APPENDIX F "ARCHAEOLOGICAL INVESTIGATIONS AT BATTERSEA: EXAMINATION OF YARD AREAS ADJACENT TO THE EAST AND WEST PAVILIONS"

APPENDIX F "ARCHAEOLOGICAL INVESTIGATIONS AT BATTERSEA: EXAMINATION OF YARD AREAS ADJACENT TO THE EAST AND WEST PAVILIONS" by Thomas F. Higgins, III and Robert R. Hunter, Jr. William and Mary Archaeological Project Center, Department of Anthropology The College of William and Mary Williamsburg, Virginia 1989

ARCHAEOLOGICAL INVESTIGATIONS AT BATTERSEA: EXAMINATION OF YARD AREAS ADJACENT TO THE EAST AND WEST PAVILIONS o 10 20 ~. lu. Prepared for Petersburg Museums City of Petersburg September 1989 6 93 The College Of '---::)- WILLIAM&MARY

ARCHAEOLOGICAL INVESTIGATIONS AT BATTERSEA: EXAMINATION OF YARD AREAS ADJACENT TO THE EAST AND WEST PAVILIONS Submitted to: Petersburg Museums City of Petersburg 15 West Bank Street Petersburg, Virginia 23803 Submitted by: William and Mary Archaeological Project Center Department of Anthropology The College of William and Mary Williamsburg, Virginia 23185 Project Director Robert R. Hunter, Jr. Project Archaeologist Thomas F. Higgins, III September 20, 1989

MANAGEMENT SUMMARY On July 18, 1989, an archaeological study was initiated at Battersea in Petersburg, Virginia, by the College of William and Mary's Archaeological Project Center (WMAPC). The purpose of this initial phase of study was to identify and evaluate archaeological resources immediately adj acent to the north end of the west pavilion and the northeast corner of the east pavilion in advance of stabilization efforts on these areas of the structure. The archaeological data recovered included evidence of a prehistoric procurement camp. The architectural remains encountered in the excavation were examined in light of their historical context with the Banister, May, and Waring family occupations. In addition, the ceramic consumption patterns of these households were evaluated and the data employed to address questions regarding room function and the use of adjoining yard spaces during these periods. ii

REPORT CONTRIBUTORS Authors: Thomas F. Higgins, III Robert R. Hunter, Jr. Graphic & Report Production Editor: Donald w. Linebaugh Graphics Contributor: Donald w. Linebaugh Artifact Inventory: Copy Editor: Deborah Davenport Suzanne Erena iii

TABLE OF CONTENTS Page Management Summary...... i i Report Contributors ~........................... iii Table of Contents...... List of Figures List of Tables i v v vi Introduction 1 Project Research Goals 4 Introduction 4 Anticipated Prehistoric Resources...... 4 Anticipated Historic-Period Resources 5 Battersea Biographies......... 5 Specific Research Issues....... 7 Site Examination Methods............ 10 Field Methods 10 Analytic Procedures 10 Archaeological Results: Interpretations of House and Yard 12 Introduction 12 West Pavilion: Excavation units A through C... 13 East Pavilion: Excavation units D through H... 25 Conclusion 36 Introduction 36 Summary 36 Recom.m.endations......... 38 References cited 39 Appendix A - Archaeological site Inventory Form 40 Appendix B - Artifact Inventory... 63 Appendix C - Feature Record........ 65 iv

LIST OF FIGURES 1 2 3 4 5 6 7 8 Project Area Location. Project Area and Environs. Location of Excavation units. Excavation unit A, North Wall Profile. Excavation units A and B Adjacent to West Pavilion (Note Extent of Foundation Wall Below Ground Surface).. Excavation Units A, B, and C Adjacent to West Pavilion with Identified Features. Excavation unit C Adjacent to West Pavilion (Note Trench Adjacent to Rebuilt Bulkhead Entrance). Remnants of Original Bulkhead Entrance in Excavation unit B.. Page 1 2 11 13 19 20 23 24 9 Excavation Units F and G, East Wall Profile..... 26 10 Excavation Unit H, North Wall Profile...... 27 11 Excavation units D Through H, Northeast Corner of East Pavili on 30 12 Excavation Units D Through H, Northeast Corner of East Pavilion 31 13 Remains of possible Waterproofing Trench (E-10, E-11) Adjacent to the East Pavilion Foundation Wall... 32 14 Excavation Unit E, South Wall Profile (Note Depth of Trench Below Foundation Wall)........... 33 15 Excavation Unit E, South Wall Profile...... 34 v

LIST OF TABLES Page 1 Inventory of Recovered Vessel Fragments 16 vi

INTRODUCTION On July 18, 1989, an archaeological study was initiated at Battersea in Petersburg, Virginia, by the College of William and Mary's Archaeological Project Center (WMAPC) (Figure 1 and 2). The purpose of this initial phase of study was to identify and evaluate archaeological resources immediately adjacent to the north end of the west pavilion and the northeast corner of the east pavilion in advance of stabilization efforts on these areas of the structure. This document presents the results of that study, including an evaluation of the archaeological resources, conducted in light of the recent historical and architectural assessment of the structure (Graham and Wenger 1988), and a discussion of specific research issues borne out by these studies. o~ MILES 'r PROJECT AREA, r-,_,/ '.,,/"""/ -' ;.! J'... /.', \ \- t ~,) / I ' I" " /" II'" :... " " '-\,. -' I..,', ~.I... I,I,,/~'~':~'~'~./~'~-> TENNESSEE NORTH.CAROLINA '.1 FIGURE 1 Project Area Location This project was conducted under the general direction of Robert R. Hunter, Jr. Thomas F. Higgins III served as Project Archaeologist and was responsible for the organization and implementation of the field program and report preparation. Archaeological Project Center staff members contributing to the field effort included Gunnar Brockett, Chris McDaid, John Fisher, Ellen Shlasko,Joe Jones, Jane Smith, Seth Pilgrim, Mike Purdy, Lee Ann Thurmin, Elizabeth Monroe, and Bruce Sterling. Laboratory processing and artifact analysis was conducted by Deborah 1

o 2000.11"0 r~~~: ~~-rv~( 0 feet I FIGURE 2 Project Area and Environs

Davenport. Donald W. Linebaugh oversaw the technical and administrative aspects of this project. Final drawings for this report were prepared by Yuj in Asai and Mr. Linebaugh. Plan drawings of Battersea were provided by the architectural firm of DePasquale & Associates of Richmond, Virginia. Fieldwork began on July 18 and was completed July 27, 1989. Field notes, artifacts, drawings, photographs, and other resources pertinent to the documentation of this project remain on file with the College of William and Mary Archaeological Project Center, Williamsburg, virginia. 3

PROJECT RESEARCH GOALS Introduction Although no archaeological investigations had been conducted on the Battersea property prior to those undertaken by the WMAPC, the property was clearly identified as an important historic resource. In addition, it was recognized that the property had a high potential for containing a rich prehistoric component. The primary goal of this field investigation was to recover potentially significant archaeological data that may have been destroyed as a result of stabilization efforts to portions of the east and west pavilions of Battersea. Although the planned excavations were limited, several important research issues were raised as to the archaeological importance of this undertaking. A brief overview of these research goals is presented below. Anticipated Prehistoric Resources Although the focus of the investigation was on recovering historic-period evidence, it was anticipated that information related to the property's prehistoric past would be encountered. Battersea is situated on a broad, high terrace fronting the Appomattox River. It is therefore an.excellent location for settlement by prehistoric groups from the Paleo-Indian (before 8000 B.C.), Archaic (8000 B.C. to 1200 B.C.), and Woodland periods (1200 B.C.- 1600 A.D.). The site type for the high terrace location would primarily be that of a procurement camp rather than a larger and perhaps more permanent base camp. Procurement camps are characterized by remains indicative of activities related to food processing, cooking, and lithic preparation. In many instances, such sites are the focus of specialized activities such as tool manufacture. These activities were limited in their extent and were carried out by groups that occupied sites for relatively short periods of time. Base camps, more frequently associated with Woodland Period groups, were usually occupied for longer periods. They are identified archaeologically by the presence of a dense concentration of artifacts, a wide variety of artifact types, and extensive remains of subsurface features. Evidence of complex activities associated with workshop areas and cooking is present. Primary lithic reduction and tool resharpening are often indicated by clusters of lithic debitage and decortication flakes, while extensive remains of pottery and charcoal reflect intensive cooking on the site. The lower floodpla.in portion of the Battersea property is most likely to produce evidence of prehistoric base camps. 4

Anticipated Historic-Period Resources The results of an archaeological investigation of an historic property such as Battersea are meaningful to the extent that they make substantive contributions to our understanding of a property's occupation for a period of over 200 years. Important elements in this study are historical documentation and architectural history. These areas of study often provide insight into the economic condition and social milieu in which people lived. Furthermore, they allow the material past to be explored from social, economic, and technologic vantage points--quite literally, from the dishes people used to the houses they built. By placing individuals and their properties in historical perspective, scholars can better evaluate the importance of an individual's contribution to an historical event or period and determine the significance of the structural remains which he or she occupied. As indicated by its listing on the National Register of Historic Places, John Banister's Battersea holds such historical importance. Battersea Biographies The following summary provides a brief glimpse of the i7thcentury botanist, John Banister; his son, John II, one of the founders of Petersburg; and his grandson, John III, who built Battersea. Individuals who owned Battersea during the antebellum period and decades following the civil War are also introduced. The summary concludes with mention of the owners of the property during this century. John Banister. Sr. Arriving in Virginia in 1677, John Banister pursued his botanical studies there with a high degree of success. His extensive education in botany and religion brought him into the elite circle of Virginia society. While having limited personal wealth, his scientific studies retained the sponsorship of wealthy and influential individuals such as William Bird I of Westover. Largely through his social and financial contacts, Banister laid the foundation on which his son and grandson would build economic, social, and political prominence in 18th-century Virginia. John Banister II. Following the death of his father in 1692, young John Banister II was raised by William Bird I at Westover. There he gained experience in plantation management and made contacts with many important people that served him well in later years. He emerged as a prominent individual in the area, as evidenced by his position as magistrate for Prince George county and vestryman for Bristol 5

Parish in 1738. John Banister, in the company of William Bird and others, helped to found the cities of Richmond and Petersburg in 1733. In addition, he purchased several properties in Petersburg, including the site that would become the Battersea estate. John Banister III. John Banister III retained and fostered his father's social and economic prominence. His social position was demonstrated in part by the large and impressive brick dwelling he built at Battersea in ca. 1770. This structure consisted of a two story central block with attached hyphens flanking either side. Adjoining the hyphens were east and west pavilions, forming the ends of the structure. The architectural style of Ba.nister' s mansion was alluded to by the traveller Marquis de Chastellux who visited Battersea in 1782 (Graham and Wenger 1988). Mr. Victor, who was still my guide, took me to the camp formerly occupied by the enemy. He expressed regret that I could not get a closer view of Mr. Banister's handsome country house, which I could see from where we were. There being no other obstacle however than the distance, about half a league, and the noonday heat, we determined that this should not stop us; and, walki~g slowly, we easily reached this house, which is really worth seeing, as it is decorated in more Italian, than American or English taste, having three porticoes at the three principle entrances, each of them supported by four columns. The structure was joined by numerous outbuildings, a formal garden, and a mill complex located several hundred yards to the north along the Appomattox River. Banister, whose millworks were quite prosperous, provided financial backing for other mill operations along the river. These investments benefited him financially and helped initiate early industrial development within Petersburg. With the outbreak of the Revolution, Banister the businessman became both solider and supplier for the continental Army. He served in the cavalry as a lieutenant colonel under General Lawson in 1781. SUbsequent to his service under Lawson, he returned to Battersea. Much of the plantation I s output was diverted toward the war effort. Bannister sold several hundred wagon loads of timber, blankets, gun powder, weapons, flour, and food stuffs to the continental Army. In addition, as much as 50 acres of his land was cleared for use as military stables and pasture. John Fitzhugh May Like his predecessors at Battersea, John May was a man of social, economic, and political prominence in the state. He was a member of the General Assembly, a judge of the Virginia Supreme 6

Court of Appeals, and a local vestryman. In the 17 years he owned Battersea, he apparently undertook maj or renovations to the dwelling (Graham and Wenger 1988). These included alterations to windows, doors, and porches. Interior spaces were also changed, including the creation of a double parlor from the east hyphen and pavilion, and the excavation of a cellar under the west hyphen. John and Catherine Waring The Warings purchased Battersea from John May in 1841. Their ownership of the property was relatively short lived owing to financial troubles and the death of John in 1847. Nonetheless, during their six years at Battersea, several alterations were made to the structure, including repairs to porches, installation of gutters, and the application of a thin layer of stucco to the exterior of the house. These improvements, specifically the repairs to the porches and the altered exterior, emphasized the classical character of the house. This architectural style was especially appealing to those caught up in the Greek Revival movement of the second quarter of the 19th century. Franklin wright Wright purchased the house in 1870. sensitive to its historic value, he made few major alterations to the structure. His efforts focused primarily on making necessary repairs rather than rebuilding. He added modern conveniences for the period, including gas lights, coal fireplaces, and a bathroom addition to the east end of the house. Dennie Perkinson and M. A. Finn Battersea was sold by the Wright family in 1905 to the trading partnership of Perkinson and Finn. Perkinson, eventually buying out Finn's interest in the property, occupied the house for over 40 years. It was retained in his immediate family until 1980 when it was passed by will to John D. McLaughlin, Jr. Over a period of 84 years, many repairs had been made to the structure. These included repairs to the stucco on the exterior of the dwelling during the first half of this century and, more recently, replacement of many of the cellar window frames and stabilization work in the east cellar. In addition, many conveniences were added during the Perkinson family occupation, including electricity and central heating. Specific Research Issues It was expected that excavation would allow data to be extracted from a variety of remains including architectural and domestic sheet refuse deposits, architectural remains, and yard 7

features. While these remains could yield data important in addressing a number of research issues, their significance in light of recent architectural interpretations of Battersea would be the focus of the investigation. It is important to note, however, that the presence of archaeological remains within small excavation areas usually generates as many new questions as it does substantive answers to the original questions. with this in mind, several general and specific research issues guided the excavation strategy. 1) As no archaeological investigations had yet been conducted within the yard area, the excavation would permit a preliminary assessment of the nature and condition of archaeological deposits and features in the yard. 2) Documentary and architectural study of the house had raised several important issues concerning the chronological evolution of the house that may be addressed by archaeological data: the archaeological excavation may provide alternative documentation of the construction date of the building (1765-1775) through the examination of the contents of the original builder's trench; evidence of remolding phases of the house would be sought as revealed through subsoil deposits removed from cellar expansion periods, deposits of architectural materials such as nails, window glass, mortar, and stucco. 3) Preliminary analysis of the architectural arrangement of the rooms had raised several hypotheses concerning the functional division of space within the Banister and May households (Graham and Winger 1988). During Banister's tenure, the main floor of the west pavilion appears to have functioned as a service areal servants' hall, while the east pavilion served as a formal dining area. The architectural evidence suggests that when May acquired the property in 1824, the function of these two rooms may have changed; the west pavilion subsequently becoming the dinning room while the east pavilion functioned as a parlor. Assuming that household refuse deposition patterns will allow an architectural examination of these hypotheses, the archaeological test implications of the shift in function of these rooms suggest that a higher percentage of refined ceramic and glass tableware should be found outside the east pavilion while cheaper and more utilitarian ware delegated to servant dining should be concentrated outside of the west pavilion for the period ca. 1770-1824. 8

conversely, refined ceramic and glass tableware of the 1824-1850 period should be more frequent behind the west pavilion. A continued presence of cheaper and more utilitarian wares would be expected given that the cellar beneath the west pavilion served as a servants' area. Those few artifacts anticipated in association with the east pavilion function during this period should relate to breakage of tea wares and other parlor-related objects. 9

SITE EXAMINATION METHODS Field Methods The initial phase of archaeological investigation of the Battersea property was undertaken within yard areas immediately adjacent to the east and west pavilions of the structure (Figure 3). These areas, consisting of approximately 282 square feet, were threatened by scheduled stabilization work on the building. This stabilization effort would effectively destroy archaeological remains within these areas. Consequently, site examination methods designed to capture archaeological data pertinent to the research questions posed above were implemented. Limited block excavation around the foundations allowed the discernment of architectural features related to the building episodes and the identification of features related to yard use. Cultural layers, many containing 18th- and 19th-century domestic and architectural sheet refuse deposits, were also recovered. The two excavation areas located at the east and west ends of the structure were arbitrarily divided into a total of eight excavation units ranging from 5 by 3 feet to 8 by 5 feet in size. These units were designated by letters A through H. CuItural layers and features identified within each unit were assigned consecutive context numbers. These deposits were recorded by detailed plan and section drawings (Appendix A), as well as black and white photographs and color slides. Analytic Procedures All artifacts were washed, sorted, and labeled by provenience. The preliminary inventory uses a standard descriptive typology for both prehistoric and historic-period materials (Appendix B). Obvious vessel and other functional characteristics were also noted. No conservation was attempted on any of the metal or faunal material. Although the excavation areas were relatively small, and hence limited the quantity of material recovered, certain analytic procedures were employed to aid in the spatial interpretation of data. In the analysis of sheet refuse, the distribution of artifacts was plotted in an attempt to observe patterns of change or continuity in the function of the east and west pavilions during the Bannister and May family tenures. More specifically, ceramics recovered from these occupations were sequenced according to period and type; these data were then examined in light of the documented transition in property ownership. 10

Bulkhead Entrance cq I I West Pavilion UP II II II II II ~- --- UP II II ----~- East Pavilion I I ~ ed I I 10, feet 20 I FIGURE 3 Location of Excavation units

ARCHAEOLOGICAL RESULTS: INTERPRETATIONS OF HOUSE AND YARD Introduction The research issues identified for the first phase of archaeological investigations at Battersea were defined largely by the results and recommendations of the previous architectural and historical study (Graham and Wenger 1988). These issues were also influenced by the limits imposed by the location and extent of the excavation area. Consequently, the results of the archaeological investigation provide only a brief glimpse into the architectural "life lt of Battersea and, to an even lesser extent, into that of its occupants. Nonetheless, data significant to the interpretation of the building's architectural history and the lifeways of its occupants can be retrieved. As stated earlier, answers to questions regarding periods of building construction, repair, and function can sometimes be gleaned from archaeological data. Architecturally related features such as builder's trenches, scaffolding postholes, and original bulkhead entrances are often present in yard areas immediately adjacent to a structure. Furthermore, dated sheet refuse scatter of architectural debris can reflect periods.of construction, destruction, or repair. These archaeological remains frequently contribute to a more accurate interpretation of the construction detail of a house (Noel Hume 1975). Much like architectural debris, the domestic sheet refuse found at Battersea, including fragments of ceramics, glass, and bone, can corroborate or refute architectural interpretations of possible room function at different periods of time. In addition, the analysis of the sheet refuse scatter within this area provides interesting data for comparison with similar deposits located adjacent to the west pavilion. Analysis includes determining the percentage of refined ceramic and glass tableware compared to utilitarian wares within this area, comparing their quantities as well as to those of similar materials recovered adjacent to the west pavilion. The prevalence of tea wares in the ceramic assemblages recovered from sheet refuse deposits adjacent to both the east and west pavilions must also be assessed. These materials, examined even in relatively small quantities, can often shed light on specific questions as well as address a broad range of research issues in reconstructing the lifeways of site occupants. Similarly, some sense of the historic yard landscape can be identified from archaeological data recovered from cultural deposits adjacent to the house. The remains of fenceline postholes, walkways and service porches, water drains, and domestic vegetation can be identified. 12

In view of the high research potential of these archaeological remains, care was taken to ensure their thorough documentation and interpretation in view of the existing historical documentation for the property. West Pavilion: Excavation units A through C cultural Layers Excavation units adjacent to the west and north walls of the west end of the structure revealed several distinct soil layers (Figure 4). Perhaps the best defined of these cultural layers was the upper deposit, which consisted of a dark brown sandy loam (Layers A-I, B-1, and C-l). This topsoil layer, ranging in thickness from.5 to.7 of a foot, contained a scatter of architectural and domestic refuse, as well as a prehistoric lithic scatter. The historic material consisted of a sizeable quantity of refined earthenware, including the base fragment of a plate marked "H(OMER LAUGHLIN}/EMPRE(SS}/160." The presence of this ceramic dates the layer to post-1906. In addition to these late period ceramics, numerous fragments of 20th-century bottle glass were recovered. Architectural debris included.a scatter of wire nails and window glass. Prehistoric material consisted of quartz decortication flakes and chipping debris. w A-I A-I A-2 SUBSOIL A-4 A-3 0 A-2 PIPES 0 I 2 ; I feet O\) BRICKS SUBSOIL FIGURE 4 Excavation unit A, North Wall Profile Sealed beneath topsoil was a brown sandy loam mottled with orange clay (Layers A-2, B-2, and C-4). This layer, measuring up to.4 of a foot deep, contained a mix of 18th-, 19th-, and early 20th-century ceramics and glass. Numerous fragments of creamware 13

and pearlware were found in association with fragments of later ceramic types, such as ironstone and porcellaneous. Over 25 fragments of 20th-century colorless bottle glass were recovered. Architectural material included a mix of hand wrought, cut, and wire nails, as well as screws, bolts, hinges, and a fragment of roof slate. Similar to the prehistoric material recovered from the upper layer, this layer contained a light scatter of quartz and quartzite flakes, including several decortication flakes. No prehistoric pottery was recovered. The presence of the late historic period ceramics and relatively recent architectural debris dates the layer to the 20th century. Beneath this layer was a light brown sandy loam (Layers A-4, B-3, and C-12). Like the second layer, the third layer was mottled with orange clay and measured.2 to.4 of a foot thick. It contained a variety of ceramic and glass, including pearlware, creamware, porcelain, and ironstone. Prehistoric material was limited to seven lithic fragments, including several decortification flakes. Found in association with this material were fragments of 19th- and 20th- century bottle glass, window glass, and fragments of lamp chimney glass. Sealed by the sandy loam and orange clay subsoil was a layer of orange sandy clay loam (Layers A-6 and B-6). This layer, not clear in profile, was relatively thin, measuring only.2 of a foot thick. As was characteristic with the sealant layers, this layer contained a scatter of late 18th-, 19th-, and 20th-century material. Fragments of creamware and pearlware were found in association with whiteware. Mixed with this material were several fragments of 19th- and 20th-century bottle glass. The orange sandy clay loam was confined to excavation units A and B and was not identified in unit C. within unit C however, sealed beneath Layer C-12, was a thin layer of yellow sandy loam. This deposit contained a dense concentration of prehistoric material. A total of 74 lithic fragments were recovered, including a sizeable quantity of quartzite flakes and chipping debris. This assemblage also included fragments of creamware, delft, and bottle and table glass. The presence of creamware dates the deposit to the third quarter of the 18th century. The cultural layers identified above show considerable evidence of having been mixed. This is indicated by the presence of colonial and early post-colonial period ceramics with recent materials within topsoil and by the presence of late 19th- and early 20th-century bottle glass in the lowest soil layers. Likewise, the presence of prehistoric material in association with the historic debris points to disturbed soil layers. The orange clay mottling found in several of the layers suggests further that the soils had been mixed, possibly through planting activity. Planting disturbance is also indicated by numerous root holes which intruded the subsoil. Domestic vegetation planted along this side 14

structure during the late 19th and early 20th centuries resulted in some mixing of cultural layers. Sheet Refuse Deposits Although disturbance to cultural layers was evident, remains of sheet refuse scatters were present. Through archaeological investigation, such deposits are frequently identified adjacent to dwellings and their support structures. They represent accumulations of domestic and architectural debris intentionally scattered from doorways, bulkhead entrances, windows, and porches. While these deposits are often present around the perimeter of a structure, they are more prevalent in yard areas where servicerelated activities occurred. Hence, they are useful in defining principal work areas (Higgins 1989; Hunter et ala 1987). Furthermore, from the types of debris contained within the deposits, a delineation of the type of activities undertaken can be made. In some instances, layered debris served a functional purpose. Areas of high activity within a yard, frequently resulting in muddy ground, were often improved and made more passable by the addition of crushed and scattered refuse (Noel Hume 1975). The sheet refuse deposits identified within excavation units A, B, and C contained a sizeable quantity of material dating to the late 18th and 19th centuries. Most notable within this assemblage were ceramics represented by 273 sherds, 21 percent of the total material recovered from this area. Table glass made up only 2 percent of the total assemblage, while bone accounted for only 5 percent. Architectural refuse, including nails, spikes, hinges, window glass, and slate, accounted for 42 percent of the total assemblage from this area. An attempt was made to analyze the material in terms of household use. Perhaps the most informative conclusions drawn from this level of study were obtained from ceramic analysis (Table 1). First, individual ceramic types were analyzed in terms of their dates of manufacture and the known periods of their popularity. Second, the minimum number of vessels present and the vessel forms represented were identified. A minimum vessel count recognizes that a certain percentage of vessels can not be reliably identified from the comparison of vessel attributes, most commonly rims and bases. The minimum vessel count, therefore, represents a conservative estimate of the total number of vessels present within an assemblage of sherds. These data are useful in that they often provide important insight into the consumption behavior of a household; that is, a family's acquisition, use, re-use, and disposal of ceramics (Hunter 1987). The following is a summary of the minimum number of vessels, vessel forms and types, and the relative costs of vessels as 15

CONTEXT WARE DECORATION PORM NO. DATE C-ll E-2 C-ll C-l B-2 C-12 E-9 B-4 Gen,A-l,E-12 C-I0 C-12 A-2,0-5,G-2 C-20 C-15 C-12,H-3 B-12 B-2 A-4 C-12 C-12 B-2;B-4 A-2 A-7 A-4 A-4 0-5 C-12,E-9 A-4 C-21 C-12 B-3,B-6 B-2 B-2 A-8 Gen,B-2,F-5,G-2 E-3 E-ll B-4 B-6 B-15 C-12 B-2 C-12,E-9 C-12 E-9 E-ll C-9,C-12 A-7,E-3 F-3 A-4,B-6 A-4,C-9 E-9 A-4 E-2 B-1 Gen,E-9 Gen,A-2,A-4 B-1 B-6 Brown stoneware Brown stoneware Chinese porcelain Chinese porcelain Chinese porcelain Chinese porcelain chinese porcelain Chinese porcelain Chinese porcelain Coarse earthenware coarse earthenware Coarse earthenware Delftware Redware White saltglazed stoneware Black basaltes stoneware Crearnware creamware creamware Creamware Pearlware Pearlware Pearlware Pearlware Pearlware Pearlware creamware Pearlware Pearlware pearlware Pearlware pearlware Pearlware Pearlware pearlware pearlware Brown stoneware Brown stoneware Brown stoneware Brown stoneware Brown stoneware pearlware pearlware Albany slip stoneware Albany slip stoneware Albany slip stoneware pearlware Whiteware Whiteware Whiteware Whiteware Yellowware Bennington Ware Whiteware Whiteware Ironstone Ironstone Whiteware Whiteware undecorated undecorated overglaze underglaze blue undecorated underglaze blue, brown-edged underglaze blue underglaze blue undecorated undecorated trailed slip undecorated undecorated slipped interior barley molded printed black undecorated undecorated undecorated painted blue painted blue undecorated undecorated shell-edged blue shell-edged green undecorated banded green painted blue, fluted dipped, embossed painted polychrome painted polychrome painted polychrome printed blue printed blue painted blue undecorated undecorated undecorated undecorated undecorated printed blue, scalloped painted polychrome undecorated undecorated cobalt-decorated printed olive green printed brown printed brown printed red printed light blue undecorated undecorated flow blue flow blue molded molded flow blue flow blue Hollowware Hollowware Can Cup Flatware Flatware Bowl? Plate Teabowl Hollowware Hollowware Flowerpot Bowl Cup Plate Teapot Hollowware Cup Lidded cup/ pot 1 Hollowware 1 Cup 1 Flatware 1 Teabowl 1 Pudding pan 1 Plate 2 Plate 1 Plate 2 Hollowware 1 Saucer Hollowware Bowl Hollowware Cup/teabowl Mug? Plate Vegetable dish Bottle Hollowware Hollowware Hollowware Ale/seltzer bottle Saucer Hollowware Hollowware Jar/pot Jar/pot Plate Plate Saucer Plate Cup Indeter. Hollowware Hollowware Saucer Bowl Cup Plate Platter 1 18th c. 1 18th c. 1 18th c. 1 18th c. 1 18th c. 1 18th c. 1 18th c. 1 18th c. 2 18th c. 1 18th c. 1 18th c. 4 18th-19th c. 1 18th c. 1 1700-1830 2 1740-1765 1 1750-1820 1 1770-18105 1 1770-1820 1770-1820 1770-1820 1780-1820 1780-1820 1780-1830 1780-1830 1780-1830 1780-1830 1790-18105 1790-1820 1 1790-1820 1 1790-1840 1 1795,..1815 1 1795-1815 1 1795-1815 1 1795-1840 4 1795-1840 1 1795-1840 1 19th c. 1 19th c. 1 19th c. 1 19th c. 1 19th c. 1 18205-18305 2 1820-1840 1 1820-1900 1 1820-1900 1 1820-1900 1 1830-1850 2 1830-1870 1 1830-1870 1 1830-1870 2 1830-1870 1 1830-1920 1 1839-18705 1 1844-1860 1 1844-1860 1 1840-1900 1 1840-1900 1 1845 1 1845 TABLE 1 Inventory of Recovered Vessel Fragments, site 44DWI06

represented within the assemblage from excavation units adjacent to the west pavilion. They have been grouped according to temporal period and households as represented by the Banister, May, and Waring families. (These data are examined more fully below in light of the minimum vessel numbers and types recovered from the east pavilion units.) Banister Period (c.1770-1824). A minimum of 22 vessels are attributable to the period 1765-1824. These include underglazed blue Chinese porcelain plates, tea bowls, and cups; a molded white salt-glazed stoneware plate; undecorated and trailed-slip coarse earthenware hollowares; an undecorated creamware cup and a transfer-printed black holloware; an undecorated pearlware pudding pan, a pearlware shell-edged blue plate, a transfer-printed blue pearlware teabowl and mug, and a pearlware printed polychrome holloware; and a black basalt stoneware molded teapot. It is apparent that a variety of vessel types and forms are present, most representing table and tea wares. This ceramic assemblage, for the most part, consists of high-status, expensive wares, as evidenced by the Chinese porcelains, the transfer-printed wares, and the black basalt stoneware. The prevalence of these materials contrasts sharply with the sparse quantity of utilitarian wares, which are limited to two coarse earthenware hollowares. May Period (1824-1841). A minimum of 9 vessels are attributable to the period 1824-1841. These include a pearlware transfer-printed blue scalloped saucer, a pearlware-printed olive green plate, a pearlware hand-painted polychrome holloware, a pearlware-dipped embossed holloware, transfer-printed red whiteware plates, a transfer-printed brown whiteware plate, transfer-printed light blue whiteware cups, and undecorated brown stoneware hollowares. similar to the assemblage attributable to the Banister period, the vessels are predominantly tablewares. utilitarian wares, likely storage vessels represented by the stonewares, are more prevalent than in the earlier period. Chinese porcelain plates, teabowls, and cups have been replaced by transfer-printed wares. These wares, equally impressive for the period, were more expensive than undecorated wares of similar forms (Miller 1980). Waring Period (1841-1847). A minimum of 4 vessels are attributable to the period 1841-1847. These include an ironstone molded cup, a Bennington undecorated holloware, a whiteware flowblue saucer, and a whiteware flowblue platter. Relatively expensive, high-status tableware, these wares are comparable in price to the transfer-printed wares identified for 17

the preceding period. In some" instances, however, the flow blue decorated wares were more expensive than the transfer-printed ceramics, suggesting the elevated economic level of the household (Miller 1980). Architectural Remains and Yard Features Important architectural remains and yard features were identified within excavation units A, B, and C (Figure 5). These included builder's trenches, a stone underpinning beneath the foundation wall, possible scaffolding postholes, and traces of an early bulkhead entrance. Several feet north of the foundation were remnants of a brick spread, possibly the remains of a walk or service porch. An early drainage trench (Feature A-7) also extended into the yard from the northwest corner of the west pavilion (Figure 6). The most revealing architectural feature identified within these units was the remains of a trench (Feature B-14), located immediately adjacent to the north wall of the west pavilion. This feature, extending almost the entire length of the wall, measured approximately.4 of a foot wide at its top, sloping inward to as little as.2 of a foot. The trench measured 1.1 feet deep. The trench fill consisted of a brown sandy loam that sealed orange clay subsoil. within the fill were several large fragments of window glass, as well as fragments of brick, mortar, plaster, and nails. Associated with this architectural debris was a fragment of pearlware, indicating that the trench was dug and subsequently backfilled sometime after 1780. Initially, the trench was thought to be the builder's trench for the west pavilion foundation wall. Closer inspection, however, revealed that the feature was not related to the construction of the building but rather stabilization efforts during the late 18th or early 19th century. This interpretation is supported by evidence visible beneath the foundation wall, the depth of the wall, construction material in the cellar wall, and the types of material recovered from the trench. The foundation wall ended approximately 1.3 feet below modern grade and 1.1 feet below the stucco (see Figure 5). Tied into the north wall several tenths of a foot below the stucco and one foot above the bottom of the foundation was a brick ledge. This feature extended for approximately three feet to the east from the northwest corner of the structure. Although the purpose of the protruding brickwork is unclear (perhaps sloppy masonary work), the similarity of the brick in color and size to those present in the foundation suggests that the brick ledge was part of the original construction. 18

-.,... -...-:-" - -.:...:-':...;.:: -:- "." "i "".".. ;.,t.~ ~:-..'.,e, -.. '..?. ;;,~,. ~~ ~;.;. " '.... '... Excavation (Note Extent FIGURE 5 units A and B Adjacent to West Pavilion of Foundation Wall Below Ground Surface)

0 4 8 : I feet BOXWOOD IN I C-IO UNIT C Stone Underpinning BRICK LEDGE TIED INTO FOUNDATION FIGURE 6 Excavation units A, B, and C Adjacent to West Pavilion with Identified Features

At two locations beneath the foundation and east of the brick ledge were several large field stones, apparently placed as underpinning at structurally weak locations beneath the wall (see Figure 6). The shallow depth of the foundation wall and use of similar stone (in conjunction with brick) in the construction of the cellar wall suggests that the construction of the cellar and efforts to stabilize the foundation were undertaken simultaneously or that the cellar walls were repaired with stone at the time of underpinning. The latter of these interpretations is more plausible based on the age of the artifacts found within the underpinning trench and the architectural and historical evidence. The materials recovered from the trench fill--window glass, nails, brick, plaster, and pearlware--were all found in direct association with the stone underpinning in excavation unit B. The broken window glass and the plaster were more typical of destruction debris as opposed to construction refuse found in an original builder's trench. Consequently, this material was most likely sheet refuse redeposited in the trench as back fill at the time of underpinning the foundation. The pearlware dates the underpinning to post-1780. Several fragments of window glass found in association with the pearlware appear to be 19th-century, most likely dating the stabilization efforts to the first quarter of the 19th century. This work was probably undertaken by John May soon after his occupation of the property in 1824. Stabilization of the foundation during or prior to cellar expansion may have been deemed necessary given the shallow depth of the foundation and the age of the structure. Intruding the top of the underpinning trench was a second trench (Features A-5, B-5, and C-6). Only traces of this feature were present, but its dark grayish brown sandy loam fill distinguished it from the earlier feature. The trench, measuring approximately.2 feet in width, was exceedingly shallow, bottoming out approximately.2 feet beneath the stucco. The trench appears to have been dug for the purpose of stuccoing the house or possibly repairing the existing stucco. The shallow feature allowed the covering to be extended slightly below ground, out of view. contained within the trench fill was a light scatter of coal, nail fragments, window glass, and stucco. While no diagnostic material was recovered, several fragments of the window glass appeared to be 20th-century. In addition, the feature cut Layers A-2, B-2, and C-2, which all dated to the first quarter of the 20th century. This may reflect stucco repair by Perkinson during the early decades that century. Careful attention was paid to other architectural remains identified within excavation units immediately adjacent to the north side of the west pavilion. One of the more interesting of 21

these was the extant bulkhead entrance, located approximately- 13 feet east of the northwest corner of the structure (see Figure 6). While the superstructure for the entrance appeared to have been rebuilt within the past 20 years, the foundation may have been original to the cellar. Immediately adj acent to the bulkhead in units Band C, excavation revealed substantial remains of a builder I s trench (Features B-4 and C-10) (Figure 7). This feature, measuring 1.0 foot in width and depth, consisted of a dark brown sandy loam mottled with lenses of orange clay. The trench fill of Feature B 4 contained a concentration of architectural and domestic refuse, including wire nails, brick, mortar, American stoneware, porcelain, creamware, pearlware, bone, and table and bottle glass. Although Feature C-10, located on the east side of the bulkhead, was disturbed by a later pipe trench it yielded similar refuse. Much _of this material (specifically the bottle glass) dates the construction of the bulkhead to the early part of the 20th century. This interpretation is supported by the age of the stone and brick used in the construction of the bulkhead foundation. The bricks appeared to be machine-made, as opposed to the hand-made brick present in the building's foundation. One of the stones in the wall had a drill hole exposed in section, indicating that it had been quarried by blasting with dynamite. This quarrying method, not in use until the late 19th century, further suggests an early 20th-century construction date. Although archaeological evidence indicates that the entrance was not original to the cellar, traces of an earlier bulkhead were identified (see Figure 6). These remains were located on the immediate west side of the extant entrance and 12.1 feet east of the northwest corner of the west pavilion. Sealed beneath and partially intruded by Feature B-4, the trench-like Feature B-9 measured approximately.7 of a foot wide and 1.3 feet deep. The fill consisted of a brown sandy clay loam. While no diagnostic material was recovered, the feature did contain a concentration of hand-made brick and a light scatter of window glass and coal. The south section of trench revealed a continuation of the pavilion wall foundation to a depth of 1.3 foot below the existing brickwork (Figure 8). While little remains of this feature (due to subsequent bulkhead construction), it does indicate the location of an earlier bulkhead, if not the original entrance to the cellar. Located approximately 9 feet northwest of this feature were the remains of two of the three yard features identified within the excavation area (Features B-11 and B-12). The larger of the two features, B-11, was a rectangular shaped posthole measuring 2.3 by 1.2 feet and 1.3 feet deep. While no postmold could be clearly defined within the feature, the posthole fill could be easily delineated. It consisted of a brown sandy clay loam mottled with 22

FIGURE 7 Excavation unit C Adjacent to West Pavilion (Note Trench Adjacent to Rebuilt Bulkhead Entrance)

.. >:'" <, ~" at ~:~ ':"~''if~, _..;i. FIGURE 8 Remnants of Original Bulkhead Entrance in Excavation unit B

orange clay. Packed within the hole were whole brick and brick bats, possibly used to secure a post. In addition to this material, several fragments of creamware, pearlware, and window glass were recovered. The presence of pearlware in the posthole fill dates the feature to post-1780. The posthole cut into a circular posthole (Feature B-12) to the west. This earlier feature measured approximately 1 foot in diameter and 1.4 feet in depth. The posthole fill, consisting of a brown sandy loam mottled with orange clay, contained a variety of 18th-century ceramics, including a black basaltes teapot lid fragment, Chinese porcelain, and pearlware. In addi'tion, several fragments of bottle and window glass were present. The pearlware within this assemblage dates the feature to post-1780, most likely the last two decades of the century. Unlike Feature B-11, the posthole contained remnants of a postmold. This feature, F-13, consisted of a grayish brown sandy loam and was.5 feet square. Materials recovered from this feature were limited to fragments of brick. The limits of the excavation area did not permit a posthole series to be identified within the yard area. While it is possible that the postholes found were part of a fenceline extending to the north and enclosing a kitchen yard, their location several feet east of the corner of the building seems somewhat odd. Their location close to the building and several feet from its corner suggests that the features may have served a different purpose, perhaps to secure scaffolding posts. The features' post-1780 date suggests the scaffolding would have been for restoration or repair work on the west pavilion, as opposed to its initial construction. The function of the posts could be identified only by attempting to locate extensions within the yard. The most substantial of the yard features identified within the excavation area were remnants of a dry-laid brick walk or service porch. While only a small portion of this feature extended into excavation unit B, probing indicated that it extended several feet to the north. The feature was contained within Layer B-2 and sealed by topsoil. Although materials recovered from sealant layer indicated that the feature went out of use during the 20th century, a fragment of brown stoneware from beneath the brick suggest that it was constructed during the 19th century. East Pavilion: Excavation units D through H Cultural Layers Excavation units adjacent to the east and northeast walls of the east pavilion contained several cultural layers, including late 25

fill deposits associated the bathroom addition. Considerable disturbance resulting from the construction of the bathroom and subsequent trenching for water and sewer lines has eliminated intact layers in portions of several of the excavation units. Nevertheless, layers with limited disturbance are present. similar to the uppermost layer identified in the western excavation units, the topsoil layer consisted of a dark brown sandy loam (Layers D-l, F-l, G-l, and H-l (Figure 9). This deposit, measuring.4 to.8 feet thick, contained numerous fragments of modern bottle glass and plastic. Prehistoric material consisted of a light scatter of lithic debitage and flakes. The presence of orange clay mottling throughout this layer indicates disturbance, possibly through planting activity. NORTH I UNIT DE~ARCATION ~ SOUTH --1------- ---- :---------~~~ G-I F-I F-3 o I I feet 2 I FIGURE 9 Excavation Units F and G, East Wall Profile Beneath this late 19th-/20th-century layer was brown sandy loam (Layers F-2, G-2, H-7), which ranged in thickness from.2 to.7 feet thick. This layer contained a sizeable quantity of material, including fragments of pearlware, creamware, transferprinted whiteware, window glass, cut nails, and brick. The presence of whiteware dates the layer to post-1830. Sealed beneath this layer was a reddish brown sandy loam (Layer H-2) (Figure 10). The layer, located approximately.7 feet below modern grade and measuring up to.5 of a foot thick, contained a scatter of architectural and domestic refuse. This domestic assemblage included pearlware and bottle glass. In addition, a concentration of architectural debris was recovered from the layer adjacent to the building foundation in excavation unit H. This small unit yielded 54 fragments of window glass, several nails, lead scraps, and brick fragments. The presence of pearlware within the assemblage dates the accumulation of the layer to sometime after 1780, most likely during the first quarter of the 19th century. 26

WEST I 'V.. H-2 17, o I ------- ~ --. ; EAST I H-I H-7 - '--'f' \. ~ H-3 '. \. / BRICK AND MORTAR FRAGMENTS 2 I feet FIGURE 10 Excavation unit H, North Wall Profile This material; specifically the architectural debris, is most likely attributable to John May's occupation of the property at the end of the first quarter of the 19th century. Architectural data suggest that May made extensive alterations to doors, windows, and porches on the structure, including the east pavilion. The concentrated architectural debris in close proximity to the structure may be the material remains of this activity. Sealed beneath this layer and overlaying sterile subsoil was a brown sandy loam (Layers H-3, F-3, and G-3), measuring.3 to.5 feet thick. Notably absent from the layer was the architectural debris found in layer Layer H-2. Although this layer contained fragments of pearlware, the assemblage suggests late 18th-century deposition. In addition to the pearlware, window glass, wrought nails, and a rim fragment of molded white salt-glazed stoneware plate were recovered. Sheet Refuse Deposits Sheet refuse deposits identified within excavation units adjacent to the east pavilion were noticeably more sparse in the presence of certain artifact types than in the scatters located to the west. Of the 1,110 artifacts recovered from this area, 90 ceramic fragments were found, representing only 8 percent of the total artifact assemblage. Table glass accounted for less than 1 percent; no bone was recovered. Architectural debris made up 78 percent of the assemblage. This heavy concentration attests to episodes of extensive repair and remodeling to the east pavilion during the 19th century. Some of this ~aterial, including mortar, brick, hand-wrought and cut nails, and window glass, was recovered from refuse scatters immediately adjacent to the foundation wall. As discussed above, 27

excavation unit H contained a concentration of such material, dating to the first quarter of the 19th century. While architectural debris often allows for the documentation of periods of structural alterations to a dwelling, domestic refuse can provide data useful in addressing a number of questions pertaining to individual households. This is especially beneficial when diagnostic material such as ceramics are analyzed in terms of minimal vessel counts, types of vessels, vessel forms, and recovery from different areas of a the site. As with the ceramics recovered from deposits identified adjacent to the west pavilion, ceramic fragments in excavation units D through G have been analyzed in terms of the vessels they represented. These vessels, in turn, have been grouped temporally and attributed to the Banister, May, or Waring households. Banister Period (ca. 1770-1824). A minimum of 7 vessels is attributable to the period 1765-1824. These include a Chinese porcelain underglazed blue bowl, a Chinese porcelain undecorated tea bowl, a pearlware shell-edged green plate, pearlware transferprinted blue plates, a white salt-glazed molded stoneware plate, and a brown stoneware undecorated holloware. similar to vessels of this period found in the western yard, this assemblage represents quite a broad range of relatively expensive tableware, including decorated porcelains and pearlwares. Perhaps most striking is the decrease in the minimum number of vessels present--only one-third of the number recovered in the western yard. May Period (1824-1841). This period is characterized by a minimum of 6 vessels. Included among these are a whiteware transfer-printed brown plate, a whiteware transfer-printed brown saucer, an Albany slip stoneware undecorated jar/pot, an Albany slip cobalt decorated jar/pot, a brown stoneware undecorated bottle, and a yelloware undecorated unidentifiable vessel. This assemblage is distinguished from those in the earlier period by the absence of Chinese porcelains and the increased presence of utilitarian wares. A similar quantity of these later (i.e., storage and food preparation-related) vessels from this period was recovered from the western yard. Transfer-printed vessels in whiteware are present but number less than half of those recovered in the western yard. Waring Period (1841-1847). A attributable to the period 1841-1847. flow blue plate, a whiteware flow blue molded bowl. minimum of 3 vessels is These include a whiteware holloware, and an ironstone 28

These ceramic types and vessel forms are similar to the period vessels identified in the western area. While their relative costs have provided some indication of the economic status of the waring household in relation to its predecessors, the minimum number of vessels present sheds light on refuse disposal patterns of these households over nearly a century. Architectural Remains and Yard Features Several significant architectural features were identified within excavation units immediately adjacent to the foundations of the east pavilion (Figures 11 and 12). Excavation unit H, for example, revealed a builder's trench (Feature H-4) along the north wall of the structure. This feature, measuring approximately 4 feet in length,.3 of a foot in width, and.8 of a foot in depth, consisted of a brown sandy loam. Careful excavation of the feature produced a light scatter of brick fragments and brick bits. No other material was recovered. The character of this feature, including the consistency of its fill and its width and depth, suggests that it was a builder's trench. Unlike the underpinning trench (Feature B-14) identified along the west pavilion wall, the bottom of the feature revealed no evidence of stone beneath the brick foundation. The bottom of the foundation, identified approximately 1.3 feet below modern grade, suggests that the cellar was dug after the building was constructed. Although there was no evidence of repair or stabilization work to the east pavilion foundation, remains of a possible waterproofing trench were identified along the east foundation wall (Figure 13). This feature, extending parallel to the wall, measured 1.4 feet in width and 1.5 feet in depth. The bottom of the trench continued approximately 1.2 feet beneath the foundation. The trench fill contained two deposits, the uppermost consisting of a culturally sterile orange sandy clay (Layer E-10) while the bottom layer was characterized by a light brown sandy loam (Layer E-11) (Figures 14 and 15). Initially, the clay fill deposit, intruding the brown sandy loam, was thought to be contained within a separate trench that post-dated the brown loam feature. Careful inspection of the feature, however, indicated that the two deposits were probably contemporary. The clay appears to have been intentionally packed within the brown sandy loam along the upper part of the trench immediately adjacent to the foundation wall. The loam contained a variety of domestic and architectural refuse, including creamware, pearlware, porcelain, Albany slip stoneware, brown stoneware, ironstone, furniture hardware, mirror glass, nails, spikes, and window glass. The presence of the 11 29

FIGURE 11 Excavation units 0 Through H, Northeast Corner of East Pavilion

f f f- I I H-4 MOOERN PIPE ' TRENCH' {i, I r-----~,'1 tv;;;, U~~5 ~8 j'1~\ I r-- SULK- UNEXCAVATED I II ~,~o ~L_~;: _ r------, - - - I-;;E-IO I I UNIT E I I UNIT G '~E-II., "(}l E-15 I I ; 'L;:::J, i 'E-14 I I I--"----,.)- - - - f-i---,---- f-./- I UNEXCAVATED LATE 19TH/ EARLY 20TH CENTURY BATHROOM ADDiTION a 4 feet FIGURE 12 Excavation units D Through H, Northeast Corner of East Pavilion

FIGURE 13 Remains of possible Waterproofing Trench (E-10, E-ll) Adjacent to the East Pavilion Foundation Wall