[Ill.u.s.tration: Ill.u.s.tration 79.--Wrought-iron colter from plow.

One-fourth. (USNM 60.88.)]

[Ill.u.s.tration: Ill.u.s.tration 80.--Hook used with wagon or oxcart gear.

One-half. (USNM 60.9.)]

[Ill.u.s.tration: Ill.u.s.tration 81.--Left, bolt with wingnut. One-half.

(USNM 60.145.)]

[Ill.u.s.tration: Ill.u.s.tration 82.--Right, lashing hook from cart or agricultural equipment. One-half. (USNM 59.2030.)]

For instance, the inventory lists 22 plows at Marlborough. Among the finds is an iron colter from a colonial plow in which the colter was suspended from the beam and locked into the top of the share (USNM 60.88, ill. 79). The colter is bent and torn from exhaustive use (Chapman, in 1731, fitted a plow "w^{th} Iron" for Mercer). From it we learn a good deal about the size of the plow on which it was used and the shallow depth of the furrows it made.

[Ill.u.s.tration: Figure 91.--FARM GEAR: a, part of collapsible-top fitting from carriage; b, chain, probably from whiffletree; c, part of bridle bit; d, iron stiffener from a saddle; e, worn chain link; f, base of handle of a currycomb; g, rivet and washer; h, piece of iron harness gear; i and j, two horseshoes; and k, chain to which a strap was attached--probably harness gear.]

Four chain traces were on the list, one of which is represented by a length of flat links attached to a triangular loop to which the leather portion of the traces was fastened (USNM 60.64, fig. 91b). The halves of two snaffle bits (USNM 59.2078, 60.67, fig. 91c; ill. 87) correspond to an item for eight "Bridle Bitts." (A "snafflebit" costing 1s. 8d. was among Mercer's purchases for 1743.) A third bit, crudely made of twisted wire attached to odd-sized rings, is a makeshift device probably dating from the 19th century. Three ox chains listed in the inventory are not distinctly in evidence in the artifacts, although a heavy hook, broken at the shank, is of the type used to fasten an ox chain to the yoke (USNM 60.9, ill. 80).

Archeological evidence of the two oxcarts and one wagon listed in the inventory is confined to nuts and bolts that might have been used on such vehicles. A long axle bolt (USNM 59.1802) measures 23 inches. A small bolt or staple, split at one end and threaded at the other, has a wingnut (USNM 60.145, ill. 81). A hook with a heavy, diamond-shaped backplate and a bolt hole was perhaps used on a wagon to secure lashing (USNM 59.2030, ill. 82). A heavy, curved piece of iron with a large hole, probably for a clevice pin, appears to be from the end of a wagon tongue, while a carefully made bolt with hand-hammered head (USNM 59.1821) and a short rivet with washer (USNM 59.1881, fig. 91g) in place seem also to be vehicle parts.

[Ill.u.s.tration: Ill.u.s.tration 83.--Hilling hoe. One-fourth. (USNM 59.1848.)]

[Ill.u.s.tration: Ill.u.s.tration 84.--Iron reinforcement strip from back of shovel handle. One-half. (USNM 59.1847.)]

The inventory listed four complete harnesses, the remains of which are probably to be found in four square iron buckles (USNM 59.1644, 59.1901, 60.131, fig. 91h), a bra.s.s ring (USNM 59.1678, fig. 83), and an ornamental bra.s.s boss (USNM 59.1878, fig. 83j).

Twelve "Swingle trees" (whippletree, whiffletree, singletree) are listed in the inventory. The artifacts include three iron loops or straps designed to be secured to the swingletrees. One (USNM 59.2042, fig. 91b) still has two large round links attached. (In 1731 Chapman fitted ironwork to a swingletree.)

Ten "Hillinghows," 17 "Weeding hows," and 8 "Grubbing hows" are listed.

In the long Chapman account for 1731 we see that Mercer then purchased "5 narrow hoes" and "2 grubbing hoes." The only archeological evidence of hoes is a fragmentary broad hoe (probably a hilling hoe) (USNM 59.1848, ill. 83) and the collar of another.

[Ill.u.s.tration: Ill.u.s.tration 85.--Half of sheep shears. One-half. (USNM 59.1734.)]

Thirteen axes are listed in the inventory. Again we find Nathaniel Chapman providing a "new axe" in 1731 for five shillings, while William Hunter sold Mercer "2 narrow axes" and "4 Axes" in 1743. One broken ax head occurs among the artifacts, worn back from repeated grinding and split at the eye (USNM 59.1740, fig. 89e).

There were four spades and an iron shovel at Marlborough in 1771. An iron reinforcement from a shovel handle occurred in the site (USNM 59.1847, ill. 84), while a slightly less curved strip of iron may have been attached to a spade handle (USNM 59.1662). Once more in Chapman's account we find evidence of local workmanship in an item for "1 Spade."

[Ill.u.s.tration: Ill.u.s.tration 86.--Animal trap. One-third. (USNM 59.1715.)]

Thirteen scythes were listed in 1771; perhaps the one excavated from the foundation of Structure H on Potomac Creek may have been among these (USNM 59.2400, fig. 90). There were eight sheep shears; half of a sheep shears was found in Structure G (USNM 59.1734, ill. 85). Of the other items on the list, a few, such as stock locks and hammers, have already been mentioned, while the remainder of the list is not matched by artifacts. An item for a chalk-line is supported by a piece of chalk (USNM 59.1683, fig. 84).

[Ill.u.s.tration: Ill.u.s.tration 87.--Iron bridle bit (see fig. 91c). Same size.]

[Ill.u.s.tration: Ill.u.s.tration 88.--Fishhook. One-half. (USNM 59.1681.)]

[Ill.u.s.tration: Ill.u.s.tration 89.--Bra.s.s strap handle (see fig. 83j). Same size. (USNM 59.1736.)]

A few specimens are not matched in the inventory. One is a springtrap of hand-forged, hand-riveted iron (USNM 59.1715, ill. 86) for catching animals. Another is a fishhook (USNM 59.1681, ill. 88), possibly one of 95 bought in 1744. An iron stiffener for the framework of a saddle is fitted with 10 rivets for securing the leather and upholstery (USNM 59.1847, fig. 91d). The third artifact is an elegantly designed bra.s.s fitting for a leather curtain or strap (USNM 59.1736, fig. 83j, ill.

89). It is fitted with a copper rivet at the stationary end for securing leather or cloth; just below the rivet is a recessed groove and shelf, perhaps to receive a reinforced edge; to the lower part of this is hinged a long handle cut in a leaf design. An iron hinge bar is part of the equipment for folding back the top of a chaise (USNM 60.178, fig.

91a). There are several horseshoes, two whole shoes and numerous fragments (fig. 91i and j). Finally, the handle shaft and decorative attachment of an iron currycomb (USNM 59.2077, fig. 91f) recalls Mercer's purchase of "1 curry comb and brush" in 1726.

XIX

_Conclusions_

Almost no exclusively 17th century artifacts were found at Marlborough; at least, there were very few sherds or objects that could not have originated equally well in the 18th century. The exceptions are the following: Westerwald blue-and-white stoneware with gray-buff paste; several sherds of delft and other tin-enameled ware, late 17th century in type, and an early 17th-century terra cotta pipestem. Otherwise, we find a scattering of things belonging to types that occurred in both centuries: North Devon gravel-tempered ware, which was imported both in the late 17th and early 18th centuries; yellow-and-brown "combed" ware, which elsewhere occurs most commonly in 18th century contexts; pewter trifid-handle spoons, the form of which dates from about 1690 but which may have been cast at a later date in an old mold (a wavy-end spoon in the style of 1710 may also have been cast later). Fragments of an onion-shaped wine bottle may date from the first decade of the 18th century, but the presence of such bottles in the Rosewell trash pit shows that bottles, being too precious to throw away, were kept around until they were broken--in the case of Rosewell for 60 or 70 years. Thus the Marlborough sherds cannot be excluded from the Mercer period. The same may be said of a late 17th-century type of fork. Thus, there is virtually no evidence of the Port Town occupation, especially as the few 17th-century artifacts that were found may well have belonged to the Mercers rather than to Marlborough's previous occupants.

The ceramics and gla.s.s are the most readily datable artifacts, and these coincide almost altogether with the period of John Mercer's lifetime. Common earthenwares are predominantly Tidewater and Buckley types, with a scattering of others, most of which are recurrent among other Virginia and Maryland historic-site artifacts. No distinct type emerges to suggest that there may have been a local Stafford potter.

Common stonewares occur in such a variety of types that no source or date can be attributed, although there is some evidence of the work of William Rogers' shop in Yorktown. Westerwald stonewares are predominantly of the blue-and-gray varieties commonest in the second quarter of the 18th century.

There is only a small quant.i.ty of delftware, but a great deal of Chinese porcelain. Evidences are that the first kinds of English refined wares, such as drab stoneware, Nottingham stoneware, and agateware, were used at Marlborough, thus pointing to an awareness of current tastes and innovations. The large quant.i.ty of white salt-glazed ware suggests that, although it was a cheap commercial product, it was regarded as handsome and congenial to the environment of a plantation house that was maintained in formal style.

Except for the white salt-glazed ware, which was probably acquired in the 1760's, most of the table ceramics date from about 1740 to 1760.

Bottles and the few datable table-gla.s.s fragments are also primarily from this period. Creamwares and late 18th- and early 19th-century whitewares diminish sharply in numbers, reflecting a more austere life at Marlborough in its descent to an overseer's quarters. Later 19th-century wares are insignificant in quant.i.ty or in their relation to the history of Marlborough. Tool and hardware forms are less diagnostic.

Most of them correspond to ledger entries and to the 1771 inventory, so, without contradictory evidence, they may be a.s.sumed to date from John Mercer's period.

In general, the artifacts ill.u.s.trate the best of household equipment available in 18th-century Virginia, and the tools and hardware indicate the extensiveness of the plantation's activities and its heavy reliance on blacksmith work.

GENERAL CONCLUSIONS

XX

_Summary of Findings_

Marlborough's beginnings as a town in 1691 cast the shape that has endured in a few vestiges even until today. The original survey of Bland and Buckner remains as evidence, and by it we are led to believe that the courthouse was located near the "Gutt" to the west of the town, near a change of course that affected the western boundary and all the north-south streets west of George Andrews' lots. Archeological excavation in the area disclosed Structure B, which subsequent evidence proved to be the foundation of Mercer's mansion, built at the pinnacle of his career between 1746 and 1750. No evidence exists that this foundation was a.s.sociated earlier with the courthouse.

Two years after the second Act for Ports was pa.s.sed in 1705, the second survey was made and was lost soon thereafter. There is evidence that the house built by William Ballard in 1708, on a lot "ditched in" according to this plat, was also in the vicinity of the courthouse. After Mercer moved into this house in 1726, it became clear that the two surveys were at odds, and a new survey was ordered and made in 1731. The maneuvers which followed make it fairly clear that Mercer's residence was encroaching upon the two acres that had been set aside for the courthouse, which by Act of a.s.sembly had reverted to the heirs of Giles Brent after the courthouse had burned and been abandoned about 1718. The 1731 plat provided a whole new row of lots along the western boundary of the town, while pushing the original lots slightly to the east. This device would have a.s.sured the integrity of the courthouse land, while relieving Mercer of the uncertainty of his t.i.tle. When Mercer's pet.i.tion to acquire Marlborough was submitted in 1747 (the 1731 plat still remained unaccepted), he offered to buy the courthouse land for three times its worth. Since Mercer was guardian of the heir, "Mr.

William Brent, the Infant," he was called upon to testify in this capacity at the hearings on his pet.i.tion. Thus the courthouse, Ballard's house, and Mercer's mansion all appear to have been involved in a boundary difficulty, and we may a.s.sume, therefore, that the courthouse during its brief career stood close to the spot where Mercer later built his mansion.

This difficulty, in particular, was influential in determining the shape of the town, the manner in which Mercer developed the property and the peculiarities that made Marlborough unique. It was not until 1755 that he was permitted to acquire all the town and by that time Marlborough's character had already been fixed. We have seen that its outstanding feature, the mansion, was architecturally sophisticated, that leading craftsmen worked on it, and that it was as highly individualistic as its master. It was lavishly furnished not only with material elegancies but with a library embracing more than a thousand volumes.

Aside from the mansion, the area most actively developed by Mercer lay between it and Potomac Creek, with some construction to the north and the east. In 1731, Mercer built two warehouses which probably stood near the waterside at Potomac Creek where his sloop and schooner and visiting vessels found sheltered anchorage. These burned in 1746, but must subsequently have been rebuilt, since Thomas Oliver in his 1771 report to James Mercer commented that the "tobacco houses" must be repaired as soon as possible. They were probably among the buildings that Mercer had constructed up to 1747, when he reported that he had "saved" 17 of the town's lots by building on them. These lots comprised 8-1/2 acres in the southwest portion of the town.

The windmill was built on land near the river sh.o.r.e, east of the mansion. It was probably located a considerable distance from the sh.o.r.e, although erosion in recent times has eaten back the cliff. In the fall of 1958, half of the stone foundations collapsed, leaving a well-defined profile of the stone construction. Fragments of mid-century-type wine bottles found in the lower course of the stones support other evidence that the mill was built in 1746.

Mercer mentioned his "office" in 1766. This may have been a detached building used for a law office. Oliver in 1771 listed a barn, a cider mill, two "grainerys," three cornhouses, five stables, and tobacco houses. He mentioned also that "the East Green House wants repairing, the west d^o wants b.u.t.tments as a security to the wall on the south side."

Besides the malthouse and brewhouse built in 1765 (which may have been situated at Structure H and the 100-foot-long stone-wall enclosure attached to Wall A), John Mercer in his 1768 letter mentioned "Cellars, Cooper's house and all the buildings, copper & utensil whatever used about the brewery," as well as the "neat warm" house built for the brewer. When the property was advertised in 1791, "Overseers houses,"

"Negroe quarters," and "Corn houses" also were mentioned.