Make your own free website on

Devereaux Cove Fieldwork
9-14 July, 2000

MaineQuadlabel.jpg (15398 bytes)

The Devereaux Cove Wreck is located in a tidal flat along the Penobscot River, in the town of Stockton Springs, Maine.  On the map below you will find Castine just to the south of the wreck site.  Known as Bagaduce during the American Revolution, this is where the rebel flotilla began its retreat up river after an unsuccessful amphibious assault on the British at Fort George.

PenobBayMapWithWreck.jpg (95856 bytes)

Because the Devereaux Cove Wreck is located in a tidal flat, the visible remains are entirely exposed at low tide and completely submerged at high tide.  Consequently, many of the vessel’s timbers become waterlogged during high tide, only to be dried out when exposed at low tide.  This cycle accelerates wood decomposition and will eventually result in timbers too unstable to be excavated or yield accurate archaeological information.  Additionally, significant seasonal temperature changes result in ice formation and repeated expansion and contraction of the waterlogged timbers, further compromising their structural integrity.  Ice formation over the site also produces "rafting", a condition where loose timbers may be worked free of the mud as the ice they are attached to drifts away from the site.   Lastly, the wreck is being eroded by constant tidal changes and the movement of water over the exposed timbers.  Those timbers not firmly ensconced in the mud bottom or attached to a more substantial part of the wreck are subject to being dislodged and carried away with the outgoing tide.   Clearly, much of the vessel has already been lost via this process, as the wreck site is comprised chiefly of floors and futtocks and little else above the mud line.   Unfortunately, the threats indicated above cannot be mitigated or reversed.  Thus, documenting and identifying the vessel at Devereaux Cove requires immediate attention before further loss of the wreck renders such an undertaking impossible. 

With the above in mind, the aim of the 2000 fieldwork was to determine the extent of the vessel's remains, produce detailed site and wreck maps, determine the site's potential for further archaeological investigation, and begin the process of nominating the wreck for the National Register of Historic Sites.  The field crew consisted of ECU Maritime Studies students Kim Eslinger, Cathy Fach, Matt Muldorf, and Mike Plakos.  The project was directed by Russ Green, who again thanks those involved for their many valuable contributions.  A special thanks must also be extended to Dr. Warren Riess of the University of Maine's Darling Marine Center.   Dr. Riess has been locating and documenting shipwrecks of the Penobscot Expedition for nearly two decades and was an archaeologist on the Defence project during the late 1970s.  Also an American vessel scuttled during the Penobscot Expedition, the Defence is one of the most significant Revolutionary War  shipwrecks found to date, and yielded much information about late eighteenth century privateering, ship construction, and shipboard life.

DevCove Flooded.jpg (157578 bytes)

The historical record suggests that more than one Penobscot Expedition vessel may be located near the town of Stockton Springs. To the left is a view of the outgoing tide along the shore line at Devereaux Cove.  Crew members are standing next to the wreck which is covered by approximately a foot of water.  Although some preliminary set up was done while the wreck was covered by water, all mapping was done at low tide while the wreck was exposed. 

Because the wreck is exposed only a few hours each day and the working days on site were limited, a simple baseline/crossline system was used for mapping.   Crosslines were positioned every ten feet and extended beyond the edges of the wreck, with the largest grid measuring 10' x 12'.  A carpenter's square was used to ensure that each crossline was perpendicular to the baseline, and the baseline/crossline system was kept a minimal distance from the highest point of the wreck to ensure level, accurate measurements.   Due to the minimal amount of relief at the site, a 10-foot length of PVC pipe, with measuring tape attached, was used as a sliding measuring bar in each grid.  The ends of the "sliding bar" were positioned under each crossline using a plumbob, and the sliding bar, resting on the floors and futtocks, was leveled.  After first taking measurements directly under the baseline, crew members then slid the bar at 6-inch intervals, or less depending on the features in their grid, toward the ends of the crosslines.  The field crew simply plotted points along the PVC pipe and produced scaled drawings of the contents in each grid.  This method proved to be an accurate and efficient means of documenting the site.

DevCoveWreck Down Baseline2.jpg (170003 bytes)

From back to front: crew members Mike Plakos, Cathy Fach, Kim Eslinger (left), and Mat Muldorf.  Here the tide is on its way back in and the site is being flooded.  The remains of the Devereaux Cove vessel cover an area approximately 60' x 24' and much of the extant framing still has outer hull planking attached.  Probing beneath the mud line at the ends of the baseline revealed that portions of the keel may still intact.  Unfortunately, the portion that should be beneath the exposed floors is severely degraded.


Mike Plakos turns the PVC pipe perpendicular to the baseline and documents outer hull planking, which remained covered by mud and water even at low tide. 

Mikemapping3.jpg (187613 bytes)


Floor Timber.jpg (163480 bytes)

A floor timber was carefully removed from the site, photographed, sketched, and returned.   This view shows the sided (or top) dimension of the timber with three trunnels extant. Interestingly, most of the examined trunnels were octagonal and none appear to be wedged.   A wood sample was taken from this floor timber, as well as from outer hull planking and two trunnels.  An analysis of these samples will reveal what type of wood each component is and perhaps tell us something about the builder's preferences.  Wood typing is also useful for discerning where the vessel originated geographically.   Stone samples from the site will be examined in much the same way.

wedge.jpg (107847 bytes)

Although no clearly dateable artifacts were found at the site, this shim, found between a floor timber and the adjacent outer hull planking, proved interesting.  A badly corroded iron drift pin was also found.  Both items were photographed, sketched, and returned.


Back at the Darling Marine Center, Mike Plakos and Cathy Fach transfer their field drawings to graph paper.  All drawings were done to scale in the field, later transferred to an individual piece of graph paper, and then slid under the larger wreck map and traced.   Click here to view the preliminary wreck map. 

Cathy Mike Drawing.jpg (147786 bytes)

The results of the Devereaux Cove Wreck Phase II survey are very encouraging.  Clearly, the wreck warrants a limited excavation, as features hidden well beneath the mud may significantly add to our understanding of eighteenth century ship construction.  Moreover, important physical evidence may be found that will help substantiate the historical record, and more closely identify the wreck with the Penobscot Expedition of 1779.  Unfortunately, time is a crucial factor, for the natural processes alluded to above will soon claim what remains of this interesting shipwreck.

Click here to return to the Devereaux Cove project start page.