Colonial Williamsburg Digital Library

Shop A-1

Without any reliable archaeological or historical data to locate or identify the early shops that Anderson operated on the property, much of what follows is necessarily based upon conjecture and cannot be construed as an absolute interpretation. The plan of the 1931 excavation raises many 20. questions which, due to the construction of the modern garage, can never be satisfactorily answered. On the site of the reconstructed building, the plan shows several features which have since been destroyed. Most notable are two sections of brick wall, one along the western property line and the other at a right angle to its southern end. At the location of the present northern wall of the garage were two brick features, one an elongated rectangular box and, to its south, two brick projections. According to the chronological key provided with the plan, all of these features are of the same period, of construction. Although many inaccuracies have been discovered on this archaeological plan, it would be presumptuous to assume, merely because the data given does not conform to conjectural ideas, that this interpretation of the chronology was also in error.

There is a discrepancy, however, in the information. It is assumed that relative dating for the structures located in 1931 was mainly on the basis of brick size and color and mortar type. If this is so, there is no apparent reason for the conclusion that the walls and brick box are of the same period. The sizes of the brick used in these two features are quite different (walls: 8½" x 3¾" x 3 ½"; [2 ½ per brick schedule 1932] box: 8¼" x 4" x 2 ¾"), possibly reflecting bricks of two different periods. Although brick size is by no means the most satisfactory method of dating a structure, it is, alas, the only data available. But because it makes little or no sense for these two features to be contemporary, we can reasonably consider the possibility that they were not.

This then leaves something of a problem. There is neither historical nor archaeological data to indicate when Anderson arrived on the property or what structures were there at that time. In light of information to be presented later, it seems likely that the walls shown on the 1931 plan may, indeed, represent a building either constructed by Anderson shortly after his arrival on the site or one which was already there. The western wall was one brick thick, as was that at the south. There is no way of determining the building's north-south length, though its east-west dimension must have been approximately 20'00" wide, based upon the location of later Walkway "B".

The thickness of the walls would indicate a frame building of one or one and a half stories. Two post holes located to the south of the structure (ER 1972B and ER 1973W) seem to be temporally related with Shop "A-1" and may have supported a stair entrance to an upper level storage or living loft. It also appears from the 1931 plan that this building was not well constructed as the southern wall is not actually straight but rather bends slightly to the north.

As cautioned above, the preceding must be taken as the most likely interpretation of the scant data available. It is hoped, however, that the succeeding pages (which are based upon more substantial information) will lend a greater degree of credibility to this conjectural first shop.

Forge A

Located to the north of Shop "A-1" and to the west of the First Period Kitchen (see p.50) was found the fragmentary 22. remains of a brick box. This feature was uncovered during both the 1931 and 1974 excavations but it was not until the removal of Forge "E"(see p.35) that it was properly identified as an earlier forge.

Although the foundation of Forge "A" was not complete, its size and shape could be estimated with reasonable accuracy. External dimensions seem to have measured 6'1" by 6'00"; the north, south and west sides were probably 1½ brick courses wide while the eastern side was a full 2½ courses in width and no doubt supported the weight of a chimney. There is, however, no evidence of a structure over this forge; thus, it is most likely that it was external and free standing.

An interesting yet difficult to interpret feature was located in the north side of Forge "A's" foundation. A small hole (approximately 2½" in diameter and 11" deep) had been drilled through the outside course of the bricks (ER 1963H), into which a sheet-iron sheathed wooden post had been set. There can be no doubt that this post was associated with the forge for its socket was sealed by the same ash deposit that covered the rest of the forge well (ER 1963G). Two explanations for the post's presence have been suggested, neither of them fully satisfactory. It may have been an added support for holding large heavy items over the fire (though its small diameter does not suggest the strength necessary for such a task) or it may have had some relation to a bellows, the mechanics of whose arrangement is not clear. As none of the other forges exhibited such a feature, one may deduce that its function was peculiar to whatever work was performed at this forge.

Around the Forge "A" foundation discovered beneath Forge "E", a spread of coal dust and ash (ER 19631) is believed to represent the working level for the forge. If so, the presence of the ash on the west side of the structure provides evidence to suggest that this forge was operated from the west. Although a construction terminus post quem of post-1745 was obtained, no artifacts recovered from this area were helpful in either dating the forge's use or identifying the work done there.

Shop B

Located approximately 25'00" south of Shop "A-1" was found the remains of another small structure which was probably associated with the early days of Anderson's occupation. Although all but the southern wall had been robbed, the lines of the walls were delineated by the location of rubble-filled "robber's" trenches (ER 1969J, ER 1970C, and ER 1975Z). The southern wall was represented by a one course wide foundation and was constructed above a level of gray loam (ER 1970Z) and an earlier backfilled privy pit (ER 1981K). Although instability of this underlying fill caused part of the foundation to settle, there was no evidence of any attempt to repair the damage.

The building itself was found to measure 12'3" by 10'00" with an east-west orientation. Located close to the center of the floor was a post hole (ER 1981C) which was definitely associated with the use of the structure (it, too, cut through the gray sandy loam). It has been suggested that the post supported a storage loft or perhaps a pyramidal-shaped roof. Outside the eastern wall of the shop was found evidence for 24. the location of a doorway. Two small post holes (ER 1975X and ER 1975W) flanked three bricks, and located as they were at the center of the eastern wall, they may represent the location of a stoop entrance.

The gray sandy loam level upon which the building was constructed contained fragments of creamware (post ca.1769), thus helping to substantiate the theory that Anderson constructed this building after his arrival on the site. Further support for that hypothesis is found in the similarity of Shops "A-1" and "B" to two structures excavated in the rear of Lot 19.³⁴ As mentioned earlier, Anderson already owned Lot 19, and it is possible that he moved both structures when he acquired the new property.

Forge B

Located 6'3" to the south of Shop "B", the fragmentary remains of yet another of Anderson's early forges were found. The southeast corner of this one course wide brick box was cut through by later Forge "G" (see p.40). Although only the northern and eastern walls of the structure were extant, the remains were 9" by 4'10". The forge itself was laid upon a level of gray sandy loam (ER 1971S) which represented the backyard fill from the pre-Anderson period of occupation, and was equivalent to the layer upon which Shop "B" was constructed.

Directly to the west of Forge "B", a depression (probably a natural gully) had been utilized as a dump for clinker and iron waste (ER 1971J). The close proximity of this pit to the forge, and the lack of indication of an associated structure, would tend to indicate that, as with Forge "A", this fire 25. box was also externally operated.

Shop A-2

The dimensions of this southern extension were 24'00" by 20'2" with a north-south orientation. Much of the western wall was extant and was 13" wide (1½ brick courses) and laid in English bond. Unfortunately, the northern end of this wall, where it intersected with the southwest corner of Shop "A-1", had been destroyed by the construction of the modern garage. The southern wall had been entirely stripped of its brick, but the level of the base of the robber's trench (ER 1974B and ER 1975B) indicated that the western wall had been sunk into the grayish loam level in order to establish a firm base on the downhill side of the ravine. The southern wall had then been stepped up to the ground level. Like this southern wall, the eastern foundation had also been robbed — with the exception 26. of one brick located directly in the southeast corner of the building (ER 1973N).

As noted above, evidence for the connection between Shops "A-1" and "A-2" had been totally destroyed by the construction of the modern garage. Because of the length of the total shop (i.e., structures "A-1 and "A-2), it seems most probable that the southern brick wall for Shop "A-1" would have been utilized when the addition was built. It is possible, for example, that the wall was dismantled to ground level, leaving a brick sill on which vertical studs could be placed to help support the weight of the structure's roof. Alternatively, the wall may have been left as a partition wall between the two shops. However, the former possibility seems the more plausible when considering the lineal arrangement of the forges within the shops which points to a strong likelihood that there was a visual line-of-sight throughout the combined workshops.

Forge C

The remains of Forge "C" were first discovered during the 1931 excavations and were re-exposed during preliminary investigations on the site in December 1974. Located within the confines of the modern garage, the forge measured 6'2" by 4'4" and survived to a maximum height of four courses. The north, south, and east sides were one and one-half courses wide while the west side was a full two courses in width, again suggesting that this side supported some sort of chimney or hood.

Though no working surface could be identified around this forge (having perhaps been removed by the previous excavations), 27. two circular pits were found in association with it. They lay both to north and south; measured 3'6" in diameter, cut 9" into the subsoil, and are believed to have been used to seat barrels of water used in quenching the hot iron. The mixed fill in these pits contained many chunks of rusted iron which, upon cleaning, proved to be unfinished waste material from the forge.

The southernmost of these pits (ER 1916B) had been cut through by the builder's trench for the modern garage. On the 1931 archaeological plan, however, the southern wall for Shop "A-1" shows a small bite out of the north face of the brickwork which may have been caused by digging the pit for Forge "C". If this is the correct interpretation, it helps to substantiate the hypothesis that when Shop "A-2" was added, the southern wall of Shop "A-1" was at least partially dismantled.

Forge D

The robbed remains of Forge "D" were first discovered during the preliminary excavations in the winter of 1974-75. Three levels of destruction fill were identified within the confines of the 6'4" by 4'9" feature and these aid in presenting a vivid picture of its dismantling. At the top was found a scatter of brick and mortar fragments (ER 1918D); and beneath it a level of coal clinker (ER 1918E) covered the entire feature. The fact that this clinker overlay the robbed walls of the forge itself is evidence that the clinker did not originate from the use of Forge "D" and was waste from another 28. forge in use after its destruction [or saved from its former use.] Below the clinker level was a layer of mortar and brick rubble contained within the rectangular wall lines of the robbed forge (ER 1918H). The robber's trench formed by the salvaging of the foundation bricks suggested that the walls had measured one and a half courses in width on the north, east, and south sides where two bricks survived, and gave some indication that the west side was two courses wide, though the evidence on that side was fragmentary. The bottom of the area originally enclosed by the walls of the forge was covered with mortar fragments (ER 1918J) which probably came from the dismantling of the structure. The forge measured 6'00" by 4'10" and stood onto the same line as others located on the site.

As with Forge "C", two large round pits were found in association with Forge "D." Unlike the former, however, these were not uniform either in depth or diameter. The hole to the southern side of the forge (ER 1918G) was excavated in 1974 and found to be 3'6" in diameter, and 2'00" deep, and contained iron waste material along with creamware sherds dating the fill after ca.1770. [is there terminal date′] The north hole measured 3'1" in diameter and was 10: in depth (ER 1973S). [′ 1973 T′ 1973 V′]

With the removal of the upper level of this feature, an ash-filled circular hole 1'2" in diameter was revealed within the larger hole. It appears that this small hole represented the actual diameter of the barrel (or tree stump′) which had been seated within the large hole around it.

As with the two pits associated with Forge "C," both of these are also assumed to have been the seatings for either water barrels or tree stumps upon which the anvil would be 29. placed. The presence of comparable features adjacent to these forges suggests that both forges belong to the same construction period — perhaps the 1777 construction billed by Humphrey Harwood.

Drain Slot

Close to the southern end of Shop "A-2", a trench 1'3" in width at the top was discovered which had apparently been added during the Period II usage in order to drain standing water from the eastern side of the building (Plate IV). The fill in this trench was stratified; the top 5" consisting of mottled clay and loam (ER 1975V). This fill was beneath (thus pre-dating) the ash floor level from the Period III usage (see p.42). Below the clay fill the trench narrowed to 1'1" in width and contained another 5" of gray-loam (ER 1975Y).

The trench ran the entire width of the shop and extended to the east of it to form an "L"-shaped catchment. It appeared that wooden planks had been laid across the bottom 5" of trough and were covered with the clay fill in order to get it back up to the shop's floor level. These planks probably rested on iron bars which spanned the lower 5" of the trench. One of these bars was found in situ while notches in the vertical sides of the trench marked the presumed location of two others. The surviving bar proved to be an interesting artifact in its own right (ER 1975). Measuring 1'9½" in length, the bar's ends showed that a blacksmith (perhaps an apprentice learning some of the basic skills) had worked it. One end was merely bent over on itself, but the other was turned at two right angles forming a "U" shape. Also at this end, a wedge-shaped 30. piece of iron had been attached — its purpose as yet undefined.

It would appear that the bottom 5" of the trench was left open to drain water through to the west side of the building. A concentration of iron artifacts within the trench to the eastern side of the structure, and the lack of finds at its western end, support such an interpretation as the entrance to the drain would be likely to accumulate heavier waste while the "downstream" end would only receive those artifacts which could be carried by the water run-off.

Along with iron waste, the artifacts recovered from the eastern end of the trench included a broken French wine bottle and part of a lead-glazed French earthenware chamberpot (ER 1975Y) This preponderance of French artifacts suggests that the filling of the drain took place ca.1781 while French troops were quartered in Williamsburg.

Shop B

During this second period of Anderson's activities the small southern outbuilding, Shop "B" (see p.23) continued to be used, but it appears from the character of the artifacts found in association with it that its use had become somewhat more specialized. Directly to the east of the structure and spreading over a surviving area measuring 7'5" by 4'6" was found an accumulation of discarded iron waste (ER 1975C). This dump was unlike any other concentration of waste found on the site, containing as it did a wide variety of objects, many seemingly related to wagons or carriages but also including two incomplete cutlass hilt guards. It is probable that this waste was derived from the main workshop (Shops "A-1" and "A-2") 31. rather than from Shop "B" which had no forge associated with it.

Analysis of the distribution of artifacts across the site indicates that two different operations had been performed in an adjacent southern outbuilding. Evidence of the first was provided by the preponderance of bone discs and the larger bones from which they had been cut which were found almost exclusively in the area to the east of the small shop. Because these discs are not found in the fill within Shop "A-2," it may be deduced that when the disc production began, that shop was already in existence. The discs themselves vary in size, but were made with a simple carpenter's bit which cut the desired sized hole into the bone, and in the process drilled a single hole in its center. It is believed that these discs were covered with cloth and used as buttons, and may have been associated with the maintenance of military uniforms. [See Anthony Artifact Rep]

The second activity apparently related to the storage and cleaning of weapons. Although gunflints were found in various areas of the site, the greatest number was found in the area associated with the small outbuilding. Again, the association can be made with the Period II occupation, placing this weapon work on Lot 18 between 1777 and 1779, when we know from the historical records that Anderson began to use the widow Hay's old cabinetshop for servicing the public arms.

Fence C and the Northern End of the Lot

With the construction of the southern extension of the main workshop (Shop "A-2") and the addition of Forges "C" and "D", Anderson apparently abandoned and dismantled his use of Forge "A" to the north of Shop "A-l." Whether he already 32. had moved into the main dwelling on Duke of Gloucester Street is not known, but it is clear that with the Period II construction, an attempt was made to separate the industrial activities of the southern end of the lot from the domestic activities at the north. This is most clearly seen by the erection of Fence "C" (see p.71) which ran east-west across the site flanking the south wall of the first period kitchen.

With this separation of activities, the area around and above Forge "A" was used as a repository for domestic trash, and later for the dumping of coal clinker from the southern forges. As shown in Figure 5, the vertical distance between Forge "A" and Forge "E" (see p.35) represents a time break between the destruction of the early forge and the construction of the Period III forge.

Shop A-3

This structure was excavated in 1931 and again during preliminary exploration mounted during the winter of 1974 (see Figure 4). It consists of two brick foundation walls on the northern and western sides enclosing a brick forge. Since the 1931 excavations a steam line was installed between the James Anderson House and the reconstructed Mary Stith Tin Shop, located on the lot to the west. This utility line was cut through the forge foundation (and all the fill beneath it) 33. and destroyed a critical portion of the shop's western wall. Other modern disturbances were extensive south of the forge and obliterated any evidence of a working surface or related features for this shop.

The North Wall

Directly north of the forge foundation, a brick wall one course in width was discovered. The foundation was not continuous, but had been cut through by the modern steam line and unidentified disturbance west of the forge. ^{[(exploration trenches 1931) See Photo #N2506]} The wall was constructed of soft-fired, orangish-red brick mortared with oystershell mortar in English bond. The upper surviving course was laid in rowlock and may have been the level upon which the shop sill rested. An attempt had been made by the mason to base the wall on a firm footing for it extended 10" below the bottom of the forge. Though earlier Colonial Williamsburg trenching had destroyed any evidence of a builder's trench outside the wall, a narrow trench was found beneath the forge (ER 1963D); unfortunately, no artifacts to aid in dating the shop's construction were found in it. However, the wall's construction had cut through a layer of clinker (ER 1963C), thus providing important evidence that Anderson had been working on the site prior to the building of the wall. This clinker was probably waste from the main shop during Period II.

The West Wall

This foundation extends a full 17'6" from the northwest corner to the point where it was cut through by the builder's trench for the modern garage. The stretch of foundation leading to the intersection with the garage and Shop "A-1" was not 34. shown on the 1931 plan, though it was found covered with modern backfill up to the recent structure.

This western wall is constructed of the same orangish-red brick found in the northern wall, though laid in English bond and one and a half bricks in width (1'1"). The similarity of brick and mortar would seem to indicate a simultaneous construction date.

Though earlier excavations against the western face of the wall had removed the fill to the bottom of the brick, a small builder's trench was found on the inside (ER 1911P), but unfortunately, no datable artifacts were recovered from its fill. Several post holes were located beneath the wall, and as will be seen later (see p.67), they are associated with fence lines dating after 1740 and 1755. Another post hole (ER 1958G) did, however, contain a bottle neck with a string rim not likely to have been produced before ca.1770, thus providing a terminus post quem for the construction of the wall no earlier than ca.1770.

The Northwest Corner

While the northern and the western walls appear to have been constructed at the same time, there is evidence pointing both to the rebuilding and strengthening of the corner and to the partial underpinning of the west wall. The brick used in the corner was much harder than that of the earlier construction. In addition, the corner had been sunk deeper than either the north or west walls, indicating a need for greater strength in the corner of the building (Plate XXI). This could be expected as the structure was located on the edge of the silted 35. and filled ravine. It is probable therefore, that early in the life of this shop, the northwestern corner settled into the ravine's unstable fill and had to be rebuilt.

Possible Northern Extension

As noted earlier, the Frenchman's Map places the "forge shops" much closer to Duke of Gloucester Street than is indicated by the archaeological data. It should be mentioned, however, that the present excavations did reveal extremely slim evidence for a more northerly extension to the shop. This took the form of mortar adhering to the northern face of the northwestern corner, suggesting that other brickwork may once have abutted. No further indications were found for this extension, and the abundance of domestic and industrial waste located directly north of the shop wall weighs against any such structure having existed.

Eastern Shop Wall

No evidence was found for the eastern wall for Shop "A-3." It may be that the wall was robbed of its brick leaving only a rubble-filled wall trench which was removed by the 1931 excavations. Such a trench was found along the eastern side of Shop "A-2," and so may have been matched by one to the north.

Forge E

This forge abutted the interior face of Shop "A-3's" northern wall. Having been previously excavated and cut through by the modern steam line, no date for the forge's destruction could be arrived at archaeologically. The foundation was built with soft, reddish-orange brick similar to those used in Shop "A-3's" 36. first period construction, suggesting a similar date for the forge. However, the forge overlay the builder's trench for the forge. However, the forge overlay the builder's trench for the northern wall and thus post-dated it. ^{[But could have been built upon completion of the wall - and would have]}

The fire box itself had external dimensions of 5'1" by 6'00"; its east and south sides were one and a half bricks in width (1'1") while the north side was only 9", or one brick, wide. The west side, however, was a full two bricks wide (1'6"). This variation in wall thickness was common to all the excavated forges whose brickwork survived. It is reasonable to assume, therefore, that these thickness differences bore some relation to the above-ground character of the forges. We do not know how substantial were the chimneys needed for the Anderson forges, though it may be significant that Harwood built considerably more "forges" than he did "forge chimneys." It is conceivable, therefore, that these fire boxes had no massive, structure above them but were actually open on two or three sides, thus allowing greater accessibility for work on large items. ^{[Intense heat from the forge fires would necessitate regular rebuilding of forges, not the chimneys.]}

At the end of the 1931 excavations, Forge "E" had been left perched on a block of stratified fill which we hoped would now provide key dating evidence for its construction (see Figure 5). Such was not the case, however, as the latest artifact retrieved from this fill was creamware (ER 1963G), and which only served to put the forge's construction date after 1770. The most important find beneath Forge "E" was the location of the remains of Forge "A" (see pp.21-22). Between them lay two significant stratigraphic layers: Forge "E" had been constructed on a bed of yellow clay which contained fragments of shell associated with Miocene marl beds found throughout FIG. 5 - SECTION SHOWING VERTICAL RELATIONSHIP BETWEEN FORGES "A" AND "E" 37. the Williamsburg area (ER 1963F). Beneath this clay, a level of domestic ashes (ER 1963G) sealed the destruction of Forge "A", and probably was deposited during Period II before Shop "A-1" had been enlarged northward.

Fence D

Post holes defining a possible north-south fence line extending from the main house to the northeast corner of the shop were located and excavated (see Master Plan). These holes (ER 1957G, ER 1921Z, ER 1946Y, and ER 1910L) are located on eight foot centers, though an unexcavated gap conceals an area where an apparent 4'00" break (possibly a gate, see p.72) was located. The post mold for the last hole to the south (ER 1910L) would appear to have abutted the northeast corner of Shop "A-3" exactly 20'00" from the northwest corner.

It seems probable that this fence was added in order to maintain separation between the industrial work in the shops and domestic life in the buildings to the north. It can easily be seen as a replacement for Fence "C" (see p.31). The artifacts recovered from the fill in these holes contained only creamware as dating evidence, but do serve to associate the fence with the Anderson occupancy of the property.

Relation of Shop A-3 to Kitchen

If our assumption of a 20'00" width for this shop is correct, it becomes obvious that its eastern wall would have run directly behind and almost abutting the west side of the chimney stack for the kitchen. Nevertheless, meager archaeological evidence indicates that both structures were standing at the same time, and the Frenchman's Map corroborates this 38. by showing a small structure (probably the kitchen) to the east of the forge shops (see Figure 1).

Shop C

Coinciding with the construction of Shop "A-3," the small southern outbuilding (Shop "B") seems to have been dismantled and replaced by a larger shop containing two forges. Although the foundations for the latter structure ^{[(ie of the shop, not forge)]} had been robbed of most of their brick, the dimensions of the building can be estimated at approximately 25'00" by 16'00" (see Figure 4). The north wall of Shop "C" apparently abutted the southern wall of Shop "A-2," and perhaps shared it. Along the eastern side of Shop "C", however, there was a 5'00" setback between the two buildings, and it was perhaps this break in the line which explains the difference in alignment of the major "shop" blocks shown on the Frenchman's Map. The eastern foundation of Shop "C" was partially extant, and proved to be one brick or 9" wide, but as only one course survived, no bond could be determined. Constructed from a mixture of whole and broken bricks, the foundation was poorly secured with oystershell mortar. The footing was not set in a trench, but rather was constructed upon the previously mentioned iron waste deposit to the east of Shop "B" (ER 1975C, see p.30).

The south foundation was represented only by a 1'00" wide robber's trench (ER 1944G) which ran directly south of Forge "G." Although most of its length had been destroyed by earlier C.W. excavations, it was estimated to have extended to a distance of 16'3"′ As the west foundation was set much deeper, it seems most probable that this one had been stepped to a higher level, 39. as was demonstrated by both the southern and northern footings for the main shop structures (Shops "A-2" and "A-3").

Most of the west foundation had been robbed, and in turn the actual robber's trench (ER 1944Z and ER 1970B) was severely disturbed by a complex of later post holes along the western property line (see Figure 15). As with the eastern foundation, it is probable that the robber's trench had extended all the way to the southwest corner of Shop "A-2," though this juncture had since been destroyed by later disturbances.

There was evidence that a lean-to extension or overhanging roof projected from Shop "C" and was supported on a row of large posts. For details see p.44. ^{[Probably [illegible] & not in 1780]}

Aside from the fact that the eastern foundation for the shop was built on top of the iron waste, there is little dating evidence for its construction. It does seem likely, however, that it occurred around 1779 when Anderson's activities were rapidly expanding.

Forge F

Shop "C" contained two forges. Forge "F" was located in the center of the building, with its southern side abutting the line of the south wall of the by then dismantled Shop "B." Although all the forge's brick had been removed, its size (as delineated by the robber's fill) was 5'2" by 4'9", and it was flanked at all corners by small round pits (ER 1969B, ER 1975S, ER 1975R, and ER 1970S) which are believed to have supported tree stumps or some other firm base upon which an anvil would have been seated. As will be seen, this feature was common during the Period III construction.

Forge G

This was located against the robber trench for the shop's southern wall and was, for the most part, in good shape, having been disturbed only by the presence of two post holes which cut through two of its corners. The external dimensions for this forge were 5'3" by 4'6", though the varying width of the sides made the internal size considerably smaller, only 2'9" by 2'4". The north, south and east walls were all a brick and a half wide while the west wall was wider by a whole brick, again suggesting that it was the west side which supported the chimney or hood. Two small pits, similar to those flanking Forge "F," were found on the north side.

Little dating evidence for Forge "G" was forthcoming — beyond the fact that it had been built on top of and cutting through the remains of Forge "B" (see p.24).

Walk B

It seems probable that at some date during Period III, this brick walkway was laid along the eastern side of the main shop complex. Constructed from both whole and broken bricks, the walkway extended from the southern end of Shop "A-2" to a point near the junction of the first period-kitchen and Shop "A-3." The presence of paving in this location would indicate that access to the shops was from the south and east rather than from the north and Duke of Gloucester Street. For more data on this walk see p.65.

Clinker Spread

Located to the east of Shop "C" another distinct spread of coal dust and clinker was found (ER 1977F). This waste must 41. have been deposited during the life of the shop because the clinker was confined to an area outside its wall. It covered an area measuring at least 17'4" by 3'6", its full extent being undetermined due to incomplete excavation.

Forge H

During the 1931 excavations, a large brick feature was unearthed at the southern end of Shop "A-3," where it joined with Shop "A-1." Although nothing of this feature survived to be re-assessed in 1975, we may reasonably deduce that it represented yet another forge. This box was quite different from the other excavated forges, however, in that it was found to measure 8'4" long by 4'2" wide with an east-west orientation. The north and south walls were 9" (one brick′) wide, while the east and west sides were 1'1" (a brick and a half′) in width.

Projecting from the south side of this structure were two brick appendages, 1'1" wide and 3'11/2" long. As the brick used in the construction of these "legs" are shown on the 1931 plan as being the same as that used in the box, we may assume 42. that the two were related. Although there is no way of knowing exactly what function these protrusions served, it seems possible that they have some relation to the superstructure of the fire box. If this was, in fact, a forge, its different size and shape suggests a very specialized use — perhaps to work on large pieces which could be laid across its long axis.

Forge J

Of this only very fragmentary evidence was found. Located directly to the south of the modern garage was an area of brick rubble mixed with coal clinker and ash (ER 1973K). Removal of the fill revealed several bricks mortared together which helped to define the shape and size of the structure. It seems that the box had measured 6'00" by approximately 4'00", though the northern edge had been destroyed by the builder's trench for the construction of the garage and the west edge by earlier archaeological digging. There was no way to determine the thickness of three of the sides though the south wall was a brick and a half (1'1") wide, and from such scant remains, no dating clues could be drawn.

Working Level in Shop A-2

A level of black coal dust and ash (ER 1973P) extended between Forges "D" and "J." This fill contained enormous quantities of scrap iron and other waste materials which had apparently fallen onto the floor of the shop during its working life. The ashy fill was approximately 2-3" deep and was associated with the use of both Forges "D" and "J." It should be noted that this level sealed the large "barrel" holes (see pp.27-28) from the Period II construction, thus giving the 43. evidence needed to separate the life of Forge "D" into two periods. ^{[Not necessarily so, as the clinker fill over the robbed remains of forge D could have been waste from its own operation, used as backfill after the dismantling. See pp. 27-28]}

Bellows Support Holes (′)

A configuration of four post holes was found cutting through the ash floor level (preceding page) of Shop "A-2" to the east of Forges "D" and "J" (see Figure 4). From the appearance of these holes (ER 1973E, ER 1973F, ER 1973H, and ER 1973C), they had held posts driven into the ground rather than being set into previously dug holes, and therefore should more correctly be identified as post molds. Larger disturbances around these molds (ER 1973D and ER 1973G) apparently represented holes dug in order to remove the-driven posts, as the fill found in both molds and holes was identical. ^{[NOT - opposition not correct. all of the fill would have been the same.]} The placement of the posts formed a square measuring 2'00" on each side, and its proximity to both Forges "D" and "J" (2'00" from either of them) indicates that it supported a structure which bore a direct relationship to the operation of the forges. One distinct possibility would be that the posts held the frame for a bellows connected to both fires. Alternatively, the holes may represent the location of a heavy work table whose legs were driven into the ground for greater stability.

One of the post removal holes (ER 1973C) contained artifacts dating after 1782, indicating that this structure was dismantled no earlier. It is reasonable to suppose that the shop was abandoned at much the same time. ^{[Shop was in use through 1798][how about forge only′]}

Anvil Stands (′)

At three of Forge "D's" corners were found roughly circular holes averaging 1'6" in diameter and approximately 3-5" in depth. The fill in these holes (ER 1917A, ER 1972C, and ER 1973M) contained fragments of discarded iron waste and cut through the remains of the ash-strewn working floor of Shop "A-2." There was some indication of a fourth hole adjacent to the southeast corner of the forge (ER 1918C), though its exact size and shape was not well defined. Present thinking has it that these holes were seatings for a heavy base, such as a tree stump, upon which the smith's anvil would be placed. Although it might not be necessary to set such stumps into the ground, there is no denying that added support would have been provided. The presence of these holes on the north and south sides of the forge indicates that it was worked from at least two directions.

Possible Lean-To Addition to Shop C

Four post holes (ER 19771, ER 1975H, ER 1975K, and ER 1975M) were found to the east of Shop "C." The line of these holes coincide with the eastern wall of the main shop building (Shop "A")^{[No! does not]} and it is believed that they supported a small shed addition or lean-to roof which was probably added at a date after the original construction of Shop "C." As such an addition would have given the row of shops a common eastern wall ^[wrong] rather than an offset, it seems likely that the posts were added after the Frenchman's Map was drawn in ca l781.

Destruction Level

^{[The post holes are 1-2 ft outside the east bldg line would have created an offset in just the opposite direction from what the Frenchman drew]}

Overlying the ash working surface in Shop "A-2" was found a layer of brown loam and brick fragments (ER 1973A and ER 1973B), 45. probably representing the remains of the post-abandonment debris. In general, the artifacts recovered from this fill can only be said to date after about 1770, though one small sherd of whiteware dated the layer (ER 1973A) after 1830. However, the proximity of Colonial Williamsburg disturbances raises the possibility that this sherd may represent a contamination from a later level. The debris also sealed the removal holes for the bellows support which, as demonstrated earlier (p.43), dated after 1782. ^{[illegible handwritten note]}

Beam Slot

Cutting through the destruction level was a rectangular slot containing mixed loam fill (ER 1972A). This slot ran east-west, directly south of Forge "J," from the western wall to the eastern wall of Shop "A-2." The digging of this trench had cut through the backfill of the large hole associated with Forge "D" in Period II (ER 1973A), thus post-dating it. Were it not for the fact that the "beam slot" cut through Shop "A-2's" destruction debris, a case could be made for its once having held a timber which aided in supporting the shop roof by running studs from beam to rafters, forming a partition wall.

Brick Drain Tunnel B

An interesting feature of the kitchen was the location of an entrance to a brick drain situated inside the northwest corner of the building (Figure 7, inset). Careful examination of the construction of this drain in relation to the kitchen's west wall revealed that the drain was added later at a date as yet unknown. All indications pointed to the northern leg of the western kitchen foundation being dismantled, except for the bottom course, and then replaced as underpinning after the construction of the drain. Furthermore, it was the drain which provided the needed evidence to establish that brick and a half (1'1") thickness of the kitchen's northern wall. The drain did not enter the kitchen directly from the north but instead had a flow line oriented slightly to the northwest. 53. The actual arched section of the drain's tunnel, however, ran straight east-west. This caused the drain entrance within the kitchen to be irregular in shape, but at the same time provided evidence for the width of the kitchen's wall. When it was added, the arch of the drain itself was used as an underpinning for the kitchen's sill.

Further data on this and associated other drains will be presented later in the report (p.56). Suffice it to say at present that it seems that this drain was added sometime late in the eighteenth century.

Related Walkway

To the north of the kitchen a section of a brick path (Walk "A", see p.64) was uncovered. This walk was laid over a shallow erosion gully which contained fragments of Rouen faience in its silty fill, a ceramic type not found in the colonies prior to ca.1775, and which thus helps date the construction of the walk. This path may well have been added in Period III and be contemporary with Fence "D" through which it most probably passed (see p.37).

Evidence of Kitchen Removed

No accurate date for the kitchen's destruction was obtained archaeologically. None of the remnants of the robber's trench contained datable artifacts and, as mentioned above, the entire area was riddled with disturbances. Possible evidence of moving, dismantling, or interior renovations was provided by a layer of plaster debris spreading north of the building (ER 1928N). Associated artifacts date this layer to post 1782, thus raising the possibility that Anderson moved 54. the kitchen from the center of the yard to its present location after his return from Richmond (and also after the drawing of the Frenchman's Map). Support for this interpretation is provided by the close similarity between the foundation and chimney base measurements of both buildings. It is possible that Anderson moved the kitchen in order to expand his shops, though as we have seen earlier, his shops seemed to have reached their northernmost extension prior to his sojourn in Richmond. The closest we are able to date the removal of this building then would be to a date between 1782 and 1810.

Summary

The evidence points to a frame kitchen structure with an external chimney having been constructed around 1760, and to a fence being erected across the southern side about 1777. Sometime after 1775 (probably long after), a drain was added to the kitchen, necessitating the underpinning of its western and northern walls. Sometime after 1782, and before 1810, the structure was replaced by a building of similar size along the eastern property line.

Box Drain

The only feature excavated during the 1975 season which had direct relation to the main dwelling fronting on Duke of Gloucester Street was a brick drain (Figure 8). This drain was located on the west side of the reconstructed James Anderson House and was apparently used to drain water from the basement of the colonial structure located on the west side of the property. The drain measures 1'8" wide at its base and was constructed in four courses. The bottom course, which served as the drain floor was built with bricks laid as headers (when viewed from the side). The two courses of the sides were laid as stretchers while the top was laid in rowlock bricks capping the two sides, thus creating a drain whose interior measured 7" by 8". The bricks were a light reddish-orange in color and FIG. 8 - PLAN AND SECTIONS OF BOX DRAIN 57. were bonded with a light gray shell mortar. Based upon the 4'6" section uncovered, the drain had a gradient drop of 5" in every ten feet.

Most of the 6'00" of fill overlying the drain had been disturbed, although it does not show on the 1931 plan. A large post hole was located beneath the drain (ER 1919D) and must have been associated with a fence line pre-dating the drain. A layer of mottled gray, sandy loam underlay the drain (ER 1919C) and contained a fragment of creamware, thus dating the drain's construction after 1770. This is very important in the light of the uncertain date for the large house, which is likely to have been constructed at the same time as the drain.

Drain Tunnel A

The main component of the drainage system installed on the site was a massive, arch-roofed brick tunnel which entered the site from the east and exited to the west. The entire length of this drain had been previously excavated, making it impossible to determine through which levels it had been cut, and thus it was impossible to date its construction.

The tunnel was well built, consisting of a brick floor approximately 3'2" wide; on either side a two course high wall was capped with an arched roof, one brick in thickness. At irregular intervals along the crown of the arch, the construction was broken by sections of brick which appeared to have been plugged into the drain, possibly serving much the same purpose as a closer in a brick wall. The internal dimensions 58. for the drain were 1'1" wide by 1'2" high. The drain was devoid of any fill, other than a few scattered chips of bone, brick, and one piece of burned pearlware.

At the eastern side of the yard, a brick box entrance to the drain was excavated (see Figure 9). The tunnel entered this box from the east and exited at a slight northwesterly angle to the west. At the base of the entrance was a single slab of cut sandstone. At the top another surface drain entered from the west and was built from another slab of sandstone resting upon the western edge of the drain entrance and flanked by two bricks. Unfortunately, all the surrounding fill had been disturbed and no further evidence of this drain was located.

The purpose of the drain's entrance would seem to have been to receive rainwater from the relatively level ground surface to the east of the yard. With the build-up of the fill within the ravine, the natural drainage (which normally would have prevented any serious problems) was cut off, perhaps causing flooding at the eastern side of the property. The addition of the drain, with an entrance located in the most heavily traveled area of the property, would have alleviated the problem.

Drain Tunnel B

As previously noted, another tunnel drain was excavated during the 1931 excavations, this beginning inside the First Period Kitchen and running northwest to intersect with Drain Tunnel "A." That intersection was not excavated during the current project, though the entrance to the drain was uncovered 59. in order to establish its relationship with the kitchen structure (see p.52).

Clearly this drain was not designed to control rainwater but was intended for the disposal of liquid wastes from the kitchen. Like the larger drain, however, no fill (other than that from the backfilling of the 1931 digs) was found in association with the tunnel; and no date for its construction, use or abandonment could be determined.

Brick used in the construction of both Drain Tunnels "A" and "B" was similar, as was the mortar. It is interesting to note, however, that two distinct types of mortar were used in the building of the drains. In both was found a mixture of a buff-colored, soft, shell mortar, containing large fragments of shell, and a much harder grayish mortar with very small pieces of shell; but no pattern could be ascertained to determine if the different mortars were used for special purposes. This similarity of brick and mortar points to a simultaneous date for the construction of the two tunnels.

Early Brick Drain (′)

One baffling feature uncovered during the year's excavation was a line of laid bricks, one brick wide and one course deep, over most of its length. This feature was previously exposed by the 1931 excavations and found to extend for a distance of thirty feet. The brickwork rests on subsoil directly south of the path of Drain Tunnel "A," and is cut through by (thus pre-dating) Drain Tunnel "B." Just west of this intersection, the brick course turns slightly to the northwest and 60. increases in height to a second course. At this point the brick reaches the eastern limits of the ravine fill and there overlies a layer of ashy fill (ER 1946Z) deposited after 1750 and thus providing a terminus post quem for the brickwork.

There is little likelihood that the brick was structural. Not only does the line angle west of Drain Tunnel "B," but the bricks are unmortared. The only reasonable explanation is that the brickwork is all that remains of an early drainage system for the eastern side of the yard. This idea may be supported by the existence of an erosion gully just south of the brick which dates from the same period and could have led to the installation of this feature.

Well A

While excavating fill from the southern side of the brick drain entrance, a large quantity of rubble lying in an apparent hole was discovered. As this area was expanded, a disturbance 61. approximately eight feet square was found abutting the south side of the drain (see Figure 9). Removal of accumulated soil levels at the surface of this disturbance unveiled a concentration of brick rubble covering the square area. As the top of this hole was already below the level of the subsoil, it was surmised that it was the upper extremity of a backfilled well shaft. Removal of the top loose rubble revealed large sections of a mortared brick lining constructed of "compass" bricks along with three large pieces of the well head. The condition of this lining was deplorable at best, having been likened to the effects of a bomb thrown into the shaft. Entire mortared sections of the lining were oriented at all different angles, with no indication of where the actual shaft was located.

There was evidence that some sort of repair work had been performed on the well head, whose orientation had apparently been shifted slightly, for in the southwest corner there was evidence of two periods of construction (Figure 9, inset).

The continued removal of the rubble fill (which necessitated the dismantling of large chunks of the lining and well head) led to a very startling and unexpected find. At a depth of approximately seven feet below modern grade, all traces of the brick ceased, leaving only an area of dirty mixed clay. It was decided that continued excavation would be dangerous until new plans could be made for the protection of workers digging in the hole. Its exploration had not been resumed at the season's end.

Excavation of the stratified fill around the surface of the well revealed a level of spread marl which may have served 62. as a paving adjacent to the shaft. It is also conceivable that this marl was derived from the excavation of the well itself. As artifacts retrieved from the marl date after 1805, they might also provide a terminus post quem for the digging of the shaft. Such dating is reasonable in the light of the presumed date for the construction or moving in of the second period kitchen, i.e., ca.1806. If, as was stated previously, the earlier well (beneath the kitchen) was used until that time, then immediately upon the filling of the shaft there would have been a need for the sinking of a new well. It may be noted in Robert Anderson's 1812 account books an entry " ... work for well..." at Nancy Camp's. Was this the reorientation of the well head mentioned above′ It seems reasonable to assume that it was, thus putting this shaft in use both prior to 1812 and for some time thereafter.

When and why was the well abandoned′ The condition of the shaft would indicate that it collapsed, rather than being purposefully abandoned and backfilled. Artifacts in the rubble fill dated post-ca.1810 (ER 1926M), though as indicated above, it seems most unlikely that the shaft was filled that early. Some of the overlying fill contained artifacts manufactured as late as ca.1830 (ER1926D) which would make better sense as a destruction date. This fill, however, was most likely added to the shaft after the original backfill had settled and compressed. On the basis of the small amount of information presently available (and without complete excavation), a destruction date between 1810 and 1840 seems reasonable.

Figure 9 - Plan Showing Relationship Between Well "A" and Drain Tunnel "A"

Well B

In the center of the First Period Kitchen foundation was found the brick-lined remains of a modern well (ER 1940). The shaft was lined to a depth of approximately 6'00" with machine-made brick neatly laid as stretchers (Figure 10). Below that level, the quality of the brickwork was poor with the brick laid as headers to a depth of 12'01" and as stretchers again below 19'06". The diameter of the shaft was 3'09" at the top, tapering to 2'02-½" at a depth of 20'00". The hole was not circular, but was instead, roughly oval in shape. The side had partially collapsed, forming a hole beginning at approximately 12'00" and continuing to a depth of 17'00" (not on section).

Fill within the shaft was virtually the same from the top to the point at which the excavation was abandoned, and comprised dark loam mixed with large quantities of trash. The artifacts included a great number of modern bottles which dated the backfilling of the shaft to ca.1928-1930, along with cups and saucers and related ceramics suggesting that this was all refuse from the cafe which stood on the Anderson House site in the 1920s. Due to the fill's late date, it was decided to abandon excavation at a depth of 23'00" below modern grade.

A large hole 6'00" by 5'00" had been dug con the construction of the shaft. The brick itself had been laid against the west side of this hole which tapered inwards until, at a depth of 7'00", it was a mere 06" to a foot wide around the east side of the shaft.

Walk A

Just north of the First Period Kitchen a short section of a brickbat walkway was uncovered (Figures 7 & 17). This paving almost certainly was found during the 1931 excavations (there was Colonial Williamsburg backfill above it), but for some reason it does not show on the plan drawn at that time. The walk was 2'3" wide and survived for a distance of only 2'6", oriented in a generally northwesterly direction from the center of the kitchen's north wall. The walk had been laid over a shallow silted erosion gully, the fill comprising washed sand containing artifacts dating after 1775, thus providing a terminus post quem for the construction of the walk.

There is little doubt that this path served the kitchen at its northern side and led to a doorway in the center of the wall line. Much more difficult to determine is the location of the path's northern end. It can hardly have led to the main dwelling as it skirted too far to the west. However, no structures dating to the fourth quarter of the eighteenth century were found in the direction the walk leads. It is possible that there was a gate in the north-south fence line which ran from the northeast corner of Shop "A-3" to the main dwelling house (Fence "D", see p.72). If such a gate did exist, Walk "A" would run right to it, thus providing access to the ravine.

Walk B

A brick walkway was uncovered during the 1931 excavations along the east side of Anderson's shops on the western side of the property (Figure 4). This walk was partially re-excavated in 1975 revealing its width in an area to the east of Shop "A-2." Haphazardly laid with a variety of whole and broken bricks, the walk had a width of 2'6" and was mortared together with oystershell mortar.

The 1931 plan indicates that this path extended from the southeast corner of Shop "A-2" northward to a point just short of the First Period Kitchen's chimney base. Two conclusions can therefore be drawn: 1) Shops "A-1," "A-2," and "A-3" were probably in existence and represented one long structure when this walk was laid, and 2) the approach to these shops was from the south, with Walk "B" terminating in the recess formed by the kitchen's west wall and Shop "A-l." It follows therefore that if the walk was laid after the construction of Shop "A-3," it belongs to Period III and to a date no earlier than ca.1779.

Walk C

During the 1939 excavations on the eastern side of the property, a section of a laid brick walkway was found extending westward from the Second Period Kitchen (Figure 17). It was at that time assumed to connect the kitchen to the forge shops; though, as has been demonstrated, this kitchen almost certainly post-dated their destruction. After the completion of the 1939 project, the brick from this walk was removed leaving only a spread of brick and mortar (ER 1925E) to mark its previous 66. location. Although artifacts from rubble fill dated after 1790, the layers beneath provide a clearer picture of the walkway's true date. A level of grayish sandy fill (ER 1925H) was found directly beneath the rubble and yielded artifacts dating after ca.1800. The most significant layer, however, was a stratum of marl, probably associated with the digging or use of Well "A" (see p. 60), and which contained artifacts manufactured no earlier than ca.1805. Thus, the walk can be dated after 1805 and may well have been contemporary with the construction of the kitchen, ca.1806-1810.

Walk D

The rotted remains of a boardwalk running in a north-south direction across the site were found during the 1975 digging (Figure 17). This boardwalk was constructed of horizontally laid cypress planks nailed to runners which kept them off the ground. The west side of the walk was found to be lined with broken bricks, but because the east side remained hidden in an unexcavated balk, there was no knowing whether similar broken bricks flanked it on that side. The remains of the walk were greatly disturbed, leaving only very slim evidence for its date of construction. One section was sealed by a level which dated after 1830 (ER 1929E), while another was laid over a stratum which dated after 1805 (ER 1928H), thus suggesting a bracket of 1805-1830 for the life of the walk.

Where this walk led is not clear, though one of the possible privy pits (ER 1945G) located south of the First Period Kitchen (see p.74) lies just west of the walk and was filled after 1810. It is suggested, therefore, that this boarded 67. walkway provided access to the privy across the muddy yard area during a period when the First Period Kitchen was being removed and the Second Period Kitchen made ready. Belief in the temporary nature of the boardwalk is supported by evidence provided by a section of Walk "C" which cut into the remains of the boardwalk (ER 1928E cuts ER 1928F). As has been demonstrated above, Walk "C" almost certainly was constructed at the same time as the second period kitchen; thus, it would appear that the boardwalk was only in use for a very short time.

Western Property Line

A great number of post holes were revealed along the western property line representing the boundary from the early eighteenth century up to modern times. As most of the holes cut through one another and represented so many fences, only the two earliest lines are covered in this report.

Fence A

The first period fence on the western property line consisted of large round holes, usually with sloping sides, with the posts located on 8'00" centers (Figure 17). In a few of these holes the remains of the post molds were located and found to have irregular shapes, somewhat resembling a log split 68. into thirds. ^{[Would not fourths seem more likely considering the difficulty involved in splitting of an odd number of sections′]} The large clay-filled holes were partially sealed by a level of dumped yellow clay which apparently dated after ca.1750. This terminus post quem is substantiated by the artifacts recovered from the holes' fill which point to a construction date of post ca.1735-1740. This is interesting because there is no indication of an earlier fence line. One of the stipulations for ownership of property in the early days of Williamsburg's history was to enclose the "Lotts or half Acres with a Wall Pales or Post and Rails within six Months after y^e Building which y^e Law requires."³⁵ From the archaeological evidence, it would appear that Ravenscroft did not fulfill that obligation. The possibility should not be ruled out, however, that he merely enclosed the lot with a simple split rail fence which would not have left any archaeological evidence. As has been demonstrated in both the history and archaeology, Lot 18 enjoyed no recorded activity prior to at least ca.1735-1740, and therefore, such a rudimentary fence may well have sufficed. With the construction of the shop on the eastern side of the lot and the smokehouse to the west (prior to 1755), there became a need or desire to improve this property line, hence the construction of the post fence. In fact, the split log vertical posts may have been re-used from the original fence.

At any rate, this fence extended the length of the western property line in all areas investigated, and it can be assumed that it stretched from Duke of Gloucester Street to Francis Street. The holes represent the earliest features excavated during the year and serve as further evidence for a lack of cultural activity early in the eighteenth century.

Fence B

All along the length of Fence "A," a series of rectangular holes were found cutting into round post holes. These later holes delineated the second period fence line along the western property line (Figure 17). Also located on 8'00" centers, they contained the typical mixed fill found in colonial post holes and yielded artifacts offering a terminus post quem no earlier than 1755. This construction date for the fence fits well into the pattern for the beginning of cultural activity on the site. It is probable, therefore, that simultaneously with the erection of the First Period Kitchen, the fence line was also replaced. These holes cut through (thus post-dating) the dumped clay fill that sealed the Fence "A" holes. At the same time these Fence "B" holes were dug, the ravine was just beginning to accumulate fill, and thus, all the holes are sealed by domestic waste dating after 1755.

Garden Fence

During the same period that Pence "B" was impaling the western property line, another was constructed to the east of it. Post holes located on 8'00" centers were found extending 62'00" in a north-south direction on a line 10'00" to the east of Fence "B" (Figure 17). This fence began at a point inside the northern end of the modern garage and extended to a point just south of the present east-west property boundary at the rear of the lot. At the southern end of this line of holes there is only a 5'00" gap separating the last two holes, then an apparent turn to the east and a gap of 4'00" to the first hole in that direction.

Although this line was definitely established on its western side, limited excavation along its possible southern and northern ends prevented positive identification of associated holes. All those located were sealed beneath the level of gray sandy loam, deposited from ca.1760 to ca.1775, and so must pre-date this deposition.

It is suggested that the fence created a rectangular enclosure within the yard area, measuring perhaps 67'00" by 62'00". Not until further excavations have been undertaken can the true nature of this fence be determined, though it could well have protected a garden area from foraging animals. It seems, therefore, that the fence was in use after Lots 18 and 19 were divided in 1760 and most probably existed throughout the following decade.

At a point 22'00" from the apparent southern end of this fence, two small post holes (ER 1985C and ER 1972X) were found in the "alley" between the western fence and the impaled "garden." Although it is not known what these holes represent, it is possible that they held a gate separating the northern end of the property from the southern end where there was a privy during this time period (Privy "B", see p.74).

Later Fence Lines along the Western Property Boundary

The next innovation along the western property line was the construction and evolution of James Anderson's shops. The western wall ran directly down the fence line and several of the holes for Fences "A" and "B" were found beneath it. All later fence lines ran just west of the wall, and as mentioned above, presented such confusion of intersecting holes that 71. it became virtually impossible to sort them into meaningful sequences. For that reason, the holes have been deleted from the overall plan of the site, but are shown in Figure 15 with a list of their terminus post quem dates.

Eastern Property Line

Only one area was tested during the season which was located along the property boundary on the eastern side of the lot. Although several post holes were uncovered and excavated, it was not possible to positively delineate a particular period of construction.

Fence C

A series of post holes was found flanking the southern side of the First Period Kitchen (Figures 3, 7, and 17). The posts for this fence were located on 8'00" centers and crossed the site east-west. Artifacts found in the post holes dated the construction of the fence line to post-1770, and therefore after Anderson's arrival on the site. The fact that this fence was added during Anderson's ownership suggests that he may have wished to separate the lot between the domestic activities to the north and the industrial activities to the south. As brought out earlier in this report (p.25), it would appear that this fence was associated with the 2nd period of Anderson's shop evolution, thus dating it to ca. 1777. It does not appear to have been in use for long, as the construction of Shop "A-3" cuts across the western extension of the fence and was probably moved to a new location.

^{[Are we to assume the entire fence was abandoned because the portion west of the Per. I Kitchen was made obsolete by the construction of Shop A-3′ The sect. east of the kitchen ... cut-off]}

Fence D

This may well represent the reorientation of the east-west fence line (Figure 17). It was delineated by the location of 4 post holes running in a line from the main dwelling to the northeast corner of Shop "A-3." Artifacts date the erection of this fence to post-1770, though its probable relation with the Period III construction of Shop "A-3" would date it around 1779.

Although the four post holes did line up, the distances between them were not constant. Between ER 1921M and ER 1921Z the distance is 8'00". There is then a gap of unexcavated fill between this last hole and ER 1946Y, followed by another 8'00" gap to the post mold for hole ER 1910P. It may be conjectured, therefore, that the posts for this fence were located on 8'00" centers, possibly with a gate located in the span which was not excavated (due to the knowledge of the entire area having been removed during the 1931 excavations). A gate would account for the 4'00" discrepancy noted in the unexcavated gap. Though this gate is purely conjecture, it should be noted that if Walk "A" (p. 64) is extended along its northwesterly axis it would pass directly through this presumed break in the fence. As we have seen earlier, this walk was dated to post-1775 and could therefore have been in use at the time the fence was erected.

Fence E

Apparently sometime after Anderson's death and the removal of his shops from the site, an east-west fence was constructed across the site. The holes which possibly delineate this 73. fence are shown on Figure 17, though due to the incompleteness of the excavation along this presumed fence line, it was impossible to determine which of the holes were related. As mentioned earlier (p.11), Nancy Camp apparently constructed a fence in 1812 and it may be reasonable to assume that Fence "E" represents the one erected at that time.

Other Post Holes

In several areas of the site (most notably, ER 1929, ER 1930, and ER 1932) the masses of intersecting holes, often further confused by modern disturbances, were impossible to interpret, though a few general trends could be seen (Figure 16).

An apparent east-west fence was located along the southern wall of the First Period Kitchen. Artifacts recovered from several of these holes dated later than 1782. It may be reasonable to assume that after the kitchen was relocated (ca.1806), a fence was constructed across the yard to intersect with the southern end of the new kitchen's chimney base. Presumably, this fence was later replaced by Fence "E".

There are some indications of a possible north-south fence in ER 1930, though no other areas were investigated in the line of these holes. It may date from the late 19th century when the lot was divided into the Canaday and Moseley properties.

Figure 16 shows the conglomeration of post holes located throughout this area of the yard.

Privy A

One of the earliest features located on the site was a circular depression filled with mixed loam (ER 1985B) found beneath the western wall of Shop "A-2." The pit itself was 74. not regularly shaped and only 6" in depth, which may cast some valid doubt on its being interpreted as a privy. The fill contained many fragments of a Westerwald chamber pot, however, which would possibly indicate some functional use.

The pit was sealed by a deposit of dumped yellow clay which was found to date post-ca.1750 (ER 1972V) and thus falls within the first period of known occupation on the site. No structural evidence in the form of post holes, brick footings or soil stains was found in conjunction with this pit.

Privy B

Located at the southern end of the alley between Fence "B" and the Garden Fence was found a rectangular pit, 4' by 3'2" and 8" in depth (Figure 17). The pit was sealed by a level of gray sandy loam (ER 1970Z) which dated post-1760. The southern wall of Shop "B" had been constructed on top of the pit and settling of the fill may have caused the partial collapse of the wall as noted on page 23.

Fill within the pit was stratified into two levels of gray sticky loam separated by a level of clay. The upper level (ER 1981K) contained numerous artifacts including fragments of Whieldon pineapple ware, providing a post-1760 date for the filling of this feature. The association of the Garden Fence and Fence "B" with the pit indicates that it was in use during the decade 1760-1770.

As with Privy "A", no remains were found to indicate a structure above this pit. It is possible, however, that such a structure could have been built on sills resting on the ground and would not have left any remnants.

Privies C, D, and E

South of the First Period Kitchen, three large rectangular pits were located (ER 1945GA, ER 1945N, and ER 1945T) (Figure 17). These three features varied in size. Privy "C" (ER 1945T) was 5'5" by 3' by 1'3" deep. The presence of creamware in this hole dated it after 1770. The pit was located stratigraphically near the bottom of the undifferentiated soil found south of the kitchen, sealed by ER 1945S which also post-dated Anderson's purchase of the property.

Privy "D" (ER 1945N) cut through the brown soil level which sealed Privy "C" and measured 2'00" by 3'00" by 10" deep. Although the artifacts from the pit (Buckley ware, delft) dated its filling after ca.1745, the fact that it cut through the post-1770 level places it temporally after the latter date.

Privy "E" (ER 1945G) was found cutting into the fill of Privy "C" and later levels of brown loam (ER 1945H). The pit measured 2'4" by 4'6" and was 1'3" deep. The mixed clay fill in this feature contained fragments of a type of pearlware not produced prior to ca.1810 thus dating this privy as contemporary with Walk "D" (see p.66) which probably gave access to it.

None of these pits were associated with any sort of structural remains. It is suggested, however, that as with both Privies "A" and "B", such buildings may not have been substantial enough to have left remains in the ground. It is believed that privies were considered temporary and that the building was actually moved from one location to another as the pit was filled. The fact that 4 of the 5 pits located 76. contained mixed clay rather than organic debris, may throw some doubt on their interpretation as privies. Such features have been found elsewhere in Williamsburg (e.g. James Geddy House) where their identification as privies was more firmly established.

Main Fill

The stratified fill covering the entire site can be conveniently broken into four major areas of deposition. As has been seen earlier in both the historical and archaeological record, there is little evidence of activity on the site prior to ca.1755-60. This is borne out by the nature of the various levels of occupation encountered.

Southern Ravine Fill

For the purposes of this report, the area extending southwards from Forge "E" is considered the southern area of the ravine. This distinction is based upon a radical difference in the nature of the fill found to the north and south of this forge. Before cultural activity commenced on the site, the ravine encompassed a much smaller section of the lot to the south and its depth was much less than that to the north (Figure 13).

At the very bottom of the fill in the ravine a level of mottled grayish sandy loam was found. This level, varying in thickness from 1"-3" was not continuous but was cut through in many areas by shallow erosion gullies leading from the flat area of the yard into the ravine. It seems most probable that this soil level represents a buried topsoil layer which covered the ravine at the time of the initial occupation of the site. Above this soil, a level of dumped yellow clay was 77. found. The origin of this clay is unknown, though it was found extending the entire length of the ravine possibly added to the ravine in an attempt to smooth out the erosion gullies previously mentioned. A majority of the areas of this clay contained artifacts indicating a terminus post quem of post-ca.1740, though several groups yielded finds pushing this deposition date as late as ca.1750.

One important aspect of this clay was the fact that it was found extending on top of (thus post-dating) the large round post holes from Fence "A". Further, the clay was cut through by (so pre-dating) the Fence "B" post holes, making it clear that this clay dump was associated with the earliest cultural activity on the site.

Directly above the clay, a level of brown loam with brick chips was found, extending as far north as the area beneath Forge "E" (in fact, it was found beneath Forge "A"—Figure 5, ER 1963N). Although artifacts in this soil level were sparse, one group (ER 1911F) yielded a deposition date as late as ca.1760. Unlike the clay fill below, it seems more likely that this level, rather than having been intentionally dumped, was, in fact, the result of a gradual buildup of soil, perhaps washing down from the southern area of the flat yard. This level was also cut through by the post holes for Fence "B", and, more importantly, by the pit associated with Privy "B" (ER 1981K). All indications point to this level having been fully deposited by ca.1760, after which we have a gradual build-up of yet another stratum.

Above this level, and found only in the areas south of the garage, was a layer of gray sandy loam. It seems that this level was another gradual deposit which accumulated between ca.1760 and 1777 when Anderson constructed Shop "A-2." It is also found beneath the southern wall of Shop "B", which we have shown was probably constructed by Anderson shortly after his arrival on the site. The upper inch or so of this level contained fragments of creamware. It is possible, indeed likely, that these sherds were ground into this soil by the increased activity in this area of the lot after Anderson commenced using it as a shop.

Northern Ravine Fill

At the lowest levels of the fill in the northern end of the ravine, the same sequence was found as that to the south; this is, the buried topsoil, covered by the dumped yellow clay. A distinct difference in the nature of the fill is noted above the yellow clay. As can be seen in Figure 11, a thick accumulation of domestic waste was found above the clay. Even the lowest of these levels seal the post holes for Fence "B", thus helping to substantiate the theory that these levels were not deposited until ca.1755-60. It is also clear that this level post-dates the deposition of the brown sandy loam to the south which we have noted was cut through by the Fence "B" post holes.

For the most part, the domestic fill found in the ravine directly to the north of Forge "E" was undifferentiated and was excavated in arbitrary 3" levels. This technique helped to establish that the deposition had occurred over an extended period of time, perhaps lasting until the construction of Shop "A-3," ca.1779. 79. Some variation was noted horizontally in the nature of these deposits. In the excavated squares nearest to the reconstructed house (ER 1921 and ER 1957) there was none of the primary clay deposits noted elsewhere in the ravine. The predominant nature of the fill from the subsoil upwards was of a type associated with water-washed silty sand. Artifacts in these levels had been broken into small pieces and were best retrieved by screening. Pipe stems from these levels date to the early l760's and therefore probably within the period of Christiana Campbell's proprietorship. Moving to the south, however, the fill became gradually less sandy and more ash filled, possibly representing the succession of dumping of fireplace ash into the ravine. It is not hard to understand this sequence. The general flow of the ravine was to the northwest and, as it appears that a majority of the trash was dumped from the First Period Kitchen, rather than from the house, this natural drainage pattern would have washed the silty fill, along with the smaller fragments of the artifacts, towards the lower areas to the northwest.

When Anderson began his blacksmithing operations on the site, he, too, made use of the ravine to the north of his shops for his industrial waste. Stratigraphically, this is the most obvious change noted. Although clinker was found almost everywhere on the site, its concentration in the area concerned was so great that there can be little doubt that this was Anderson's primary refuse dump.

Fill dumped into the ravine in the Anderson period raised the grade to much the same as it is at present. No exact date for this final filling is available from the archaeological record, as the clinker levels contained few datable artifacts other than a few scattered sherds of creamware. It may be reasonable to assume, however, that the ravine was filled by the late 1780's or early 1790's. Apparently, this change in the site's topography caused many drainage problems, necessitating the installation of the brick drain (p. 57).

Yard Levels North of the First Period Kitchen

The area along the east side of the property was not affected by the topography of the ravine and was found to be relatively flat at the natural level, thus likely to be the most used portion of the lot (Figure 12). In fact, it was discovered that cultural activity had left its mark on the stratigraphic record in all excavated areas to the east and north of the First Period Kitchen. It is interesting to note that even in this apparently choice spot for occupation, the archaeological remains point to virtually no activity prior to 1740 (ER 1923K) and probably even later than that. Although the 1975 excavations did not investigate the area directly south of the east wing of the house, which was probably the earliest structure on the site (possible dating to the 1730's or 1740's), it seems that much of Lot 18 was left in its natural state until the mid-eighteenth century.

Yard Levels to the South of the First Period Kitchen

In the two areas excavated south of the First Period Kitchen, further evidence was found to substantiate the theory that little activity took place on Lot 18 prior to ca.1755-60. 81. In fact, there was a profound difference in the fill south of the kitchen compared with that found to the north and east of it. This latter fill was well stratified into levels relating to different activities; whereas, the fill to the south was basically undifferentiated brown loam from top to the bottom. The only means of distinguishing different layers was by use of post holes on the assumption that holes cutting from the same level were of similar date. The basic dissimilarity in the fills on either side of the kitchen would point to a variation of the activities performed in those area. This difference may have been imposed by the construction of the Garden Fence, ca. 1760, and later by the erection of Fence "C", ca.1777. It is significant to note that both of the areas investigated, creamware was found in all levels down to the natural subsoil, thus indicating a lack of activity in the southern portion of the lot prior to ca. 1770.

Figure 13 - Plan of Erosion Gullies

SKELETON E.R.1942L-10.G, SEX (FIG.18)

On November 5, 1975, work was commenced in removing fill from the northernmost burial trench, E.R.1942L-10.G. In the process, the remains of a hexagonal coffin were delineated. From the extant configuration, it was assumed that the body had been interred with facial orientation to the east. Subsequently, fill removal was intensified at the western end of the coffin so the skull could be quickly exposed for analysis.

Osteologically, the skull yields fairly reliable information in the determination of sex. Based on the initial assumption that the average male is more robust than the average female, it is further assumed that all male cranial processes for the attachment of musculature will display a more rugged character. Two sexually diagnostic regions of the cranium are the mastoid processes (occurring bilaterally) and the external occipital protuberance.

The mastoid process, located behind the external auditory meatus (ear canal) on the temporal bone, serves as the insertion base for the sterno-cleido-mastoid muscle of the neck. Conical 103. in shape (viewed anteriorly), it is generally wider (viewed laterally) with deeper vertical projection in the male. Even without metric confirmation, the left mastoid process of E.R.1942L-10.G exhibited a rugosity strongly indicative of male physiognomy.

The external occipital protuberance, located on the squamous portion of the occipital bone, is a bulbous prominence for the attachment of the nuchal ligament. It is generally larger with greater horizontal projection in the male. All features on the occipital bone of E.R.1942L-10.G, including the external occipital protuberance and the inferior nuchal line, clearly demonstrated marked robusticity reinforcing the male sex determination.

Once the skeleton was completely exposed, additional regions were checked for corroborating sexual data. The general size of all long bones and the configuration of the pelvis confirmed that ER.1942L-10.G was an adult male.

E.R.1942L-10.G, RACE

From the outset, the assessment of race proved to be of major interpretive interest.

To facilitate diagnosis, the skull was removed to the laboratory for additional cleaning. Once completely exposed, the anterior aspect of the skull was observed to have undergone no major alteration from earth pressure, thus minimizing significant interpretive error. The face, which demonstrated a true narrow configuration, did not conform to Negroid dimensions. Similarly, the frontal bone did not exhibit any degree of recession, a feature which occurs quite frequently among Negroes. 104. The most important factor to repudiate the initial assumption that this was a Negro interment involved the absence of alveolar prognathism. Due to the lack of this most characteristic Negroid trait, it was necessary to conclude that E.R.1942L-10.G was the remains of an adult Caucasian male.

E.R.1942L-10.G, AGE

The assessment of individual age was determined from observations of dentition, epiphyseal union, and sutural obliteration.

Full permanent dentition was present in E.R.1942L-10.G, indicating a minimum individual age of 21 years at death. Observations of molar occlusal attrition provided an additional measure of age. By comparing the character of molar grinding surfaces with charts compiled by Brothwell,¹ the dentition of E.R.1942L-10.G was suggestive of a range between 25 and 35 years of age.

Epiphyseal union involves the full closure of bones of the post-cranial skeleton. By noting the complete ossification of certain bones, it is similarly possible to assign a minimum of individual age for such physiological involvement. All extant bone in E.R.1942L-10.G when compared with closure schemes posited by Genoves,² was indicative of an age range in excess of 25 years.

The last means of assessing age, sutural obliteration, has long been of controversy among osteologists. Like epiphyseal union, the scheme involves the closure at margins but refers only to those bones of the skull. Opponents of the technique maintain that marked cranial variation obviates any attempt to temporally classify material. Still others maintain that variation does not discount the viability of the technique as a general base. In support of the latter argument, cranial development in E.R.1942L-10.G exhibited no major variation. Diagnostic divisions of the coronal and sagittal sutures indicated a minimum individual age of 30 years.

By calculating an average from the above observations, we may estimate that the individual died at age 27.

E.R.1942L-10.G, STATURE

The living stature of an individual can be attained by applying certain long bones measurements to calculations posited by Trotter and Gleser.³ The equations, based on mathematical regression formulae, provide for distinction of both race and sex. Assuming that E.R.1942L-10.G was the remains of an adult white male, the following statural estimates were reached:

• 1) 1.26 (Max. length, cm., FEMUR + Max. length, cm., TIBIA) +67.09 = STATURE
  • 1.26 (48.4 + 38.1) + 67.09 = S
  • 1.26 (86.5) + 67.09 = S
  • 108.99 + 67.09 = S
  • 176.08 + S
• 106.
• 2)2.32 (Max. length, cm., FEMUR) + 65.53 = S
  • 2.32 (48.4) + 65.53 = S
  • 112.28 + 65.53 = S
  • 177.8 = S

These statural estimates when converted and averaged indicate a height approaching six feet. From the general size of all long bones, viz., femora and tibiae, we may further conclude that this individual possessed a massive weight to complement his stature.

E.R.1942L-10.G, PATHOLOGY

Pathological conditions in E.R.1942L-10.G were confined to the teeth. Both the upper and lower dentition were offset by slight alveolar resorption. This condition, indicative of periodontal disease, was most likely caused by the extensive deposition of calculus still active up to death.

The upper dentition demonstrated an interesting array of anomalies. In addition to an advanced occlusal caries of the right first molar, the left third molar exhibited impaction and the anterior teeth had rotated clockwise as a result of palatal crowding. The ante-mortem loss of both second premolars suggests the possibility of therapeutic removal for the provision of additional dental space. Whatever actions transpired before death, the condition of the mouth was most assuredly the cause of great physical pain.

The absence of certain pathological conditions reinforced the assessment of age. None of the vertebrae demonstrated any degree of osteophytosis. This process of intervertebral lipping accompanies the degenerative effects of age and arthritis. Clinically, arthritis first begins to complicate the vertebral 107. column at age 30. With no such conditions having been observed in E.R.1942L-10.G, the pre-30 age assessment thus seems entirely plausible.

There were no osteological indicators to explain the cause of death of the individual. From statural observations noted earlier, it might be possible to infer death was caused by physiological exertion leading to mio-cardial infarction. Historically, however, diseases and dysfunctions of the heart are conditions of technological man and seemingly did not plague our Colonial American predecessors. From extant medical accounts of Virginia in the eighteenth and nineteenth centuries, it is probable that death was incurred at the hand of one or more recurring diseases. Furthermore, these pathogenic diseases would leave no trace effects on the skeletal material.

E.R.1942L-10.G, BURIAL POSTURE

A major problem in the interpretation of E.R.1942L-10.G involved the posture of the skeleton. The displacement of some bone was initially suggestive of activity associated with secondary reburial. (see Figure 18)

Historically, there were two modes of secondary reburial. The first method entailed the exhumation and subsequent relocation of an intact coffin. The second procedure involved the removal of bone from an unstable or decomposed coffin with direct transfer back into the ground or replacement in a new coffin.

Albeit spatially disordered, all bones in E.R.1942L-10.G were located in the correct relative position. Unless the grave digger relegated to exhuming the original burial proceeded to extricate and relocate each bone separately in the new coffin, 108. collation is doubtful. This unlikely possibility thus obviates consideration of the latter proposal.

The first mode of secondary reburial, i.e., the exhumation and subsequent relocation of an intact coffin, similarly does not sufficiently explain the displacement of bone. The position of the vertebrae and bones of the left leg are suggestive of relocation during some aspect of coffin movement. Yet any movement sufficient to shift Vertebrae and the formidable bones of the lower leg should similarly have played havoc with the original position of all bones. The simple fact that no other bones exhibit relocation, especially the gracile bones of the left lower arm, repudiates the claim of secondary reburial.

Discounting secondary reburial, it becomes necessary to examine causative factors of bone displacement associated with primary inhumation. The majority of data, notwithstanding the activities of earth pressure or burrowing animals, suggests that the relocation of bone in E.R.1942L-10.-G was effected by induced burial posture.

The foot end of the coffin measured 9-1/2" in width. This limitation in space, regarding the statural robusticity of the individual, would have prevented the normal parallel extension of the legs. The angle of the left tibia and fibula, and the fact that only the proximal ends of these bones are located beneath the right leg, strongly suggests that the left leg was flexed behind the right knee. The left downward angulation of the pelvis, as illustrated in the burial, would be the physiological concomitant of such directional leg flexion. Furthermore, the placement of vertebrae exhibiting both primary and secondary curves would similarly result from such pelvic angulation 109. The separation of the left femur from the left tibia and fibula is probably indicative of shift accompanying decomposition. To flex at the knees in the noted manner would probably entail raising the left side of the pelvis. Decomposition, first occurring at the knees, would result in disarticulation. With this subsequent reduction in the length of the leg, the pelvis could have been impelled to relocate in the prone position carrying the still articulated femur. Once the pelvis had come to rest, the femoral-acetabulum articulation would decompose, leaving the femur at its present illustrated location.

E.R.1942L-10.G, SUMMARY

E.R.1942L-10.G was the remains of a tall, robust, white male approaching 30 years of age. While pathological evidence did not explain the cause of death, we may assume that disease, a frequent visitor to the eighteenth and nineteenth century urban environments, was the culprit.