 22 at the Garrett‘s Chance site were submitted for macro-botanical analysis. Feature 22 describes a pit feature formed in the early 18th century and relating to domestic contexts within a dwelling of earthfast construction. The analysis of archeobotanical remains was undertaken in order to better understand historic plant use as well as landscape form and function at the site.

A. Flotation Sample (Sample No. 1) The soil sample was individually processed using a Flote-Tech flotation system equipped with 0.325mm fine fraction and 1.0mm coarse fraction screens. The Flote-Tech system is a multi-modal flotation system which facilitates the separation and recovery of plant macro-remains from the soil matrix by agitation in water. Processing resulted in two (light and heavy) fractions of material. Floted portions were air dried.

In addition to botanical remains, the flotation sample contained an array of small artifacts. These included brick, daub, iron hardware, and faunal remains (bone, shell, teeth).

All plant remains recovered through flotation were combined and passed through a 2mm geological sieve, yielding fractions of 2 different sizes for analysis. The greater-than-or-equal-to 2 millimeter fraction was examined with a binocular microscope under low magnification (10X to 40X) and sorted into broad categories of material (wood, seed, cultigen, et cetera.). The less-than 2 millimeter fraction was examined under low magnification and the remains of seeds and cultivated plants were removed for analysis. Each category of vegetative material was quantified by weight and specimen/fragment count. Weights were recorded using an electronic balance accurate to 0.01 grams.

Identifications were attempted on all seed, nut and cultivated plant remains, and on a sub-sample of 20 randomly-selected fragments of wood charcoal in accordance with standard practice (Pearsall 2000). Identifications of all classes of botanical remains were made to the genus level when possible, to the family level when limited diagnostic morphology was available, and to the species level only when the assignment could be made with absolute certainty. When botanical specimens were found to be in such eroded or fragmentary condition that complete examination or recognition was impossible, a variety of general categories were used to reflect the degree identification possible: General wood categories within the analyzed assemblage include „ring porous‟ where specimens exhibited differences between early and late wood growth; „semi-ring porous‟ where specimens revealed a less distinct division between early and late wood growth; „deciduous taxa‟ where a porous structure was apparent; and „unidentifiable‟, where specimens were so fragmentary or minute that no clear section could be obtained upon which to base identification.

All identifications were made under low magnification (10X to 40X) with the aide of standard texts (Martin and Barkely 1961; Schopmeyer 1974; Edlin 1969; Panshin and deZeeuw 1980) and checked against plant specimens from a modern reference collection representative of the flora of the project area.

114

B. Waterscreen-Recovered Sample (Sample No. 2) The waterscreen-recovered sample contained abundant carbonized plant material within a matrix of crushed daub, brick, faunal remains and dust. In order to most efficiently liberate the plant material from this matrix, the entire sample was subjected to flotation processing using the Flote-Tech system and following the procedure described above.

As Sample No. 1 and Sample No. 2 were secured from the same provenience, an abbreviated analysis was undertaken: The recovered plant remains were described and identified, but a quantitative analysis (counting and weighing of specimens) was not performed. The heavy and light fractions were passed through a 2mm geological sieve, yielding fractions of 2 different sizes for analysis. The greater- than-or-equal-to 2 millimeter fraction was examined with a binocular microscope under low magnification (10X to 40X) and all classes of plant material were noted and described. The less-than 2 millimeter fraction was examined under low magnification for the remains of seeds and cultivated plants.

Identifications were attempted on all seed, nut and cultivated plant remains, and on a sub-sample of 20 randomly-selected fragments of wood charcoal in accordance with standard practice (Pearsall 2000). Identifications of all classes of botanical remains were made to the genus level when possible, to the family level when limited diagnostic morphology was available, and to the species level only when the assignment could be made with absolute certainty. When botanical specimens were found to be in such eroded or fragmentary condition that complete examination or recognition was impossible, a variety of general categories were used to reflect the degree identification possible: General wood categories within the analyzed assemblages include „ring porous‟ where specimens exhibited differences between early and late wood growth.

All identifications were made under low magnification (10X to 40X) with the aide of standard texts (Martin and Barkely 1961; Schopmeyer 1974; Edlin 1969; Kozlowski 1972; Panshin and deZeeuw 1980) and checked against plant specimens from a modern reference collection representative of the flora of Prince Georges County, Maryland.

3. RESULTS OF ANALYSIS A. Flotation Sample (Sample No. 1) Ten liters of flotation-processed cultural fill produced 6.24 grams of carbonized archeobotanical material (a mean average of 0.624 grams of material per liter of soil). Recovered archeobotanical remains were both abundant and diverse, and the condition of recovered organic remains was excellent. A variety of economically important cultivated and wild plants were represented in the analyzed assemblage: These include deciduous wood charcoal (dominated by oak species); corn; seeds from weedy and forest taxa; and miscellaneous plant materials including a monocot stem fragment, small buds and amorphous carbon. An inventory of flotation-recovered plant remains from Sample No. 1 is presented in Table 01.

115

Table 01: Contents of Sample No. 1 (flotation-recovered)

sample number 1 feature 22 stratum 1 half south soil sample volume (liters) 10 total weight carbonized botanical remains (grams) 6.24

WOOD CHARCOAL (carbonized) (number of fragments) 427 total weight (grams) 5.67 Carya spp. (hickory) 2 Castanea dentata (American chestnut) 3 Fagus grandifolia (American beech) 2 Quercus spp. (white oak) 9 semi-ring porous 1 ring porous 1 deciduous 2 total identified fragments 20

CULTIVATED PLANT REMAINS (carbonized) (total count) 25 total weight (grams) 0.36 Zea mays (corn ) total specimens 25 cupule 4 cupule fragment 20 cupule with glumes 1

SEED REMAINS (carbonized) (total count) 14 total weight (grams) 0.02 Acalypha spp. (copperleaf) 3 Liriodendron tulipifera (yellow poplar) 1 Phytolacca americana (poke) 2 CYPERACEAE (sedge family) 7 POACEAE (grass family) 1

MISCELLANEOUS PLANT REMAINS (carbonized) (total count) 21 total weight (grams) 0.19 monocot stem fragment 1 small buds 3 amorphous carbon 17

B. Waterscreen-Recovered Sample (Sample No. 2) The archeobotanical remains contained within Sample No. 2 were abundant, diverse, and extremely well preserved. The sample contained all classes of plant material identified within Sample No. 1, confirming a reliance on oak woods, corn, and the presence of a variety of wild seeds. Sample No. 2 was unique in containing fragments of black walnut shell. An inventory of botanical remains from Sample No. 2 is presented in Table 02.

116

Table 02: Contents of Sample No. 2

sample number 2 feature 22 stratum 1 half south

WOOD CHARCOAL (carbonized) present Carya spp. (hickory) 9 Liriodendron tulipifera (yellow poplar) 1 Quercus spp. (white oak) 8 ring porous 2 total identified fragments 20

NUT REMAINS (carbonized) present Juglans nigra (black walnut) shell fragments present

CULTIVATED PLANT REMAINS (carbonized) present Zea mays (corn ) total specimens present cupule present kernel present cob segment present cupule fragment present cupule with glumes present

SEED REMAINS (non-carbonized) Liriodendron tulipifera (yellow poplar) present Rubus spp. (blackberry or raspberry) present

SEED REMAINS (carbonized) present Phytolacca americana (poke) present Rhus spp. (sumac) present CYPERACEAE (sedge family) present

MISCELLANEOUS PLANT REMAINS present deciduous twigs present small galls present rindlike material present amorphous carbon present

117

Appendix D: Credentials James G. Gibb, Ph.D 2554 Carrollton Road Annapolis, Maryland 21403 (443) 482–9593 Education 2003 Certification in Computer Aided Design and Drafting, Anne Arundel Community College, Department of Engineering Technologies. 1994 Ph.D. in Anthropology, Binghamton University 1985 M.A. in Anthropology, Binghamton University 1978 B.A. in Anthropology, State University of New York at Stony Brook Continuing Education 1996 University of Nevada Cultural Resource Management Program: Assessing the Archaeological Significance of Historical Sites. 1994+ Washington Ceramic Seminar: Master Class in Antique Ceramics. Professional Experience Twenty–seven years of archaeological field and laboratory experience in six eastern states and Arizona, on sites ranging in age from early prehistoric to late 19th century. Author of more than 60 technical reports. Twenty years of supervisory experience and eleven years as Principal Investigator in Sole Proprietorship consulting firm.

Select Publications 2000 Imaginary, But by No Means Unimaginable: Storytelling, Science, and Historical Archaeology. Historical Archaeology 33 (2): 1–6. 2000 Reflection, Not Truth, the Hero of My Tale: Responding to Lewis, Little, Majewski, and McKee and Galle. Historical Archaeology 33(2): 20–24. 1999 A Layperson‟s Guide to Historical Archaeology in Maryland. Archeological Society of Maryland. (Editor and contributor) 1997 Selby Bay Phase Subsistence Strategies at the Smithsonian Pier Site, Anne Arundel County, Maryland. Maryland Archeology. 33(1&2): 59–76. (with Anson H. Hines) 1997 Necessary but Insufficient: Archaeology Reports and Community Action. In ―In the Realm of Politics: Prospects for Public Participation in African–American and Plantation Archaeology,‖ edited by Carol McDavid and David W. Babson. Special Issue of Historical Archeology 31(3): 51–64. 1996 The Archaeology of Wealth: Consumer Behavior in English America. Plenum Press, New York. 1995 The History of Helb Barn. The Calvert Historian 10(2):5–18. (with Matt Croson) 1994 Dated Window Leads from Colonial Sites in Anne Arundel County, Maryland. Maryland Archeology 30(2):23– 28.(with Al Luckenbach) 1994 English Trade Tokens from a 17th Century Colonial Site in Southern Maryland. Maryland Archeology 29(1 & 2):55– 60. 1994 “Dwell Here, Live Plentifully, and Be Rich”: Consumer Behavior and the Interpretation of 17th Century Archaeological Assemblages from the Chesapeake Bay Region. UMI, Ann Arbor Michigan. 1993 Dutch Pots in Maryland Middens; or, What light from yonder pot breaks? Journal of Middle Atlantic Archaeology 9:67–86. (With Wesley J. Balla) 1993 Publishing in Local History Journals. Journal of Middle Atlantic Archaeology 9:41–48. 1991 Gender, Activity Areas and Homelots in the 17th Century Chesapeake Region. Historical Archaeology 25(4):109-131. (with Julia A. King)

118

1990 Making Cheese: Archaeology of a 19th Century Industry. Historical Archaeology 24(1):18-33. (with David Bernstein and Daniel F. Cassedy) 1989 History Exhibits and Theories of Material Culture. Journal of American Culture 12(2):27-34. (with Karen Lee Davis) 1988 Unpuzzling the Past: Critical Thinking in History Museums. Museum Studies Journal 3:41-45. (with Karen Lee Davis) PUBLICATIONS: PUBLIC INFORMATION AND INTERPRETATION submitted Fischer‘s Station on the Chesapeake Beach Railway, Anne Arundel County, Maryland (1908–1935). The Calvert Historian. 2000 The Lesson of the Selby Bay People: Oysters eaten 1,800 years ago have a moral for our times. Bay Weekly November 16–November 22, 2000. 2000 Linden: An Urban Farmstead in Prince Frederick, Calvert County, Maryland (1868–1988. The Calvert Historian 26: 39–55. 2000 Animating History at Colonial London Town. Chesapeake Life Magazine (January–February): 92–95. (with John Kille) 1999 Revolutionary Spirits: A Play in Two Acts. Performed at London Town Historic Park by the London Town Publik House Players, April 1999. 1998 Ghosts of London: A Play in Three Acts. Performed at London Town Historic Park by the London Town Publik House Players, October 1998; reprised October 1999. 1998 Letters from London: A Provident Visit. The New Bay Times August 6–August 12, 1998. 1998 Letters from London II. The New Bay Times June 25–July 1, 1998. 1998 Letters from London: Sheriff Rawlings Expected Trouble; He Found it. The New Bay Times May 28–June 3, 1998. 1997 The Dorsey–Bibb Tobacco Flue: Innovation and Entrepreneurship in Southern Maryland Agriculture. The Calvert Historian 11(2): 4–20. 1995 Helb Barn: A Pennsylvania German Barn in Calvert County. The Calvert Historian 10(2): 5–18. (with Matthew E. Croson) 1994 Railroad Ghosts. The New Bay Times 2(10): 14–16 (May/June 1994). Reprinted in The Calvert Historian 21(1): 63– 70. 1993 Chesapeake Bay Life: Finding History through Garbage. The New Bay Times 8(1):10 (July 29–August 11, 1993). Reprinted as ―Archaeological Clues to Life in Colonial Calvert County: The William Stephens Land Site, c.1660– 1680,‖ in The Calvert Historian 21(1): 7–16. 1990 A Road Without Rails: The Baltimore and Drum Point Railroad, 1868–1891. The Calvert Historian 5(2):20– 35.(With Paula F. Mask) 1990 Using Calvert County's Agricultural Censuses. The Calvert Historian. 5(2): 9–17.

119