The International Journal of Nautical Archaeology (2002) 31.2: 228–236 doi:10.1006/ijna.2002.1045 Shipwreck in a swamp: the Burroughs site at Edenton, N. Carolina, USA Bradley A. Rodgers & A. Corbin Program in Maritime Studies, East Carolina University, Greenville, North Carolina, USA This preliminary site analysis of a suspected 18th-century shipwreck located in Edenton, North Carolina, USA, helps call to question the roll of abandonment in the examination of ships as artefacts. Abandonments often provide an easily accessed and inexpensive means of fleshing out knowledge of ship construction in the past. Abandonments are typically located near historically well-used commercial ports and harbours and may now be hidden by shallow water, marshland or land fill. These areas, in many instances, are inaccessible to normal remote sensing survey techniques.  2002 The Nautical Archaeology Society Key words: derelicts, abandonments, coaster, 18th century. Introduction he Edenton project began as a predisturbance reconnaissance. The site (0007EDS) was discovered and first reported in March 2001 by a local resident, Mr Gil Burroughs, after it became exposed by an unusual confluence of wind and tide (Fig. 1). In May 2001 East Carolina University Maritime Studies Program (ECU) archaeologists travelled to Edenton to assess the site for its historical and archaeological potential and to evaluate its potential as a summer field-school project (Rodgers, 2001). Abandonments have time and again provided East Carolina University (ECU) researchers a glimpse at the average maritime workhorse, the unremarkable vessels, seldom discussed in archaeological circles, but invaluable as a means to examine both ship construction and the material culture left within.[1] The wreck-site was easily accessible from shore and support facilities, and was found to contain several diagnostic details which dated it to an earlier era than originally suspected. Field assessment determined that the site warranted further study in the form of a documentation survey of the remains without the use of invasive techniques. The Phase II study in June 2001 comprised an overall assessment of the shipwreck remains including construction details, a plan of the port T 1057–2414/02/020228+09 $35.00/0 Figure 1. The Burroughs site as exposed at low water ( Program in Maritime Studies, 2001).  2002 The Nautical Archaeology Society B. A. RODGERS & A. CORBIN: SHIPWRECK IN A SWAMP Figure 2. Map of North Carolina showing the location of Edenton (John Hopkins University Applied Physics Laboratory, 1995, modified by the authors). side ceiling, frame ends, and keelson, a map of the wreck’s orientation to the immediate shoreline, and an assessment of diagnostic artefacts located after sweeping debris from the port side. The site is commonly referred to as the ‘Burroughs Site’. Historical background The town of Edenton, North Carolina, was founded in 1712 and served as the colonial capitol until 1746, when the seat of government was moved to New Bern. Edenton is a protected port, situated well inland on the Albermarle Sound (Figs 2 & 3). During the second quarter of the 18th century it became a busy and prosperous seaport. First established between the mouth of the Pembroke and Queen Anna creeks, Edenton had a sufficiently large natural harbour to accommodate a large number of ships. More than 60 vessels arrived during the last six months of 1729. Trade to Edenton came from the Northern colonies, the West Indies, and England. Commodities traded included tobacco, naval stores, lumber staves, headings, shingles and planks, foodstuffs including corn, herring, and pork, and cotton, linen, silk, shoes, hats, china, rum, salt, coffee, sugar, and molasses (Angley & Cashion, 1980: 4–7). According to the Sauthier map of 1769, the area around the Burroughs site lay beyond a swampy point near the mouth of Pembroke Creek and, apart from most commercial traffic, representing an ideal abandonment area (Sauthier, 1769). As is often the case when working vessels near the end of their useful lives, they are disposed of through abandonment. The abandonment areas are usually convenient to a commercial centre but not close enough to create an eyesore or hinder local navigation (Babits & Figure 3. Map of Edenton Harbor on Albermarle Sound, North Carolina ( Program in Maritime Studies, 2002). 229 NAUTICAL ARCHAEOLOGY, 31.2 Corbin, 1997). Unsurprisingly, Edenton was no exception. Edenton’s prosperity waxed and wained with economic trends throughout the 18th century but generally remained sound. Even the American Revolution proved a boon to the port. Its well protected, inland position, combined with the fact that no large warships could pass over the outer banks bar helped to ensure that Edenton was safe from attack and virtually impossible to blockade. In addition, the British blockade of larger east coast ports literally drove commerce to Edenton (Angley & Cashion, 1987: 15). However, by the 19th century both man-made and natural factors combined to slow commerce to a trickle. After the first quarter of the century the catalyst for decline became the opening and widening of the Dismal Swamp Canal in 1805, 1819, and 1825, respectively. The canal moved the state’s economic focus eastward to Elizabeth City. It also proved an efficient link between the northern and southern colonies by passing commerce inland and bypassing the ever dangerous coastal waters of North Carolina. North Carolina is still known as the graveyard of the Atlantic because of its hurricanes, shifting shoals, and conflicting ocean currents. In 1828 Currituck Inlet silted over, forever closing the old trade route to Edenton. As a result, Edenton lost much of its commercial revenues, surviving instead on fishing and agriculture. Its main crops were peanuts and cotton, and continue to be so. Today, however, it seems that Edenton’s curse has also become its blessing. Many of its 18th-century buildings have been preserved and the town’s ambience has attracted a thriving tourist industry. Project area The wreck is located at the mouth of Pembroke Creek, approximately 0·96 km (0·6 miles) west of downtown Edenton. The vessel has a north-south orientation with the bow facing the shore and the stern facing the creek (Fig. 4). The bottom consists of soft silty mud and a large quantity of lumber offcuts that originated from a nearby 19th- and early 20th-century sawmill. Water depth on site varied from 0·61 m to 1·22 m (2 to 4 ft) with minimal tidal fluctuations. The creek bottom is stable and relatively flat with a gradual increase in depth toward the channel. This is a typical inland river-mouth site with black tannic water and visibility ranging from 0 to 0·61 m (0 to 230 Figure 4. Location of the Burroughs wreck-site at the mouth of Pembroke Creek, North Carolina ( Program in Maritime Studies, 2002). 2 ft). The area is infrequently disturbed by winddriven currents and boat wash. Vegetation consists of a concentration of water lilies, cypress trees, and river algae. Marine and marsh life encountered included freshwater mussels, leeches, snakes, crabs, fish, and birds. Phase II testing After establishing the site parameters, a datum, and laying a baseline, the project crew commenced with clearing a 3·05 m (10 ft) wide transect across the vessel amidships from the port side turn of the bilge to the starboard side turn of the bilge, crossing the keelson. Most of the interior debris was found to be detritus such as aluminium cans, which had clearly drifted in and out of the site with changing water conditions. Accumulated mud and organic material in the hold area never exceeded 10 cm (3·94 ins). Though all dredged material was sifted for artefacts, little historically significant material remained in the test swath and those that were located were primarily intrusive items associated B. A. RODGERS & A. CORBIN: SHIPWRECK IN A SWAMP Figure 5. Cross section of the Burroughs site, taken from the stem looking forward ( Program in Maritime Studies, 2002). with local 19th-century logging operations. It seemed clear that the ship had been nearly stripped of artefacts before it was burned and abandoned. Despite this, the vessel is remarkably preserved from the keelson to the turn of the bilge. Since there was no stratigraphy it was decided to sweep clean the entire port side of the wreck, but sift the overburden for diagnostic material. Description of findings The lower portion of the white oak vessel was largely intact from the keel/keelson to the turn of the bilge. Most notable of the hull details included: the sacrificial pine planking attached to the outer side of the hull, the lack of a centre-board, and scarf chocks between the floors and futtocks. In addition, the floors were notched over the keel and the keelson was notched to accept the floors. The floors were connected to the futtocks via five-sided scarf chocks (Fig. 5). The keelson comprised the main keelson plus a decayed half rider or hog that sat on top of it to provide additional longitudinal strength, and to accept the mast-steps. Only one mast-step survived; this was 3·05 m (10 ft) from the sternpost. Iron drift bolts secured the rider, the keelson, and the keel through the floors. The keelson was scarfed with a recognizable ‘hook and wedge scarf’ with the keys still in place. White oak ceiling planking 6·35 cm (2·5 ins) thick survived from the keelson to the turn of the bilge and there was clear evidence that periodic repairs were made to it. Outer hull planking 51 mm (2 ins) thick was recorded protruding from the mud at the turn of the bilge; no doubt it is intact below this as far as the keel. The sternpost and inner sternpost were intact and the rabbet to accept the outer hull planking was still visible above the side strakes. The lower gudgeon strap is secured in this area and the lower part of the rudder was attached to the gudgeon by means of a pintle. The lower stem, cutwater, and gripe were also intact along with the rabbet to accept the hull planking. The lower cant frames were also intact in the bow although obscured at their base by a cypress root that had extended itself through the bow (Fig. 6). The wreck clearly demonstrated evidence of fire with substantial amounts of charred wood in the dredged spoil and charring along the frame ends. An overall lack of artefacts and ballast suggests that the ship was brought to shore to be repaired or was abandoned after being found unworthy of repair. At that time any remaining cargo and personal items could have been removed and the vessel burnt in order to salvage the fastenings. The location of the wreck in a shallow easily accessible environment may also have contributed to stripping of artefacts by townsfolk over the years. The first futtocks begin only 152 mm (6 ins) from the keelson, a possible indicator of early 18th-century construction. It had minimal deadrise and a shallow draft of only 1·83 to 2·75 m (6 to 9 ft), suggesting a coasting design. The depth 231 NAUTICAL ARCHAEOLOGY, 31.2 232 Figure 6. Overall site plan of the wreck found at the Burroughs site ( Program in Maritime Studies, 2002). B. A. RODGERS & A. CORBIN: SHIPWRECK IN A SWAMP Figure 7. A variety of iron fasteners found at the wreck-site ( Program in Maritime Studies, 2002). today suggests that considerable silting has taken place in the area since the 18th century if the vessel was burnt in situ. Artefacts The largest group of artefacts discovered included intrusive chains, log offcuts and log staples, all to be associated with the local lumbering operations. These artefacts were scattered over the wreck and seemingly intrusive. They occur over a wide area and are clearly associated with local activities unconnected with the wreck. A number of iron and wood fasteners were associated with the vessel including treenails and wrought-iron nails and spikes. The iron fasteners were handmade and include primarily roseheaded nails with chisel points. The iron fasteners are in varying states of preservation and are of varying dimensions; no two are alike (Fig. 7). Based on standard dating and evaluation criteria for hand wrought nails it appears that the fasteners from the site date from the mid 18th century and are certainly no later than 1805. Samples from the wreck compare closely with samples extracted from a Louisiana building of the same period (Meverden, 2001: 12–13). All the treenails found on site were eight-sided and in varying states of decay. Many showed signs of burning. An interesting corollary to the fasteners involves a concentration of what can only be described as scrap iron hask found in the port side bow area of the wreck. This ‘hask’ consists of small sections of scrap nail plate and cut nails Figure 8. A two-sheave, or double block located in the wreck-site ( Program in Maritime Studies, 2002). seemingly melted and welded together in a slurry of raw iron ore. Evidence of the burlap bag in which this material was stored exists in the form of burlap impressions on some of the scrap iron. Since much of the material in the hull is intrusive it cannot be stated with certainty that the cut nail evidence relates to the wreck. However, if it did belong, it would push the date of the vessel’s destruction to the later 18th or early 19th century. Farriers have suggested that such scrap and hask collections may signal emergency collection, a contingency used in time of war (Smyth, pers. comm., 2001). Material associated with the ship’s rigging also make up the artefact collection. Most notable were the large amounts of charred rope and a complete, two-sheave block (double block), made entirely of wood, with a rope strop (missing), lignum vitae sheaves, and oak block cheeks (Fig. 8). Additional pieces of disarticulated blocks and severely charred sheave remains were also located. The intact block measured 12789 89 mm (53·53·5 ins). One side of the block was badly damaged by fire but the other face was diagnostically sound. Further evaluation of the block suggests that the piece is made entirely of wood with the internal sheaves being made of lignum vitae (Guiacum sanctum), commonly known as ironwood due to its hardness. The block also shows signs of being machine-made (a post1760 development). Current research suggests a possible date-bracket for this type of block as the late 18th to early 19th century (Goodall, 2001: 4–8). 233 NAUTICAL ARCHAEOLOGY, 31.2 Figure 9. Grape shot and Spanish or ‘peak shot’ associated with the wreck-site ( Program in Maritime Studies, 2002). Three pieces of ceramic were found, all creamcoloured earthenware. More specifically, one was a piece of pearlwear with a moulded rim. It was hand-painted with an indistinguishable ‘feather’ pattern using the underglaze technique. The second piece was pearlware with a clear glaze. The third piece was creamware, also with a clear glaze. Initial analysis dates the fragments between 1770 and 1830 (Derby, 2001: 10). The most interesting, and ultimately the most diagnostic, artefacts were the items of weaponry that were found. Several pieces of canister or grapeshot and two pieces of ‘Spanish shot’, also known as star-shot or spikeshot, were found. The star-shot is an elongated object tapering to point at both ends (Fig. 9). It has a four-sided shaft which resembles two overly elongated pyramids joined at their bases and may have been produced simply by welding two spikes together at their heads. A ring, approximately 39 mm (1·55 ins), encircles the shaft at the midpoint and may indicate the approximate calibre of the gun intended to fire it. The grain lines of the central ring run counter to the shaft indicating that the ring was wrapped around the shaft in a forge. The overall length of the star-shot after conservation is 250 mm (9·8 ins) (Lewis, 2001: 3). Little published documentation regarding starshot ordnance like that found at the Edenton site exists. The apparent rarity of such items in the historical record in the United States prompts speculation, both concerning the ordnance and its association with the shipwreck. Neither of the pieces found had any visible marks indicating date or manufacturer which would have aided identification. The closest example of star-shot recovered archaeologically comes from the excavation of a 234 Figure 10. A wooden carpenter’s bevel with the owner’s initials ‘ID’ carved on the body ( Program in Maritime Studies, 2002). 16th-century English vessel, wrecked on Alderney, one of the Channel Islands. At the Alderney site, nine objects were reported which were described as ‘about 3 in. in diameter with 5 in. spikes on either side’ (Davenport & Burns, 1995: 34, fig. 5). These examples are similar in design to the Edenton star-shot, although the latter was not cast. Additionally, the Alderney star-shot was described as anti-ship and anti-personnel ordnance (Bound, 1998: 71). The unique shape of the star-shot allowed incendiary material to be wrapped around the protruding ends and then fired at the hulls of ships or into their rigging. The points would penetrate the hull while allowing the fire and fuel to catch and burn the vessel. The 16th-century starshot and the Edenton starshot have several similarities worth noting. Both consist of a double-ended iron rod, pointed at both ends. The Alderney ordnance is approximately 100 mm (4 ins) shorter (Lewis, 2001: 3–5). Both samples contain a centrepiece attached after the shaft was forged. In the case of the 16th-century example the central piece is a two-part sphere approximately 79 mm (381 in) in diameter, rather than a wrought-iron ring as found on the Edenton specimen. Unlike the Edenton piece the Alderney star-shot is marked with an ‘R’ (Lewis, 2001: 5). It seems likely that the Edenton spike-shot represents a ‘homemade’ piece of ordnance and is unlikely to represent Royal Naval or US Naval ordnance. An additional significant artefact is a wooden carpenter’s or shipwright’s bevel used to mark angles on wood (Fig. 10). It has an overall length of 241 mm (9·5 in), is 25 mm (1 in) wide by 19 mm (0·75 in) deep and is personalized with the inscribed initials ‘I D’. Although it is unclear if this artefact was lost with the ship or is intrusive, B. A. RODGERS & A. CORBIN: SHIPWRECK IN A SWAMP it is the type of tool likely to have been carried on board.[2] Finally, a number of seeds were recovered during cleaning the ceiling. Prominent among these samples are peanuts and cotton seeds indicative of the ship’s environment or cargo. Conclusions The wreck remains at the Burroughs site clearly represent a seagoing wooden vessel designed for coastal trade with a waterline length of about 24·38 m (80 ft) and an overall length of about 29·26 m (96 ft). It had a 7–7·6 m (23–25 ft) beam and, therefore, had a 3·5–3·8 m (11·5–12·5 ft) mean depth of hold. These figures in Old Measure Tonnage (1773 rules) indicate a capacity of between 231 and 269 tons. Construction details are consistent with early to mid 18th-century techniques as are the fasteners, some of the ceramics, and the block. The relative dearth of fasteners and other artefacts suggests the vessel was intentionally abandoned, scavenged, and burnt for its iron in the late 18th or early 19th century. This is consistent with historical and archaeological accounts of other abandoned vessels of this date. Comparative artefact analysis in conjunction with known structural features narrow the approximate date of build to the first or second quarter of the 18th century and destruction in the later part of the 18th or early 19th century. The few artefacts recovered associated with the wreck corroborate the vessel’s long life. The ordnance and hask suggest a wartime influence, perhaps that of the American Revolution. Further investigation is certainly warranted in this case as only a handful of 18th-century vessels have been located and investigated in North Carolina waters. Additional work is planned and the site will be the focus of Kate Goodall’s ECU masters thesis. Acknowledgements The authors wish to express their thanks and appreciation to Mr Gil Burroughs who discovered this site and reported it to ECU for possible investigation. This work and field school would not have been possible without the assistance and support provided by the people and city of Edenton, North Carolina, particularly Mayor Roland H. Vaughn and Town Manager, Anne-Marie Knighton. Much appreciation also goes to Mr Jeff Knox and his wife Rae for their hospitality and logistical assistance. Fieldwork was made possible with the help of our crew: Assistant P. I., Frank Cantelas, crew chiefs Matt Lawrence and Deborah Marx, and crew members Kate Goodall, Marc Porter, Carrie Bell, Steve Williams, John Hart Asher, Jason Paling, Mike Overfield, Alena Derby, and Keith Meverden. Thanks and appreciation to Kate Goodall for mapwork and historical background and to Christopher F. Valvano for photographing the artefacts. Notes [1] (See Babits & Corbin, 1995; Babits et al, 1995; Babits & Corbin, 1997; Kjorness & Babits, 2000; Merriman, 1997; Rodgers, 1995; 1999; Rodgers & Corbin, 2001; Watts & Hall, 1986). [2] Ross, L. A., 1979, Underwater Archaeologists Vade Mecum to the Companies, Duties, Tools and Related Supplies of the 18th-Century British and French Naval Vessels. Unpublished MS in the possession of the author. References Babits, L. E. & Corbin, A., 1997, Locating small boats in the archaeological record: a model from the North Carolina Sounds. Museum Small Craft Association, Transactions, 3: 34–45. Bound, M., 1998, A wreck off Alderney from the late Elizabethan Period: an analysis of the artefacts. In M. Bound (Ed.), Excavating Ships of War, 2: 6–83. Oswestry. Daventry, T. G. & Burns, R., 1995, A sixteenth century wreck off the island of Alderney. In M. Bound (Ed.), The Archaeology of Ships of War, 1: 34–40. Oswestry. Kjorness, A. C. & Babits, L. E., 2000, A derelict small boat survey Pamlico Drainage, North Carolina, USA. In J. Litwin (Ed.), Down the River to the Sea, 8th International Symposium on Boat and Ship Archaeology, Gdansk, Poland, 1997, 193–196. Merriman, A. M., 1997, The Cypres Landing Shipwreck of Chocowinity Bay: A North Carolina Sail Flat. Research Report No. 9, East Carolina University, Greenville. Rodgers, B. A., 1995, The 1995 Predisturbance Wreck Site Investigation at Clafin Point, Little Sturgeon Bay, Wisconsin. Research Report No. 10, East Carolina University, Greenville. 235 NAUTICAL ARCHAEOLOGY, 31.2 Unpublished reports on file in East Carolina University, Greenville and elsewhere Angley, W. & Cashion, J. C. (Eds), 1980, Historical Research Reports, Series I—State of North Carolina, H629 series 1, #19 microfilm, North Carolina Collection, East Carolina University, Greenville. Babits, L. E. & Corbin [Kjorness] A., 1995, A Final Report on an Archaeological Survey of the Western Shore of the Pungo River from Wades Point to Woodstock Point. Report on file North Carolina Department of Cultural Resources, Raleigh, NC. Babits, L. E., Morris, J. & Corbin [Kjorness] A., 1995, A Survey of the North Shore Pamlico River: Bath Creek to Wade’s Point. Report on file North Carolina Department of Cultural Resources, Raleigh, NC. Derby, A., 2001, Conservation of Ceramic and Brick from the Burrough’s Site, Edenton, NC. Report for the East Carolina University Conservation Lab, Grenville, NC. Goodall, K., 2001, Conservation of a Wooden Block from Edenton Site. Report for the East Carolina University Conservation Lab, Grenville, NC. Lewis, R., 2001, Conservation of a Starshot. Report for the East Carolina University Conservation Lab, Grenville, NC. Meverden, K., 2001, Conservation of Iron Fasteners from the Burrough’s Site (0007EDS), Edenton, North Carolina. Report for the East Carolina University Conservation Lab, Grenville, NC. Rodgers, B. A., 1999, The 1999 Bullhead Point Stone Barge Investigation. Interim Report. Technical Publication No. 5, State Historical Society of Wisconsin, Madison. Rodgers, B. A., 2001, Burroughs’ Site, Edenton, North Carolina (0007EDS): Preliminary Site Analysis Conducted by East Carolina University’s Program in Maritime Studies, June 2001. Report for the North Carolina Underwater Archaeology Unit. Rodgers, B. A. & Corbin, A., 2001, Birmingham’s Site Analysis, Scow Schooner Dan Hayes, Sturgeon Bay, Wisconsin. Report for the State Historical Society of Wisconsin, Madison. Sauthier, C. J., 1769, Plan of the Town and Port of Edenton in Chowan County North Carolina, 1769. Historic Map files, North Carolina Division of Archives and History, Raleigh. Watts, G. P., Jr. & Hall, W. K., 1986, An Investigation of Blossom’s Ferry on the Northeast Cape Fear River. Research Report No. 1, East Carolina University, Greenville. 236