On July 5, 1993 ten students and six staff/faculty members from East Carolina University, Program in Maritime History and Nautical Archaeology arrived in Jacksonville for a four week summer field school. During the previous field school investigation in 1992, participants documented 55 feet of the forward deck (Cantelas 1993). The goal of the field school in 1993 was to excavate and document the engineering spaces on the starboard side in accordance with the site research design (Cantelas 1992:9-10). Upon completion, approximately two thirds of the vessel had been examined on the starboard side.

The Maple Leaf site is located in twenty feet of water and buried under mud 4 to 8 feet deep. Tidal currents are strong and carry a suspended silt layer 1 to 2 feet thick along the bottom. These conditions make underwater work difficult and diver visibility is restricted or non-existent. Coping with these conditions often slows work and makes accurate recording difficult. Several techniques were developed to improve underwater conditions and recording.

An experiment tried in 1992 successfully created an underwater structure that provided better working conditions on the site. A fence-like structure was built around the bow extending aft 50 feet using 4 by 8 foot galvanized wire panels covered with geotextile silt barrier fabric. The barrier deflected the silt layer and kept sediment and trash from filling in the excavation. It also allowed clearer river water to descend down to deck level, providing significantly better visibility. Three modular "silt barrier units", 24 feet square, were built to enclose the bow (Cantelas 1993:19).

The modular design was abandoned in 1993 in order to build a single large structure around the field school work area. This offered the advantage of completing the barrier before field school started allowing students more time to document the site. During three days in early June 1993, three divers from the East Carolina University Dive Safety Office and the Project's principal investigator erected a barrier around the engineering spaces amidships. Figure 5 illustrates the shape and location of the silt barriers built in 1992 and 1993.

The corners of the proposed barrier were marked by triangulating their position from the cable baseline. The baseline is a 1/4 inch steel cable attached to the stem post and rudder post, and marked at ten foot intervals. After placing the corner posts (8 foot lengths of 1 1/2 inch galvanized pipe), a line was stretched between each corner with marks every 8 feet to correspond with the placement of the silt barrier panels. Next, additional upright poles were placed at each mark to hold the barrier panels. Once in place, panels were set against the poles and pushed into the bottom approximately 1 foot. Large plastic tie wraps were used to secure the panels to the poles. Some areas of the barrier required additional strengthening using concrete blocks to anchor the panels against the current.

When finished, the barrier enclosed the midships portion of the Maple Leaf in a large square. It crossed the baseline at 55 feet and 120 feet, extending 21 feet to port and 40 feet to starboard. The enclosed area measured 65 by 61 feet or roughly 3965 square feet. The barrier required periodic maintenance during the investigation to fix gaps caused by the current.

Field school diving operations took place from the SJAEI dive platform consisting of a 28 foot spud barge rafted to a 28 foot pontoon boat. These vessels are named Tater and Mud Puppy, respectively. The spacious area housed a drawing table, two dredge pumps and a large work area for documenting recovered materials. Large concrete mooring anchors were placed off the starboard and port sides of the Maple Leaf's engine spaces to anchor the platform over the work area. The Maritime Program's 24 foot Privateer and a 15 foot Boston Whaler, loaned by a SJAEI volunteer, provided transportation to and from the site.

University students and staff completed all underwater work using modified scuba gear incorporating a redundant air system and a full face mask with wireless communication. This system offered additional safety on a site with many entanglement hazards. Communications allowed a surface tender to monitor diver air consumption and activity. It also allowed two or more divers to work together to complete complicated tasks.

The silt barrier successfully diverted relatively clear river water down to the main deck level providing better visibility than expected on the river bottom. Within the barrier, lights provided visibility ranging from 6 inch to 3 feet. Divers used an array of hand held and head mounted underwater lights. During times of extremely low visibility, divers used clear plastic bags filled with water to take measurements. The bag was sandwiched between the face mask and tape measure with a light shining in from the side. This technique proved very accurate but time consuming. During actual dredging operations large amounts of suspended silt temporarily eliminated all visibility and required excavation by feel alone.

Approximately 4 to 6 feet of river bottom sediment covered the engineering spaces. Using two water induction dredges, divers moved approximately 180 cubic yards of overburden. Although the silt barrier kept river-borne debris and silt out of the excavation, periodic cleaning was necessary to remove silt deposited by diver activity during site documentation. In accordance with state and federal environmental permits, water turbidity from the dredge outflow was monitored every 2 hours during dredge operation. Water samples were taken 100 meters down current from the dredge exhaust and compared to background samples taken at the same time. Permit stipulations require dredging operations to cease if the turbidity of the dredge plume exceeds 29 Nephelometric Turbidity Units (NTU) above background levels. Comparative measurements normally averaged 1 NTU or less. The highest recorded level was 9.17 NTU's, far below the permit limit.

Field school students began removing the overburden along the east wall of the silt barrier, effectively digging a trench from the center of the ship out to the starboard guard. From this point the excavation proceeded aft until it reached the paddle shaft. The area exposed included portions of the forward deck, the upper portion of the engineering spaces and the intact starboard guard. The open unit extended 45 feet along the length of the ship, from 60 to 105 feet along the baseline, and stretched 22 feet across the ship, from baseline to the edge of the starboard guard. Trenches were dug at the forward and aft end of the excavation. The short forward trench relocated the forward cargo hatch found in 1992, providing continuity between the two areas. It also revealed considerable scouring caused by river currents over the intervening year. Eight feet of mud covered the cargo hatch in 1992 and only 4 feet covered the area in 1993. Aft, a trench exposed the rear edge of the engineering section and located the walking beam and paddle shaft connecting rod.

During the excavation phase, many buoyant, disarticulated deck planks rose to the surface. The planks and loose timbers littered the engineering spaces as a result of Ross's work in 1883. Explosives used to remove the A-frame and paddle wheels caused tremendous damage to the deck structure. When excavators removed the overburden holding the debris in place, a great deal of it floated to the surface. The surface support crew kept a continuous lookout for "floaters" and retrieved many pieces with a small boat. The recovered wood was given a general provenience number designated for the overburden removed from the engineering spaces. The recovered material was then identified, drawn, described and photographed in black and white.

In 1992, similar material was redeposited on the wreck for on-site storage. The volume and extreme buoyancy of many large pieces made on-site storage unfeasible for the current investigation. After documentation, simple conservation procedures were implemented. The wood was wrapped in polyethylene sheets to slow dry. All diagnostic material from this group was removed for further study and alternative conservation techniques. Since the results of slow drying are not predictable only redundant and non-diagnostic material was treated in this manner.

Other artifacts, exclusive of large timbers and the most recent glass, metal and plastic debris, were listed under the general overburden provienence with a brief description of where they were found. They were then taken to the SJAEI conservation laboratory for documentation and treatment. Post sinking site formation processes destroyed the original provenience of artifacts on the deck or in the upper levels of the engineering spaces. The original silt deposited in the engineering spaces following the sinking was badly disturbed or completely removed by demolition work in the 1880's. The activity also moved and redeposited artifacts over a wide area of the site and allowed debris deposited on the site after the sinking to intermix with site artifacts.

Documentation did not begin until the entire starboard side of the engineering space was completely exposed. A recording grid was setup using the permanent cable baseline originally installed in 1989. The old baseline broke several times and was replaced during the field school with a new 1/4 in. galvanized wire baseline. The ends are secured to metal clamps placed over the stem post and rudder post. The site datum is the inside face of the stem post where the cable attaches to the clamp. All measurements taken on the ship are in reference to this point. The only exception is work conducted inside the aft cargo hold which used a subdatum. The cable is marked at 10 foot intervals with bronze clamps.

To create a mapping grid, transverse lines were extended perpendicular to the baseline. Polypropylene lines were fastened to the baseline at 10 foot intervals and stretched to the starboard guard. The end of each transverse line was triangulated from the base line to locate each grid. Because of the high vertical relief from the baseline to the guard, measuring tapes were placed on the ship directly under the transverse lines using a plumb bob. The polypropylene acted to guide students to mapping grids while actual measurements were taken from the tapes. Each transverse line received a letter designation for recording purposes.

The narrow units created by the transverse lines running athwart ship were further subdivided. Individual mapping grids were laid out between the transverse lines at 5 foot intervals. The transverse lines established grid units measuring roughly 5 by 10 feet extending from the baseline to the edge of the guard. For record keeping purposes, each grid was identified by the letter designation of the transverse lines at each end and the distance from the baseline. One student was assigned to each grid. Within the mapping grid, features were documented using offset measurements and triangulation from measuring tapes placed around the border. Each evening, the day's data was transferred to the site map (Figure 6). Any errors or questions were resolved the following day.

Work beyond the scope of the field school included continued excavations in the aft cargo hold. SJAEI volunteers accomplished the work throughout the summer of 1993. While an analysis of the cargo and its spatial distribution are presented below, architectural elements will be discussed here. Currently, this is the only interior examination of ship structure conducted inside the vessel's cargo holds. The structural investigation was limited by the area cleared of artifact material in 1988, 1989 and 1993 (Figures 19 and 20).