FIRE, SMOKE, AND STONE AT THE POLLOCK WORKS
Robert V. Riordan
Wright State University
Ohio Archaeological Council © 2002
The Pollock Works enclosure of Greene County, Ohio, was constructed in the Middle Woodland period using earth and stone embankments to isolate a limestone plateau on the south side of Massie's Creek (Figure 1). The gradually rising land surface on the west side is traversed by three meters high embankments which are penetrated by three openings or gateways. Field research conducted between 1999 and 2001 by Wright State University crews concentrated on the environs of the central gateway.
A small test had been dug in the first field season at the site (1981) that ran across the gateway's opening. It demonstrated that the gateway surface had been paved with stone to a depth of 30-40cm. Probing done in 1982 had further suggested that there was a consistent layer of stone under the narrow gateway passage (ca. 1.5m wide) that extended an unknown distance to the west, outside the enclosure. Work in 1999 therefore started with exterior shovel tests that were intended to find the limits of the exterior stone deposits.
While the exterior surface area immediately south of the gateway was disturbed by 19th-20th century quarrying activity, the tests revealed a "pavement" of stone that extended some 10 m to the north of the gateway and about 10 m to the west. The stone was covered by only several centimeters of soil, which presumably had formed in the approximately 1,800 years since the last construction at the enclosure. After the pavement's limits had been detected, a 3x3 meter area just to the exterior of the gateway was completely stripped of its soil cover to expose the stone. It was composed of rough limestone, ranging in size from chunks of ca. 0.25 kg up to massive blocks weighing as much as 68 kg. A narrow (80cm wide at the surface) profile trench was then set up, oriented so as to run directly in the direction indicated by the opening of the gateway itself (and not on the regular site grid). Starting at the interior side of the gateway entrance, it extended 16m to the west; later it was extended an additional 3m to the east, inside the enclosure.
The result was to expose the limestone that had been deposited as part of the pavement. It was at its thickest within the gateway proper, that is, between the adjacent embankments traversed by the passage. Even 10 m to the west, however, the pavement was generally two to three stones deep, not just the surfacing of stones in a single thickness that had been expected. At the eastern end of the trench, a 3m long by < 1m wide set of fitted limestone formed a smooth surface "ramp" into the enclosure. This and the rough exterior pavement appear to be prehistoric features, and both are connected to the final prehistoric stone paving within the gateway.
A perpendicular trench, 2 m wide was extended into the embankment to the south of the gateway. This effort, incomplete after the 2001 field season, has so far exposed two layers of limestone. The uppermost was connected to that which was added to the gateway surface. This layer lies directly above a layer of burned limestone which is embedded within and lies above a distinct stratum of burned soil. We have previously explained the presence of burned soil in association with carbonized stockade posts and horizontally interwoven timbers as a burned mud plaster that had been applied to the lower reaches of the stockade. Because there is a tremendous amount of it beside the gateway (>40-65cm to ~1m thick), work in 2002 may confirm this hypothesis or indicate another explanation. While no radiocarbon dates have yet been processed on material from this locus, elsewhere the stockade has been consistently dated to the late second to early third centuries A.D.
Squier, Ephraim G., Edwin H. Davis
1848 Ancient Monuments of the Mississippi Valley. Contributions to Knowledge No. 1. Smithsonian Institution, Washington, D.C.