Colonial Williamsburg Digital Library

Welcome to Nassau Street

In the fall of 1940 Colonial Williamsburg architectural draftsman James M. Knight and his team of excavators encountered an unexpected foundation at the corner of Duke of Gloucester and Nassau Streets (Figure 1). Guided by the Frenchman's ca. 1782 map and other eighteenth- and nineteenth-century documents, Knight and his men had been hunting down the foundations of eighteenth-century buildings as part of Colonial Williamsburg's reconstruction. The trenching and complete excavation of Block 14's brick foundations yielded, among other finds, the Bryan House, which stood on Duke of Gloucester Street between the mid-eighteenth century and the end of the nineteenth century. Knight also located and examined the Bryan House's four smaller dependencies. But the structure located half in the Bryan house yard and half under Nassau Street did not fit the profile of an eighteenth-century Williamsburg building.

For one thing, at least half of the cellar's foundation would have stood in the middle of Nassau Street-a street laid out during the town's earliest days. What's more, the foundation sat off the town grid with its west end farther south of Duke of Gloucester Street than its eastern end. The 1940 excavations of the western side of the cellar's rubble fill disclosed a large Figure 1. Location of excavation. 2 number of lead window cames which contained seventeenth-century manufacture dates. Excavators uncovered the nearly intact remains of a seventeenth-century casement-two windows laying flat on the floor (Figure 2). Further excavations in 1942 added more evidence supporting Knight's conclusion that this anomalous building was a seventeenth-century structure dating to Middle Plantation, the community existing before Williamsburg's 1699 founding.

In the decades since Knight's Nassau Street find Chesapeake archaeologists have examined the remains of dozens of seventeenth-century buildings and homelots on sites located all over the region. Consequently archaeological understanding of the material remains of seventeenth-century settlement and the sophistication of possible research questions has grown along with the size of the data base (Brown 1998; Muraca and Levy 2000; Pickett 1996).

In 1999 the town of Williamsburg celebrated the three hundredth anniversary of its founding. In 1699 Governor Francis Nicholson and members of his council presided over the relocation of Virginia's colonial capital from the blighted marshes of Jamestown to the high grounds of what was then known as Middle Plantation. From its first settlement in the 1630s this portion of the colony, situated midway between the York and the James Rivers, had been home to several large planters, including several members of the governor's council. By the time of the capital's relocation Middle Plantation had grown into a prosperous and well peopled rural crossroads which, except for the predominance of tobacco agriculture and the presence of almost one hundred African slaves, strongly resembled the English countryside. The creation of Williamsburg-the new colonial capital-transformed this landscape of flat terraces, deeply cut ravines, and winding roads into a rectilinear cityscape (Billings 1986; Coombs 1999).

As part of the town- and museum-wide commemoration of Williamsburg's founding, the Colonial Williamsburg Department of Archaeological Research wanted to conduct a public excavation of a site dating to Figure 2. Casement windows found on floor of Nassau Street cellar (from Noel Hume 1963:78). 3 the period of the city's founding. The seventeenth-century brick structure Knight located in the 1940s was the natural choice. Its Nassau Street location placed it right against Duke of Gloucester Street, and all previous research on the building suggested that it dated to the time of the capital's big move. In late May-just as the museum's visitor summer season began-College of William & Mary students and Colonial Williamsburg archaeologists began a comprehensive examination of the building underneath Nassau Street. Funding for this project came from the City of Williamsburg, the Colonial Williamsburg Foundation, and a generous donation from Joan Key from Glen Ellyn, Illinois.

Excavation Background, Goals, and Questions

The summer 1999 excavation at Nassau Street picked up where Knight and his team left off, while bringing current archaeological thought, techniques, and technology to bear on the structure (Figure 3). The project had three principal goals. The first was to provide Colonial Williamsburg visitors with a chance to watch a modern archeological excavation in progress. Colonial Williamsburg amassed a staff of well-trained volunteers to field questions from the hundreds, if not thousands, of anticipated visitors. Two costumed interpreters portraying Governor Nicholson and the Reverend James Blair-both key figures in the movement of the capital-conducted walking tours of Middle Plantation sites and acted as living connectors between the Nassau Street dig and the Dewitt Wallace Gallery's Figure 3. Excavation areas at Nassau Street site. 4 three hundredth anniversary exhibit "1699: When Virginia Was the Wild West" (Hill 1999).

As part of the site's public mission project archaeologists provided a "site diary" on Colonial Williamsburg's web site at www.history.org. This web site contained digital photography chronicling the excavation's progress and featured regular updates on the dig and on the methods and principles the staff employed.

The project's second goal was to understand the lifespan of the structure. The dates of the building held the key to its largely mysterious ownership. Traditional wisdom holds that this building was Francis Page's Middle Plantation home, which was leased out as an ordinary after his 1692 death. Francis Page was the son of Colonel John Page, a prominent local land owner and merchant. The latter used much of his capital and position to acquire land, and owned several tracts concentrated in the Middle Plantation area. In 1662 the elder Page built an impressive cruciform brick home on Bruton Heights overlooking his land, much of which would become Williamsburg in 1699 (McCartney 1999; Metz et al. 1998).

Francis Page was the elder son of John and his wife Alice. Born in 1657, Francis schooled in England like many a well-off Virginia-born Englishman. By 1680 Page was back in the colony and had followed his father into a career of land- and officeholding. In 1680 Page served as York County's Justice of the Peace and also as a member of the Bruton Parish vestry. In 1684 he entered the House of Burgesses, where he served from York County until 1691. From 1686 to 1691 he was the Clerk of the Assembly.

In 1679 John and Alice Page granted their eldest a 168-acre tract of Middle Plantation land, comprising three tracts of land John bought in the early 1670s. All three of the tracts sat on the western side of the 1634 palisade. Soon afterward Francis Page began his own family by marrying Mary Digges in 1682. Mary soon gave birth to their only child, a daughter named Elizabeth. In 1692 John Page died, leaving his holdings and home to his wife Alice and naming Francis as his reversionary heir. But only a few months after his father's death, Francis himself took ill and died, leaving his holdings to his daughter (in the care of the widow Alice Page) and to his brother Mathew Page. Francis requested that his home, brick barn and brick malt house near the road in Middle Plantation, and all his other buildings be rented out to tenants until Elizabeth reached maturity. It seems that she never did, dying a few years after her father's demise (McCartney 1999).

In August 1695 one Poynes Weldon was listed as running a tavern in the "house lately belonging to Fra. Page and now in the possession of Captain Mathew Page" (York County Deeds 6:128, 9:126-127). The same building had been occupied before Weldon by one Thomas Taylor (York County Deeds 18:189).

The story of Page's home-cum-ordinary is a well documented one, but there was in fact very little to tie the documentary story to the physical structure underneath Nassau Street. Furthermore, local scholars have used reconstructed property lines to argue that the Nassau Street building sat to the east of the palisade, placing it off Francis Page's holdings and therefore not his home.

The ambiguity surrounding the structure and the lot's ownership stem from its actual physical location in James City County, a county with few surviving colonial records. Several colonial scholars have meticulously reconstructed James City County's property lines and titles, but with varying results. The key to understanding the location of Middle Plantation properties is the location of the 1634 palisade, which probably ran right through the area and served as a property line reference long after its pales had rotted and fallen over. Using surviving deeds and maps, local topography, and surface features, scholars have offered several possible locations for the palisade. With each new location comes a new set of property lines. Excavations at Bruton Heights in 1994 located a 2500-foot-long portion of the palisade, identified by its depth, size and artifacts. But this roughly east-west stretch did not offer clues as to the location of the north-south stretch-the stretch which most directly effects the Nassau Street property. Seventeenth-century patent maps suggest that the north-south stretch of the "pales" ran somewhere between Nassau and Henry Streets, but an error of a few yards can mean very real changes for the supposed owner of a given building. In short the surviving documentation offers little clear cut and authoritative evidence of who may have owned the Nassau Street building (Metz et al. 1998; Muraca and Brudwig 1993).

At the time the dig commenced there were three competing possibilities for the structure's ownership. The first was that it was Francis Page's home. The brick foundation suggested an owner of some means, and the fact that the Nassau Street building seemed to be close to John Page's land make Francis Page a strong contender. But two bits of evidence worked against Page's ownership. The first is that the building is in James City County and not York County, Page's identified county of residence. This situation is further confused by the fact that this building sits virtually atop the modern James City-York county line, making its seventeenth-century county location up for grabs. The second argument against Page's ownership is that Page in his will stated that he owned both a brick barn and a brick malt house near the Horsepath in Middle Plantation. Although Knight's team located the remains of a brick building they found no hint of other brick buildings in the immediate area. Research suggests that seventeenth-century plantations buildings generally clustered fairly close together. If this was Francis Page's Middle Plantation home, where were the two documented outbuildings?

The second possible owner was actually a sequence of owners. In 1654 Francis Peale patented a 50-acre parcel of land along the west side of the palisade. In 1662 Robert Weeks repatented the same tract and by the 1690s the land appears to have become part of the large Bray family holding. The Peale-Weeks connection offered the tantalizing possibility of studying the well-constructed home of men who were not high level colonial elites like the Pages. But it is also unlikely that men as obscure as Peale and Weeks would have had the resources to construct a home as large and imposing as the one mapped by Knight.

The third possibility is that this building was part of the Bray family's many holdings. The Brays owned a considerable amount of land in Middle Plantation and much of that land became the town of Williamsburg in 1699. James and David 6 Bray were both active in the last quarter of the seventeenth century and were contemporaries of Francis Page.

Establishing the dates of the building was the best way to begin to establish its owner. Establishing dates for the structure's raising and razing places it within the life span of one of the possible owners, while locating other features might provide clues to ownership, occupancy dates, and activities. This strategy may not be able to use a non-personalized domestic assemblage to delineate the elite Pages from their elite neighbors the Brays. But the structure's dates would be able to test the possibility of a Peale-Weeks connection, while the presence or absence of tavern trash deposits would test the Francis Page possibility. This way the structure's dates not only suggest who the owner was, but also serve as a test for the validity of the reconstructed property lines. The dates of artifacts recovered during the excavation of the cellar's rubble fill provided dates for the building's demolition while artifactual data for the builders' trenches provided dates for the building's construction.

The project's third goal was to define what the area immediately surrounding the building looked like. During the seventeenth century Virginia went through several re-conceptualizations. From company compound to private fiefdoms, to royal colony, to slave society, the economic and social order of the colony changed, and with it the landscape. The organization of Virginia homelots-the shapes of yards, placement of outbuildings, and divisions of use spaces-are both formative and reflective of these changes in colonial Virginians' ways of life. In order to delineate the surrounding landscape excavators examined portions of the Bryan House yard as well as an area beneath Nassau Street extending about 100 feet to the south of the building's foundations.

Pre-1999 Excavations at Nassau Street

In 1942 Knight secured permission from the city to close traffic on Nassau Street and conduct a short-term excavation of the east side of the building which was not examined in 1940. Whereas the 1940 excavation (Figure 4) consisted of the complete removal of all of the structure's soils and rubble within the Block 14 bounds, the 1942 investigation consisted of trenching around the foundations wall in order to create a complete plan of the structure (Knight 1942; Figure 5).

Knight's excavations revealed a spread-foot brick foundation measuring 43' 7" east-west and 19' 4" north to south. The remaining walls had been badly robbed: in some places only the footer course survived while in others as many as nine courses remained. The complete bricks measured 9" x 4" x 2 ½" and were held in what Knight identified as English bond with shell mortar. The walls measured from 18 ½" to 19", with the spread-footers measuring 22". The foundation had "Dutch" in-set hearths at each gable end and the east side hearth had a brick pad to its north that Knight misidentified Figure 4. James Knight's completed work in 1940 on the west half of Block 14's seventeenth-century building. The Bryan House foundations and Duke of Gloucester Street are in the background, Nassau Street to the right. 8 Figure 5. James Knight's 1942 north wall trenching episode in Nassau Street. The newly constructed Bryan House sits in the background. as part of a stairway. The west side of the structure was fully paved with brick pavers laid in a single herringbone pattern (Figure 6). The only deviation from this pattern was in the area in and around the west hearth, which was covered by pavers laid in a staggered pattern. The paved floor extended to the midpoint of the building where it ended-the east side had a dirt floor. Figure 6. James Knight's plan of the Nassau Street foundation. Image is rotated so north is at top. 9 Knight found no evidence of brick removal in the east half and his conclusion that only half the cellar was paved is supported by the fact that the terminating edge of the pavers is marked by brick bats. These suggest that the pattern ended by design rather than having been removed. Knight only saw a small portion of the terminus of the pavers as they appeared only in his 1942 trenching areas and not in his 1940 full excavation.

Knight recovered few artifacts from the cellar's fill. His finds included clay pipe fragments, some earthenware sherds, and many pieces of "broken ale bottles" which are now identified as wine bottle. Knight also noted that wood ash covered the flooring of the eastside hearth.

The 1942 excavations left intact a large block of rubble fill in the center of the eastside of the structure. Knight also did little damage to the builder's trenches on the structure's east side and made no mention of soil features associated with the building.

The next time excavators looked at the structure was in 1983, when city utility crews put in a new sewer line main down the length of Nassau Street. In order to protect the foundation remains, Colonial Williamsburg archaeologists removed a portion of Nassau Street's asphalt to be sure that the new pipe missed the structure. They mapped the surface level plan of the brick wall and identified part of the south-side builder's trench as well as some other modern utility trenches. The excavation crew conducted no further excavation of the site.

Methods Then and Now

Knight and his team used a fairly straightforward method which well suited their straightforward goals. Knight approached Williamsburg's Historic Area block by block or lot by lot and trenched each one at a 45 degree angle to the town grid. Knight's crew dug their trenches one shovel blade width thick and one shovel length apart. They generally dug to the subsoil although in some areas they stopped at the top of large features. The principal goal was to locate the remains of brick foundations. Once the crew hit a wall they then followed its length to locate corners and other walls. Once a building had been completely exposed, mapped, and photographed the team then returned to trenching.

These methods were well suited for the job at hand-finding brick foundations quickly and thoroughly. But times change, and the methods and goals of historical archaeology are now drastically different. The 1999 excavations used the current Colonial Williamsburg method, based on "open-area" techniques developed in Britain and elsewhere. Before the digging began excavators laid a metric grid over the whole site. Each excavation unit measured one meter square, and each layer, cut, and fill received its own unique "context" number. Excavators assigned "master context" numbers for contexts that spanned units such as fill layers in the cellar or for event features or layers that spanned the whole site such as the remains of Knight's excavation trenches.

Figure 7. Brick-by-brick drawing of the seventeenth-century building.

Excavators mapped all located features and drew profiles for all those features they excavated. The team created two sets of maps-one made feature by feature and a second one made on larger mapping paper which captured the plan relationships between features. During excavation the crew took chemical and phytolith samples from every layer and fill. Excavators also removed some layers for flotation. All non-saved soils went through a ¼-inch screen with exception of one portion of the builder's trench fill which excavators ran through a water screen in order to improve the rate of artifact recovery.

The cellar fill was the only intact feature on the site to yield any significant quantity of animal bones. Unfortunately the number of recovered faunal remains was too low to enable any more than the most basic of examinations. The bones nevertheless were counted and identified.

The Cellar Excavations

The location of the remains of the cellar underneath Nassau Street with its wide-open location allowed visitors to see the excavation from all sides, and the fact that it represented the largest concentration of seventeenth- and eighteenth-century deposits made it the excavation's centerpiece for both visitors and archaeologists (Figure 7, right). But before cutting into the previously unexcavated cellar fill the crew first removed the 1940 backfill covering the portion of the cellar that lay in the Bryan House yard to the west of the Nassau Street sidewalk (Figure 7, left; see also Figure 3 for exact excavation 11 boundaries). Since its backfilling numerous small utility lines and one huge sewer pipe cut the cellar and the 1940 fill. Excavators removed this fill, exposing the pipes and utility lines, for two reasons. The first was to locate any builder's trench remains for future excavation, and the second was to expose as much of the floor plan as possible for overhead photography. Knight's excavation left no trace of what must have been a fairly small builder's trench, but archaeologists did take several overhead photographs after completing the cellar excavations.

The Nassau Street sidewalk covered a large previously unexcavated portion of the cellar's fill, but to the sidewalk's east sat the mostly untouched road-covered fill. Knight's 1942 excavation in the road chased the structure's walls but left an untouched 420 by 301 cm block of rubble in the center of the cellar. The remains of a tree stump did not allow the excavation of the southwestern corner of the block. This block of rubble fill extended between units 1144N/498E and 1141N/502E, but the fill did not extend to the cellar walls as these had been trenched by James Knight in 1942.

After removing the 1942 backfill excavators divided the remaining untouched block into quadrants roughly measuring 210 by 160 cm and began the stratigraphic removal of the rubble fill, taking chemical and phytolith soil samples from each layer. Five "master context" layers spanned all four of the excavation quadrants and several smaller localized layers appeared in each of the quadrants (Figure 8). Although the composition of these master layers was fairly consistent across the whole cellar excavation, their depths varied dramatically both within quadrants and across the whole cellar block.

MC-13 was the first layer and consisted primarily of mortar rubble with inclusions of small brick fragments. This layer measured between 10 and 41 cm with its deepest portion in context 14BC-266 in the northwest quadrant. In the southeast quadrant MC-13 sealed an 11-cm-deep layer of mixed brick bats and mortar rubble. MC-14 sat beneath MC-13 and 14BC-266. This master layer was a substantial concentration of brick rubble composed of whole and broken brick as well as small portions of intact brick wall. MC-14 measured between a shallow 15 cm in the southeast quadrant and 32 cm in the northeast quadrant. In the southeast quadrant excavators encountered an intact chunk of the structure's brick wall (context 14BC-268). This fragment measured 59 cm long, 23 cm wide, and 20 cm deep. Context 14BC-268 showed evidence of having been plastered on three of its sides, suggesting that it may have part of an internal partition wall from a story above the cellar (Figure 9). Excavators and conservators removed the fragile wall intact for further analysis.

In the southeast quadrant sat a 90 by 70 cm pocket of light brown soil which was 10 cm deep. The light color of the fill and its lack or artifacts suggested that this was redeposited subsoil. Beneath this layer and MC-14 sat MC-15, a spread of mortar and brick rubble which measured between 10 cm in the northwest quadrant and 18 cm in the southwest quadrant. This master layer sealed another portion of collapsed brick wall-14BC-298-which sat in the northwest quadrant and measured 115 by 100 cm. Unlike 12 Figure 8. Harris matrices with context numbers and soil descriptions of various layers comprising the cellar's fill. 14BC-268, 14BC-298 was not composed of mortared bricks. Instead, these bricks lay flat atop the supporting layer in a pattern reflecting the three courses of wall they once were. These bricks, although disturbed and damaged during the structure's demolition, appear to have been part of sheathing that was not fully tied into the rest of the wall.

MC-16 was a brick rubble layer that sat beneath MC-15, 14BC-298, and a 10-cm brick layer (14BC-307) and a subsequent 7-cm mortar layer (14BC-308) in the southeast quadrant. MC-16 measured between 12 cm deep in the southwest quadrant and 23 cm deep in the northeast quadrant. This layer consisted of loose brick rubble with chunks of mortar throughout. A large number of complete bricks came out of this layer. Beneath MC-16 the remaining layers became thinner and more complex.

Figure 9. Close up of intact brick wall chunk 14BC-268. Note the plaster visible at the outer edges.

A 1 to 3 cm thin layer of soil with heavy ash concentrations-14BC-314-covered most of the southeast quadrant. This ash was of a bluish-gray color and showed numerous charcoal and coal inclusions. This layer had a roundish plan shape and was a thin spread through which were visible patches of lower layers of plaster and mortar. In the northeast quadrant MC-16 covered a 12-cm-thick layer of mortar rubble (14BC-356), which in turn covered a 110 by 100 cm spread of collapsed wall (14BC-365). Like 14BC-298, this portion of brick wall appeared to be a collapsed portion of a wall's sheathing. These bricks were not mortared into place-instead they lay where they fell. A 3-cm layer of mortar rubble covered the southwest quadrant beneath MC-16. This layer in turn covered a 6-cm-thin layer of scattered brick rubble which sealed another brick wall portion (14BC-391) (Figure 10). This wall was by the far the largest and best preserved of all the collapsed wall layers. It covered almost all of the southwest quadrant, measuring 187 by 144 cm and was 11 cm deep. The fine condition of its bricks offered the best glimpse of how the wall had been constructed. All the other portions of brick wall were spreads of no-longer-mortared brick. But the bricks of 14BC-391 were still mortared together and only its size and obvious fragility kept excavators from attempting to remove it intact. Instead excavators left this wall in situ after recording and photographing it.

Beneath all this sat a layer of broken plaster, MC-17, measuring 1 cm. This plaster was not a solid covering but rather a loose scatter spread evenly over the entire excavation block. The plaster faced downward, suggesting that it resulted from the collapse of the wall which became the rubble and brick chunks above the plaster. Excavators retrieved numerous plaster 14 Figure 10. Collapsed portion of the building's north wall, 14BC-391. The capped pipe is for the Bryan House's oil tank. samples from this layer. Removal of the plaster layer revealed the full extent of the layers visible where no plaster lay. In the northwest quadrant a 2 to 5 cm layer of artifact-rich loamy soil (14BC-302) covered the unit. This layer contained numerous pipe fragments and considerable charcoal inclusions. The northeast and southeast quadrants were partially covered by a roughly circular layer of ash and charcoal. This layer (14BC-386) measured 160 by 216 cm and was 2 cm deep (Figure 11). Ash and charcoal also appeared on the floor of the eastern hearth. Excavators removed ten liters of the hearth ash fill and 100% of the ash layer 14BC-386 for flotation. Knight's team exposed and recorded the hearth ash in 1942 and it was consequently somewhat Figure 11. Plan of ash layer and mortar spread. 15 contaminated, whereas the excavation block ash layers were more intact and offered the prospect of a better glimpse of the ash associated with the building's demise. The eastside hearth also contained two burned spots. These features, 14BC-387 and 14BC-388, appeared to be burns associated with the use of the hearth and were of a hard burned clay consistency. Neither were excavated. Context 14BC-387 measured 17 by 16 cm, and 14BC-388 measured 25 by 23 cm.

A layer of soil and ash surrounded the ash layer and extended into and covered the rest of the excavation block's northeast and southwest quadrants. This layer-MC-18-exhibited considerable brick, mortar, and shell inclusions and was 3 cm deep. The removal of the full expanse of the soil/ash and ash layers exposed the cellar's hard subsoil floor (14BC-404) and three large deposits of mortar (14BC-395). Although these deposits were not contiguous, they were clearly of the same fill and created at the same time. Context 14BC-395 was 4 cm deep and was composed of granular mortar with shell, brick, and charcoal inclusions. Digging the deposits tuned up a large number of artifacts including numerous pipe fragments and several mending pieces of a Challis milk pan. Interspersed throughout the mortar layer were small pockets of sandy silt (14BC-394). These contained no artifacts and were not in regular shapes. Instead they were small localized deposits set in the mortar and were between 1 and 3 cm deep.

Layer 14BC-404 was the hard subsoil floor of the cellar. The western portion of the cellar was covered by brick pavers laid in a double herringbone pattern, but Knight's 1942 dig showed that these did not cover the whole floor. The 1999 excavations revealed that there never was a brick floor on the east side of the cellar. The level of the floor (even with the bottom of the brick walls), the presence of pressed-in sheet refuse, the lack of robbing evidence, and the presence of burned areas in the east hearth all suggest that this half of the cellar had a soil floor while the west had a paved one. This division in turn suggests that at one time there was a partition wall separating the two rooms.

Some Conclusions About the Building

The artifacts recovered from the cellar's layers all suggest a building that came down sometime at or just before 1710. Pipe bowls and stems all date to the 1680-1710 period. One pipe stem recovered from the lower ash layers had a "I F" maker's mark, indicating James Fox, a pipe maker operating at the turn of the eighteenth century. The datable wine bottle fragments and ceramic sherds, including the foot rim of a delftware punch bowl and several fragments of Challis type milk pan, also confirmed the early eighteenth-century destruction date. The layering of the cellar rubble suggests that the building came down fairly quickly. The lack of organic layers or silt layers as seen in other brick cellars points to a quick demolition, not allowing time for the accretion of trash or silt brought in by rain storms. Furthermore, excavators encountered several intact portions of the building's brick walls. These 16

Some of the most spectacular finds during Knight's 1940s excavation were windows fragments and window casements. Although the 1999 dig found no complete windows, excavators did recover a large number of window leads all from the cellar fill (Figure 12). Many of these leads contained legible dates. Four complete dates appeared in the leads: 1670, 1674, 1693, and 1695. The most frequent of these was 1674, which appeared twenty-one times. The year 1693 was the second most common, appearing seven times, while 1670 and 1695 appeared once each. Other partial dates included 168[?] and 16[?]3. The initials "WM," "WC," "WB," "BG," and "P" also appeared on many of the leads. All of this data point to a building which stood during the last quarter of the seventeenth century.

The building itself was a 43' 7" x 19' 4" brick structure whose subterranean portions were laid in a mix of English and junk bond. The cellar was divided into two rooms and the first floor above the English cellar probably mirrored this pattern. There was little solid evidence to suggest the height of the structure but an intact chunk of wall-14BC-268-may be an important clue. This wall fragment was one brick in thickness and measured 59 by 23 cm. There was evidence of plastering on both of the long sides as well as on one of the short sides. This suggests that this fragment was part of a door jamb in an interior wall. There was no evidence of such internal brick partitions in the unplastered cellar, therefore this wall was from a first or even 17 Figure 12. Dated window lead similar to those found in the Nassau Street cellar. second floor partition. The plastered wall fragment and large quantities of plaster at the bottom of the cellar rubble indicate that the upper floor walls were well plastered, whereas the cellar's interior wall was bare brick with scored mortar.

Some portions of the wall were filled with brick and mortar nogging rather than with well-laid brick. This suggests that portions of the wall were quickly or poorly built. The southern wall in the eastern half of the cellar also showed evidence of having been repaired and repointed. The portions of the collapsed wall which appear to have sheared off during demolition suggest that much of the wall was either held together by mortar or were made of poorly fired bricks.

Knight's 1942 excavation provides the first clues that only half of the cellar was paved in brick. The 1999 excavations substantiated that conclusion and found further evidence that the east side of the cellar was in fact never paved. This side appears to have been a work area and ash deposits in the hearth and on the soil floor surface hint at the cooking tasks that took place there. Knight identified a brick pad north of the east heart as a stair pad. While this is a possibility and consistent with other seventeenth-century studied buildings, three problems nevertheless remain with this identification. The first is that the width of the pad is too narrow to have sustained a full-sized stair case. The second is that the lack of bulkhead entrance means these stairs would have had to have ended in an above-ground stair box attached to the outside of the building. No archaeological features remain to support this possibility. The third is that the portion of the hearth cheek that would have formed the wall of the staircase does not show evidence of having been finished as such. The feature that Knight identified as a stair pad may instead have been the footer of a brick oven attached to the north side of the hearth. This would account for the difficulties with the stair identification and would also be consistent with the evidence of the area having been a highly used work space. Rich Neck Plantation's Structure B had the remains of a hearth-oven combination very similar to the one at Nassau Street, and several reconstructed kitchen in Williamsburg's Historic Area also have the same type of oven.

A Look at Roof Tiles

In 1997 Colonial Williamsburg excavators examined the remains of a subterranean brick kiln on the Bruton Heights property just north of 18 Williamsburg's Historic Area. The kiln was part of John Page's complex of building and, in up to five firings, contributed bricks and tiles to the Page buildings. As part of the analysis, Colonial Williamsburg archaeologists oversaw a series of acid extraction and xeroradiography tests on flat roof tiles from the Bruton Heights and several other Williamsburg and Tidewater sites (Figure 13).

Flat earthenware tiles have appeared on several sites in Williamsburg's Historic Area (Figure 14). Fragments came from a boundary ditch near the Chiswell House, from a natural gully near the Coke-Garrett House, from the seventeenth-century buildings in the Wythe Yard, and from the John Page home site and the Nassau Street site's 1940s excavations. Tests on these tiles aimed at determining the range of similarities and differences between all these artifacts at the elemental level. Acid extraction tests conducted at the College of William & Mary entailed removing of 200 mg of the tile and subjecting each sample to a series of chemical reductions. The result of these tests were a series of tables showing comparing the various relative relationships between the tiles' contents of magnesium, iron, zinc, and calcium. Xeroradiography is a less destructive procedure that entails subjecting the tiles to x-rays in order to ascertain their construction density and uniformity of inclusions such as temper (Blakely 1989; Metz et al. 1998).

These tests revealed that the Nassau Street tiles had the same density and rate of inclusion as the Bruton Heights tiles taken from both the kiln and the domestic site. Acid extraction showed that the chemical makeup of the Nassau Street tiles was for the most part like those found at the Page site. The strongest correlations were on the charts showing calcium vs. iron and magnesium vs. iron (Metz et al. 1998).

Figure 13. Flat ceramic roofing tile made in John Page's kiln.

The Builder's Trench

Figure 14. Anthony Hay House in the Historic Area, showing a flat tile roof.

One of the most important features excavated by the crew in 1999 was the builder's trench-14BC-328-which surrounded the eastern portion of the cellar. Knight's 1942 excavations did little damage to this feature, which measured between 39 and 41 cm wide. A pre-1942 utility line cut through the south side of the trench and cellar fill (doing extensive localized damage to the brick wall) but did little harm to the builder's trench apart from cutting through a 70 cm section near the southeast corner. Excavators removed two portions of the trench in hopes of finding datable artifacts which could provide dates for the structure's construction. The first section between units 1140N/504E and 1139N/502E, a 100 by 40 cm area south of the structure, revealed a flat bottom at 79 cm. Excavators water-screened the soils from this area through 1/8-inch screen but no artifacts resulted. A second portion between units 1143N/503E and 1141N/504E, revealed the same flat bottom at the same depth, but soils run through a regular ¼-inch screen yielded several datable artifacts. The most significant of these was fragment of a leaded crystal wine glass.

A small and very shallow posthole sat in the builder's trench's southeastern corner in unit 1141N/504E. This small feature aligned with the structure's foundation and measured 20 by 18 cm and was 2 cm deep. It contained a small circular postmold measuring 10 cm across.

The builder's trench held very few artifacts, consistent with its having been filled fairly soon after it had been dug. But despite its low artifact count a few very telling artifacts did present themselves. The most significant of these was a small fragment of leaded table glass excavators found in the 20 builder's trench against the cellar's eastern wall. This glass type has a TPQ date of 1676, indicating that the building went up some time after that date.

The Rest of the Site

Finding evidence of the seventeenth-century building and its environs was the primary mission of the summer dig. But a combination of grading, erosion, and construction left little evidence of the seventeenth-century landscape. Excavators worked in three areas surrounding the building's foundations-the road south of the cellar, the area south of the Bryan Kitchen, and the area to the south of the Bryan House itself. A few features had orientations that hinted at a seventeenth-century date, but their lack of artifacts made it impossible to say for sure. But excavators did find numerous features associated with the lot's later eighteenth-, nineteenth-, and twentieth-century inhabitants.

The first documentary appearance of a building at the corner of Nassau Street and Duke of Gloucester Street is the 1781 Frenchman's map. The map shows a rectangular structure sitting on the northeastern corner of the lot. Three smaller squarish buildings sit behind it along Nassau Street. The ownership of these buildings is unclear, although they may have belonged to John Lewis, a New Kent and James City County merchant who owned a few Williamsburg properties in the second half of the eighteenth century. The Williamsburg land tax records for 1795 show a parcel of land transferred from John Lewis to a William Bryan. If this was the same parcel as the one identified on the 1810 Bucktrout map as "Bryan," then this means that the buildings on the Frenchman's Map were Lewis's. Regardless of the colonial ownership of the lot, the reconstructed building currently sitting at the corner is a copy of the well-documented Bryan house. This house appears on several Mutual Assurance policies and was owned by William Bryan's widow Frances until 1818, when it passed to Jesse Cole. Cole was the owner of several other neighboring lots and rented out the home to a host of tenants, some of whom used the building as a store as well as a home. By 1830 John Hutchings owned the lot and used it as both home and store until well after the Civil War. Sometime after the war Hutchings or the next owner had the Bryan house moved to the back of the lot and turned to front onto Nassau Street (Farish 1940). Sanborn fire insurance maps of the block show the newly-moved Bryan House sitting at the rear of the lot adjacent to the First Black Baptist Church (Figure 15). Two early 1900s photos of the refronted Bryan House exist (Figure 16) and these guided the plan of museum reconstructers in 1940. By 1905 a new building stood on the corner of Nassau and Duke of Gloucester Streets. This was the Binn's Store and adjacent home, which occupied the corner until removed by the museum. By 1921 the old Bryan house had been replaced by an auto garage and concrete parking lot.

The archaeology of the area found ample evidence of the lot's changing shape and use. Excavators found hundreds of early twentieth-century artifacts ranging in type from ceramics and bottle glass to architectural remains 21 Figure 15. The Sanborn fire insurance map from 1904, showing the relocated Bryan house, the First Black Baptist Church and the newly built Binn's Store at the corner of Nassau and Duke of Gloucester Streets. Figure 16. Circa 1900 photo showing the old Bryan House newly set next to the First Black Baptist Church and facing Nassau Street. and chunks of poured concrete. The yards of the Bryan House and its outbuildings also showed considerable evidence of the site's history of construction, destruction, and reconstruction. But despite all this change excavators located a few small features dating to the seventeenth and eighteenth centuries and one unexpected find from the nineteenth century.

The Area South of Bryan Kitchen Excavations

Excavators laid the Colonial Williamsburg grid over the yard area south of the Bryan Kitchen and began excavating the area in checker-boarded one 22 meter units. The dig concentrated on the portion of the yard that sat to the west of the garden path and avoided areas whose digging would have hurt the existing vegetation. Although the layers in these units were heavily disturbed by Colonial Williamsburg activity, excavators discerned three basic layers. The first was a topsoil layer that averaged 11 cm in depth. The second layer was a CW fill layer of between 12 and 15 cm, and the third layer was a mix of sheet refuse and CW fill averaging 14 cm in depth. It was in this third layer that excavators located the site's 1940 James Knight cross-trenches which covered most of the area. With a few exceptions the bottoms of these units contained either clean subsoil or twentieth- century concrete and gravel.

After having seen much of the yard behind the Bryan Kitchen excavators opened a sixteen-square-meter portion of the most promising area. This open area, between units 1121N/490E on the northwest and 1118N/484E on the southeast, showed the same three-part topsoil, CW fill, and CW fill/sheet refuse stratigraphy that the other one-meter units exhibited (Figure 17). Several features sat in the subsoil at 34 cm deep. Most of these appeared to be small posts or driven posts, and their orientation and fills suggested that they dated to the nineteenth and twentieth centuries. Only two of these, 14BC-186 and 14BC-191, were oriented off the town grid, thereby suggesting that they may be seventeenth-century features. Context 14BC-186 was a square posthole with no visible mold which sat in unit 1120N/491E. This hole measured 27 by 28 cm and was 11 cm deep with a flat bottom. Its fill registered 10YR3/2 in the Munsell color book. Hole 14BC-191 sat in unit Figure 17 .Collection of small postholes located just south of the Bryan House Kitchen. 23 1121N/492E and measured 38 by 19 cm with a flat bottom at 13 cm deep. This hole's fill had a Munsell color of 10YR3/3. The artifacts from each of these postholes provided TPQ dates of 1787 and 1775, respectively.

Excavators also examined a large round feature, 14BC-75, which was most likely one of the two large dark round features identified by James Knight in 1940. This feature in unit 1119N/492E had a rounded bottom at 36 cm deep and measured 154 by 68 cm. This feature's dark fill containing pieces of concrete and sewer pipe and the feature's round bottom all suggested that this was a modern planting hole for a tree or a large bush.

The Area North of Bryan Kitchen

A grass lawn and brick patio covered the area immediately north of the Bryan Kitchen. After removing the topsoil and the bricks with their underlying sand bed, excavators began examination of the area between 1137N/490E at the northwest and 1131N/495E at the southeast corner (Figure 18). This area sat just to the south of the 1940 excavated portion of the seventeenth-century building and was cut for its whole length by a 44-cm-wide sewer pipe trench dating to the 1950s or 1960s. A 10-cm-deep layer of eighteenth-century and nineteenth-century century sheet refuse covered the portions of the area not sealed by the brick patio. Four 1940 James Knight cross-trenches appeared in this layer.

A large number of features were located in the area's subsoil, and although only a few were excavated, all were mapped. A large CW planting bed covered the west side of the excavation area bordering the Nassau Street sidewalk. This feature, 14BC-299, was 50 cm wide and 6 cm deep and had a dark loamy fill containing numerous twentieth-century artifacts. This feature ran the full length of the excavation area and also covered the backfilled eastern portion of the seventeenth-century building. The removal of feature fill 14BC-299 revealed a line of postholes all marking the western boundary of the property. One of these, 14BC-363, was the fill for a modern post but the other four-14BC-359, -361, -366, and -368-all appeared to be eighteenth and nineteenth-century features. None of these features were excavated.

Excavators examined two of the area's larger features as their orientation and visible fills suggested that they might have been related to the seventeenth-century building. Context 14BC-334 was a large linear stain running 200 cm on a roughly northwest-southeast axis and terminating at 14BC-336, which cut it. Excavators removed a 10-cm section of 14BC-334, which revealed a flat bottom at 5 cm deep. This feature's fill registered 10YR4/4 on the Munsell soil color chart and contained no artifacts except some shell inclusions.

An overlapping set of brick-filled features sat to the east of and partly underneath the reconstructed Bryan House well. These features in units 1137N/490E, 1136N/490E, 1137N/491E, and 1136N/491E were either round or off the town grid suggesting that they might be of seventeenth-century 24 Figure 18. Overall plan of the features sitting south of the Bryan House, north of the Bryan House kitchen and west of the well. Note the large 1960s era utility trench that cut through the whole area. origin. Feature 14BC-330 measured 86 by 86 cm with a fill that was principally brick rubble. The feature had a bowl-shaped bottom at a depth of 10 cm. The removal of this feature revealed the relationship between 14BC-332 and 14BC-378 the edges of both of which were visible before 14BC-330's excavation. Context 14BC-332 cut 14BC-378, which in turn was cut by root disturbance (14BC-380). Context 14BC-332 measured 112 by 69 cm and had a round-based bottom at 4 cm deep. The fill of this feature had a Munsell color of 10YR3/3 and had numerous small brick chip inclusions. The removal of this feature revealed a more complete view of 14BC-378, which measured 119 by 28 cm and had an irregular bottom at 4 cm deep. This feature's fill had a Munsell color of 10YR4/3 and contained no artifacts or inclusions.

The Nineteenth-Century Hearth and Related Features

A C-shaped brick hearth sat at the southwesternmost corner of the large excavation area in units 1134N/493E and 1132N/493E (Figures 19 and 20). This hearth measured 167 by 84 at the cheeks and 30 cm at the back of the firebox, and was constructed of reused seventeenth- and eighteenth-century bricks held together by a loose sand mortar. The shape and size indicated their age, as did the light covering of shell mortar on many of the bricks. 25 Both of the hearth's corners had been robbed three courses into the ground, making a full four courses of brick visible after excavation. Although the hearth was a brick construction, its bricks were very poorly laid showing no effort to bond the courses and little attention paid to the levelness or evenness of the bricks. The hearth's mortar beds were also very thick, in some places measuring as thick as 3 cm.

Excavators removed a large modern feature-14BC-311-which obscured the robbed northern corner of the hearth. This feature measured 38 by 31 cm and was 11 cm deep. Feature 14BC-309 was a small round pit that sat inside 14BC-311, although the two did not seem to be related. Context 14BC-309 measured 21 by 20 cm and had a bowl-shaped bottom at 10 cm deep. Context 14BC-311 cut into 14BC-346, which was the builder's trench of the hearth, but excavators did not remove this builder's trench. In plan it measured 6 cm wide and appeared along the entire outside of the hearth.

Little evidence survived of the structure that was once associated with this small hearth. Two round pits to the hearth's northwest may have been wooden pier postholes for an ephemeral building which stood above ground level. Excavators bisected both of these features but left half of the fill in situ. The easternmost of these features-context 14BC-337-located in unit 1134N/493E and measured 59 by 59 cm with vertical sides and a flat bottom at 23 cm deep. Excavators found three fill layers in the hole. Context 14BC-336 was 10 cm deep and had a Munsell color of 2.5Y5/4. This fill had a some brick chip inclusions but no artifacts. Context 14BC-357 was 9 cm deep, contained some shell inclusions but no artifacts, and was of a darker color than 14BC-336, registering 10YR4/3 on the Munsell soil chart. Context 14BC-358 was only 2 cm deep and contained no inclusions or artifacts and had a Munsell color of 10YR4/2.

Figure 19. Photo and plan of the loosely constructed nineteenth-century hearth. Note the large mortar spaces between the bricks.

Figure 20. Overhead view of all the features associated with the nineteenth-century structure.

The second-westernmost of these two features was 14BC-271 in unit 1134N/490E. This feature had a single fill of a 10YR4/3 loam and measured 72 by 70 cm. One small piece of feather-edge creamware gave this fill a TPQ of 1762. The sides of this feature were somewhat bowl shaped although the south side was relatively vertical. The bottom of the feature was flat.

The two-brick-wide remains of a pier may also be associated with the missing ephemeral structure. This small brick feature, 14BC-406, sat in eighteenth-century sheet refuse (14BC-304) in units 1134N/487E and 1135N/487E. Excavators only mapped and recorded this feature, but its orientation, alignment, and its composition of reused seventeenth- and eighteenth-century bricks all suggest its association with the nineteenth-century hearth.

In 1968 and 1987-89 Colonial Williamsburg archaeologists conducted investigations of the Thomas Everard House in the town's Historic Area. The 1987-89 excavation found the remains of a nineteenth-century slave home that sat on a portion of the lot once owned by the Tucker family. Subsequent research concluded that this cabin had been the home of Polly Valentine, a domestic slave of the Tuckers in the first half of the 1800s. All that remained of the structure were a brick hearth and the small surviving portions of three brick and stone piers which supported the sills and framing members of the 15-by-25-foot home. The Nassau Street nineteenth-century structure was held above ground by at least two posts and one brick pier but was nevertheless had in most respects the same architectural plan. Both buildings had a small brick hearth which sat outside one of the structure's gable 27 end walls. The corner posts on the hearth gable end roughly aligned with the front ends of the hearths' cheeks (Edwards 1990).

While 1999 excavations identified, mapped, and photographed the hearth and excavated two of the associated postholes, other features possibly associated with the structure and the hearth's builder's trench remained untouched for future examination.

All indications point to this being a mid- or early nineteenth-century slave cabin or low-cost dependency. The sand mortar is a nineteenth-century construction technique, but the fact that the bricks formed a firebox suggest that this structure pre-dated the common use of stoves later in the 1800s. The firebox was constructed out of reused seventeenth- and eighteenth-century bricks which still had evidence of having once been shell mortared into another construction-a common cost-saving measure employed in slave architecture.

The building would have sat above ground on piers and posts and been constructed of either logs or framing and split wood weatherboards. During the first half of the nineteenth century John W. Hutchings owned the Bryan House. At various times he used the building as a home and a store and at other times he leased it out to tenants. The Mutual Assurance Society revalued the property several times between 1820 and the Civil War, but none contained a mention of this rough-constructed building. The revaluations repeatedly estimated the value of the "dwelling and store" at $600 in 1830 and 1860, and each of the plot maps showed both the home and a kitchen. It seems that this kitchen was the same one excavated by James Knight and indicated on the 1810 Mutual Assurance policy. But unlike the excavated and reconstructed Bryan Kitchen the policy maps show a kitchen whose gable ends align east and west rather than north and south. The maps on the policies are not to any strict scale and are somewhat impressionistic-but it is curious that the maps show a kitchen aligned the same direction as the nineteenth-century structure and not the same as the eighteenth-century kitchen excavated by Knight in 1940.

The Road

Colonial Williamsburg and City of Williamsburg backhoe teams removed Nassau Street's blacktop and the sand and hard pack that supported it (Figure 21). The subsoil level beneath the road layers dropped dramatically towards the southern half of the road's excavation area. In unit 1146N/505E, for example, the 9-cm-deep blacktop sat directly on clean subsoil, whereas in unit 1118N/505E the black top sat atop 12 cm of sand and hard pack, 9 cm of old road surface, 38 cm of redeposited clay, and a 2 cm deep layer of silt between the clay and the subsoil. The depth of the subsoil and the presence of deep redeposited clay suggest that the road had been heavily graded at the time of its first paving. Early twentieth-century road crews removed much of the older mud that was the surface of the unpaved Nassau Street and created a hard bed of clay to support the new tarmac surface. One result of this 28 Figure 21. Cross section of Nassau Street's paving and road-supporting layers. grading was the near-complete removal of archaeological features in the road bed.

A few very shallow features hinted at what might have been in the area before the road-related destruction (Figure 22). Three small postholes sat in units 1136N/499E and 1135N/499E. The smallest of these, 14BC-349, measured 16 by 17 cm and had a flat bottom at 3 cm deep. Posthole 14BC-321 sat to the northeast of 14BC-349 and measured 14 by 13 cm. The posthole showed a round postmold (14BC-351), which measured 11 by 13 cm, and the entire posthole was 15 cm deep. Posthole 14BC-319 sat directly west of 14BC-349. It measured 32 by 44 cm and was 6 cm deep. This posthole also had a mold (14BC-350), which measured 15 by 16 cm and was 4 cm deep. The fill of none of these postholes contained artifacts, making their dating difficult. The only artifact was found in postmold 14BC-351, which contained a small fragment of a locally manufactured pipe stem. One posthole on the east side of the excavation area appeared to be roughly in line with the western three postholes. This posthole, 14BC-323, sat in unit 1137N/505E and measured 21 by 22 cm. It had a flat bottom at 3 cm deep but its fill contained no artifacts.

What We Learned

All the evidence accumulated in the seventeenth-century building's cellar and builder's trench suggests a structure that was built around 1676 and came down by 1710. The site's low artifact count and the lack of related seventeenth-century features make it difficult to determine high points of occupation, work areas, and the various activities in and around the building. The dates of the structure rule out Peale and Weeks as possible owners, and the sophistication and obvious costliness of the architecture mean that this was no easily overlooked tenant structure.

Figure 22. Plan map showing all the features which Nassau Street's paving covered.

A few finds seem to tie the building to the Page family. The first is the presence of Page roof tiles, which are common to both the Nassau Street building and the John Page 1662 home. But the presence of these tiles may be a reflection of local supply networks rather than implying the owner's membership in the Page family. The building's brick makeup and its proximity to John Page's land and the 1684 Bruton Parish Church-a Page family project-suggest that it may have been a Page family property. But the large amount of brick architecture distributed throughout Middle Plantation agues against this type of defacto connection. The problem of the missing brick buildings remains as well. If this was Francis Page's home, where are the remains of the other two brick buildings he identified in his will? Knight's excavations turned up no other seventeenth-century brick remains in the immediate area. It is possible that later construction demolished what ever had remained of the two documented outbuildings, or that they still sit in some as yet unexcavated plot or in Duke of Gloucester Street, but these are remote possibilities. The fact that the 1999 excavations encountered no evidence at all of out buildings-be they brick or post-argues against Francis Page's ownership of the Nassau Street building.

Furthermore, excavators found no substantial trash deposits or trash distribution fields like those found on other tavern sites both in and outside Williamsburg (Blick 1999; King and Miller 1987). While there is nothing to suggest that Weldon or any other tenant's tavern operation was a successful one, the complete lack of tavern refuse argues against the site having sustained an ordinary at any time.

Instead the archaeology of the site suggests that neither of the two documentary ownership scenarios are correct. A third owner therefore is a strong possibility. Once again the location of the 1634 Palisade is vital for determining a possible owner. If the Nassau Street building sat to the west of the 30 palisade, then it may have been on the Peale/Weeks tract. The structure's dates rule out either Peale or Weeks as the owner, but Robert Weeks falls out of the York County records in 1674. If this was close to the time of his death, then it is a safe assumption that the tract changed hands at a time that segues well with the construction of the Nassau Street building. David Bray is also a possible owner, but given the paucity of James City County land records and the site's lack of substantial trash deposits, a definitive answer may be an impossibility.

Nevertheless, even without an identified owner the archaeology of the site offers many important insights about the owner, the building he built, and its relationship to the Middle Plantation community.

Obviously the building of a brick home required capital. Owners of brick buildings generally had that capital although not all owners were men of especially high status. Homes like the Custis's Arlington, the Ludwells' Rich Neck, or Sir William Berkeley's Green Spring were large and imposing structures which announced the owner's wealth and position as well as their embodiment of English standards and tastes. But Middle Plantation was a community that had a disproportionately large number of brick buildings. Not all of the identified seventeenth-century brick building were the seats of great Virginia families. For example, two buildings sit on the George Wythe property and underneath Prince George Street. The orientation and remains of the structures indicate that they are seventeenth-century buildings. During the seventeenth century they would have sat on land belonging to John Page, but could not have been Page's residence which was on Bruton Heights. Furthermore all of other possible owners seem to be accounted for in the York County records. These substantial buildings therefore may have been used by Page tenants or even servants. While it is highly unusual to see servants or tenants housed in structures of such expensive materials, the Pages themselves were unusual in their extensive and well documented use of brick. These two buildings suggest that in Middle Plantation brick buildings were the homes of a wide range of occupants and that while the Nassau Street builder needed access to some capital to build in brick, the amount of that capital may have been lower than in other parts of the colony where the architectural infrastructure was less well developed.

The 1999 excavations located very little evidence of the building's surrounding landscape within our excavation area. Nevertheless, some of these absences are telling. The first was the total lack of slot fences connecting to the brick foundation. Virtually every other seventeenth-century site has yielded at least one episode of fencing which related to the main dwelling. The absence of fence lines at Nassau Street probably result from the site's grading and erosion during the centuries between its original occupation and today. The surface level in front of the Taliaferro-Cole House-a home that has stood intact since the 1740s-is 58 cm higher than the subsoil surface layer in the road adjoining the building's remains. The combination of natural slope and grading has removed more than enough soil to account for the absence of relatively shallow features like slot trenches.

The few surviving seventeenth-century features clustered to the south of the structure. This admittedly thin evidence suggests that the building faced north with the southern features being evidence of work areas behind the dwelling. None of these few postholes were large enough to be structural components of outbuildings suggesting smaller dug or punched holes for work purposes. If the front was to the north it may have fronted the Middle Plantation Horse Path. This orientation of the building would support Knight's assertion that the Horse Path ran along Duke of Gloucester Street between the ravines on both ends of Nassau Street.

The artifactual evidence retrieved from the cellar's rubble fill indicate a quick destruction taking place in the first decade of the eighteenth century. Without a solid connection to a particular family or individual it is difficult to explain single event as potentially capricious as the decision to demolish a brick home. It is also important to note that this demise was untimely. If the building went up soon after the builder's trench TPQ date of 1676 and came down by 1710 this means that the building stood for as few as 34 years. This is a decidedly short life span for a brick building that did not end catastrophically. Looking at town events and other contemporary Middle Plantation buildings suggest possible scenarios for the Nassau Street structure's demise.

In 1699 Theodorick Bland surveyed the 220 Middle Plantation acres that were to become the new capital city of Williamsburg. Soon after the town began to grow lot by lot. During the first few years of the eighteenth century town development hung to the east side of the city. The Capitol building was well underway by 1704 and a few nearby new homes fronted Duke of Gloucester Street. Regulations tightly controlled the development of these lots stipulating the size, materials, and time frame of construction and as early as 1709 Williamsburgers owned title to as many as fourteen lots. But this development all took place in the east side of town near the new capitol. In this area workers eliminated vestiges of the old Middle Plantation landscape by realigning some homes and pulling down interfering buildings like the four Page-owned brick structures that capitol builders removed from Duke of Gloucester Street (Hellier 1989:1-19).

In the west of town, however, the older Middle Plantation landscape remained largely intact. The Wren Building, the 1684 Bruton Parish Church, and several other local homes and taverns coexisted with the new east side for at least a decade. Why then did the Nassau Street building come down before development pressures seem to have made themselves felt? One possible explanation lies in a connection between the Nassau Street building and the 1684 Bruton Parish Church.

This Church was first proposed in the 1670s to replace an older wooden parish church not in Middle Plantation. The land for the church came from John Page who was a prominent vestry member. After the vestry rejected the original church design Francis Page became the principal builder. Like so many other Middle Plantation buildings the plans were for a brick church, and its location on Page land within sight of John Page's 1662 mean that the 32 bricks probably came through Page sources. The building was completed in an imposing Jacobean gothic style with curved parapet gables and sturdy looking buttresses. But as soon as the building was completed the vestry voiced concerns that the building was not constructed in a "Workmanlike manner." In order to keep the building in good condition they empowered Francis Page to see to its maintenance (McCabe 1856:594).

But despite ambitious hopes the 1684 church turned out to an albatross around the vestry's neck. The now-lost Vestry book contained a litany of repair woes. In 1693 the vestry noted that "Whereas ye inside work on ye Church ought to be rectified and repaired, it is therefore ordered that the... Churchwardens provide an able workman to effect the same, and that it be done as soon as they can." Then again in 1699, on the eve of Williamsburg's founding they recorded that "the Church again stands in need of repairs which are ordered." In 1702 "repairs in brick and wood ordered upon the Church" and the following year "a new pulpit is required, the pews ordered to be repaired, the floor raised, &c" (McCabe 1856:596, 597).

By 1706 this constant round of repairs had begun to try the vestry's patience and they began to look for an alternative. They wrote "the Vestry considering ye great charge ye parish hath been at for ye repairing of ye church, and how bad a condition it is still in,-ordered, that twenty thousand pounds of Tobacco be levied this year for and towards building of a new church." By 1714 plans were underway to replace the old church with a new one-the one that now stands on Duke of Gloucester Street. The vestry's new plans hinted at some of the problems they faced in the old building. They required that "the Church wardens take some convenient care to make the Church tite against the wether."

The 1684 Church was a heap and by 1715 the new church was underway, hopefully better built and less vulnerable to winds and rains than the old one had been.

The new church was larger than the old, but only by about 15 feet not including two wings (it was later expanded to full size it has today). In fact the vestrymen initially planned to build a church of the same size as the original, writing in 1710 that "the matter being debated it was Resolved, that a Church of ye same dimensions of ye old church will be large enough." Building decay is the principal reason the vestry chose to rebuild rather than repair (Goodwin 1903:31-32; McCabe 1856:610).

The 1999 digging season unearthed evidence suggesting that the Nassau Street building underwent a similar cycle of decay and repair as the Church that sat only a few dozen yards to its northeast. Very little of the structure's walls survived the demolition. But the southeastern portion of the wall of which fifteen courses remained showed evidence of both repairs and poor construction techniques. Several course of the brick were repointed with different colored mortar and a large portion of the wall was filled with brick bat and mortar nogging (Figures 23 and 24). These small bits of evidence suggest that, whether by building incompetence, poor quality building materials, or limited resources, the Nassau Street structure was not one of the 33 best-built dwellings in the area. This may account for the building's rapid destruction before development pressures changed Williamsburg's west side from Middle Plantation into the new capital.

This architectural inference combined with the documentary evidence of the gradual decay of the 1684 Bruton Parish Church suggest that while Middle Plantation may have had more than its share of brick buildings, not all of these were up to the standard of the best homes, like Page's 1662 Bruton Heights home or the Ludwells' Rich Neck. Although there is no smoking gun connection between the Nassau Street building and the 1684 Church their nearness in time and space and their similar, though differently documented Figure 23. Inside portion of structure's south wall. Note the unevenness of the brick bond and the pronounced evidence of repairs to the wall's mortar. Figure 24. Overhead view of the south wall. The photo clearly shows the mortar and rubble fill built into the walls. 34 fates suggests that some of the same builders or building materials went into both edifices.

In his Notes on the State of Virginia, Thomas Jefferson wrote that some of Williamsburg's best known buildings were little more than "rude, mis-shapen piles, which but that they have roofs, would be taken for brick-kilns." Had Jefferson been alive one hundred years earlier and stood between Bruton Parish Church and the Nassau Street building perhaps he would have reversed his observation, noting instead how well-formed these buildings were, although in fact some were no more stable than brick kilns (Jefferson 1782).

Where Do We Go From Here?

By the fall of 1999 City of Williamsburg road crews replaced Nassau Street's surface and Colonial Williamsburg groundskeepers replaced the Bryan House's brick patio. Very little of the seventeenth-century site-apart from the foundation-remain for curation. But excavators left almost all of the site's nineteenth-century component untouched. A complete excavation of this site would require removing the brick patio at the door of the Bryan Kitchen and digging right up to the Kitchen's foundations. Unfortunately much of the area abutting this foundation was probably heavily disturbed during Knight's 1940 excavation. A future excavation should be able to locate enough of the nineteenth-century building's posts to create a full outline, and the excavation of the hearth's builder's trench should provide a construction date.

The 1999 dig did not see all the possible areas for seventeenth-century outbuildings. The 1940s digs at the Bryan House and in the Taliaferro-Cole yard yielded no signs of seventeenth-century buildings. But as yet there has been no systematic examination of Duke of Gloucester Street. A dig there may find evidence of the Nassau Street building's surroundings.

The 1999 dig was the Foundation's first attempt to conduct a major seventeenth-century excavation in the public eye. The overwhelmingly positive public response to the dig and its interpretation-both costumed and uncostumed-indicates that an active and engaging interpretation of Williamsburg's seventeenth-century history has an audience. Furthermore, rather than clashing with the museum's primary interpretation a glimpse of the more distant past enlarges and enlivens the museums other interpretive programs. There are numerous understudied seventeenth-century sites in the Historic Area. These would all be great candidates for a public dig done in the style of the 1999 excavations. Not only would the public benefit for these possible digs, but both the archaeology and history of the region would continue to explore new territory.

During the archaeological investigation of the Bryan Lot in October, 1940, the foundation of a very early colonial building was discovered. This foundation was only four feet from the southeast corner of the Bryan house foundation, lying partly in Nassau Street and partly in the Bryan Lot. Only the portion in the Bryan Lot was excavated in October, 1940, because of the impracticability of closing and barricading Nassau Street at that time. In April, 1942, permission was granted by the city to close a part of the street for further investigation of this foundation. This work was done at once and the area backfilled as soon as all available data was secured.

The foundation was of a 43' - 7" x 19' - 4" brick building determined by the wall thickness. A large amount of broken brick and shell mortar composed practically all of the basement fill. It had a full basement with the western half paved with well-preserved brick. No evidence of brick paving in the eastern half was discovered. There was an inside chimney on each end, of which the eastern chimney appeared to have had a basement fireplace. This was proven by the presence of a layer of wood ashes as well as by the fact that the brickwork of the sides and hearth showed signs of exposure to fire. No indication of fire was discovered in the western chimney so that the chimney mass above (now missing) was supported originally on a relieving arch. Thus this end probably did not have a fireplace other than on the first and second floors. It is to be noted that the foundations of these chimneys do not center in the building, but are slightly to the south of the centerline. (See Archaeological Drawing, Block 14, area B-2).

Entrance steps to the basement were located on the east end near the northwest corner. Several bricks which were remains of the bottom tread and a strip of shell mortar which showed the imprint of the wood nosing were exposed. All other evidence of the steps had been destroyed.

The wall thickness of this foundation ranged from 18½" to 19", with the bottom course acting as a 22" spread footing. Very few whole bricks were used in the interior of the wall. It would appear that in order to utilize the bats and broken bricks, these were employed as fillers for the inside of the wall while the whole bricks were used on the interior and exterior face.

No evidence of outside steps to the first floor could be found on either the north or south side of the building. These were probably destroyed when the building was demolished and all traces of them have since disappeared.

The artifacts found in the basement fill were composed mostly of broken ale bottles and pieces of lead came. The ale bottles were of the early colonial variety, very similar to those found at Jamestown, Virginia. Practically all of the lead cames were found in two locations within the basement. In the central portion of the paved area were found almost the complete remains of the lead work of an early leaded glass window lying flat on the paving. There were also many broken pieces of window glass which originally formed a small rectangular and diamond-shaped panes. This window has been restored and will be placed in the Old Court House Museum. About five pounds of lead cames were found in the extreme southeast of the basement, in so damaged and distorted a condition, however, that it was impossible to restore original shapes or sash sizes. Several pieces of broken clay pipe were also discovered among the debris of brick and mortar. Very few pieces of broken china or earthenware were found, and from the general appearance, it seems that this basement was cleared of all contents or rubbish, before the building was demolished.

There being no record of any structure on this site it is almost impossible to state what dwelling this was or anything in regard to its early ownership.

The building that was here was not parallel or perpendicular to any of the Williamsburg street lines and it originally faced either NNW or SSE.

In the seventeenth century when the early settlers first began to inhabit Williamsburg, then known as Middle Plantation, there was a traveled way or road on the peninsula known as the "horse path." This traveled way followed the high ground or ridge between the ravines that drained to the York and James Rivers. Should this road be laid out again along the ridge, the seventeenth-century building foundation would be on the south side and practically parallel to it with the NNW side of the building facing the front.

It may be mentioned that the early church foundation located in the Bruton Churchyard and uncovered in 1938 would be just to the north of the road as it winds eastward.

Photo N6531. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.

Photo N6533. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.

Photo N6882. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.

Photo N6883. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.

Photo N6884. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.

Photo N6885. Photograph on file, John D. Rockefeller Library, Colonial Williamsburg Foundation.