Return to Naval Historical Center home pageImage of an anchorReturn to Underwater Archaeology Papers
Flag banner

Sinking of CSS Alabama

WASHINGTON DC 20374-5060

Whitehall Project 1995: A Preliminary Report on the Excavation and Study of the U.S.N. Row Galley Allen

by Eric B. Emery

(Material contained herein is made available for the purpose of peer review and discussion and does not necessarily reflect the views of the Department of the Navy or the Department of Defense.)

The winter of 1813/14 in the Lake Champlain Valley was a season of strained vigilance for U.S. Navy Master Commandant Thomas Macdonough. On June 18, 1812, a fledgling United States government had declared war on Great Britain over the issues of free trade and the impressment of American sailors. By February, 1814, Macdonough's shipwrights at Vergennes, Vermont, were engaged in an arms race with the British at Isle-aux-Noix, Canada. The prize was control of Lake Champlain. This strategic waterway was the key to supplying American efforts to invade Canada, and to British plans for splitting New England from the rest of the United States. Over the course of their struggle, both sides adopted the use of a variety of small war-craft, including the row galley.

Numerous reports from Canada in January, 1814, told of British efforts to expand their gunboat flotilla. It was believed that the British intended to resume raiding in the Champlain Valley by early summer. This prompted U.S. Navy Secretary William Jones to grant Macdonough his request for a row galley Squadron. The galleys were to be built according to the plans and specifications of the Chesapeake Bay row galleys designed by William Doughty, Chief Naval Constructor at the Washington D.C. shipyard. Jones described the Doughty galley as "the most perfect of [its] kind" measuring "75 feet long and 15 feet wide... carry[ingl a long 24 and a 42 pound carronade, row[ing] 40 oars, and draw[ing] but 22 inches water."

Row galleys boasted a number of advantages for service on Lake Champlain. They could be quickly built and were relatively inexpensive; they had a shallow draught which allowed them to travel almost anywhere on the lake. The galleys' low freeboard made them difficult to hit from a distance, and they were capable of being operated by sails or sweeps.

On February 14, 1814, the Navy Department contracted Noah Brown, one of New York's finest shipwrights, to build Macdonough's galley squadron. In less than two months, Brown constructed, armed, and launched a total of six of these vessels: Allen, Borer, Burrows, Centipede, Nettie, and Viper. Allen was launched onto Otter Creek below the falls at Vergennes in late April. It was immediately employed in protecting Fort Cassin, a makeshift battery at the river's outlet into Lake Champlain. Allen was manned by 40 officers and seamen under the command of Sailing Master William Robbins.

During the spring and fall of 1814, Allen cruised Lake Champlain looking for smugglers, and assisted Macdonough's fleet in its victory over the British at Plattsburgh Bay, on September 11th. When the war ended in December of 1814, the Navy's squadron was put in ordinary at Whitehall, New York. Three years later, Allen was recommissioned for patrol duty on the lake under the provisions of the Rush-Bagot Agreement. When the Navy Department closed the Whitehall station in 1825/26, the vessel was sold. Situated out of the way in the Poultney River, about a mile and a quarter northeast of Whitehall, the exposed portions of Allen's hull were removed by salvagers while the submerged portions remained well-preserved.

The 1981-82 Survey and Investigation

The remains of a row galley--thought to be Allen--were discovered and first investigated by Kevin Crisman and Arthur Cohn (founder of the Lake Champlain Maritime Museum) in August of 1981. The wreck was situated on the New York side of the river, downstream from the remains of the U.S. brig Eagle and the British brig Linnet. Approximately fifty percent of the original hull was determined to be intact. This included the keel (roughly 70 feet [23 m] in length) and the keelson (67 feet [22.1 m] in length), portions of the stem and stempost assemblies, and most of the starboard side. A full-scale investigation of the wreck promised to yield a considerable amount of new information on early naval life and the design and construction of War of 1812 vessels on Lake Champlain. It was further concluded that it would be possible to reconstruct, in the form of line drawings and construction plans, the hull's shape and appearance from the existing remains.

An archaeological survey of the hull was carried out in 1982. The project was supported by various local organizations, including the New York State Education Department, the Vermont Division for Historic Preservation, and the Champlain Maritime Society. There was evidence at this time to suggest that this wreck was in fact the row galley Allen. First, the wreck was found in close proximity to the remainder of Macdonough's squadron in the Poultney River. Captain James T. Leonard-commander of the Whitehall Station on Lake Champlain after 1814 had the remaining vessels Eagle, Ticonderoga, and Linnet moved into the Poultney River where they would be less likely to obstruct boat traffic heading north from Whitehall. Allen was the only operating vessel from the 1814 squadron at this time. When the station closed in 1825/26, Allen was probably taken into the Poultney River and auctioned off to local salvagers.

Secondly, the collection of ballast found in the wreck's stem suggested that it was being used to counterbalance a large weight in the bow. This was unusual because ballast was typically placed on either side of the keelson along most of its length. This kept the vessel trim in the water and added stability to the hull. When Allen was patrolling Lake Champlain during the latter part of its career, it mounted a long 12 pounder on its bow. The weight of this gun would have required sufficient ballast in the galley's stern to float the vessel on an even keel.

Finally, a silver-plated uniform button from the U.S. Army's 13th Infantry Regiment was found between the vessel's keel and keelson during the 1982 survey (INA Quarterly 22.1). It was known through historical documentation that Macdonough had used American soldiers stationed at Plattsburgh and Burlington to man the oars of his row galleys. There were soldiers from the 13th Infantry on Lake Champlain during the Battle of Plattsburgh Bay on September11, 1814. An American soldier may have lost this button from his uniform while bending oars for Plattsburgh aboard Allen.

Whitehall Project 1995 - Project Objectives

The second and final phase of Allen 's study was designed to gather the data needed to complete the analysis of the hull (thereby permitting its reconstruction). This needed to be done before the arrival of the Zebra Mussel in the Poultney River. The Zebra Mussel is a small mollusk that was introduced into the American Great Lakes roughly ten years ago (INA Quarterly 22.1). It is indigenous to freshwater lakes and streams in Europe. These creatures form colonies on both wooden and metal shipwrecks and obscure the details of their design and construction.

Five main objectives were defined for the 1995 project: (1) to gather sufficient information to develop a complete site plan; (2) to complete the documentation of the keel, keelson, and stem and stempost assemblies; (3) to record the construction and shape of frame sections at as many points as possible; (4) to conduct an intensive study of the vessel's starboard side including its ceiling planking; and (5) to map and record the vessel's ballast arrangement in the stern on both sides of the keelson and remove samples for conservation.

The 1995 fieldwork took place in the form of a joint field school between Texas A&M University and the University of Vermont from July 10th to July 29th. There were ten field school divers (in addition to the project directors, a local divemaster, and the author) to assist with the excavation and recording of the hull, removal and cataloging of artifacts, and photography. Besides this full-time staff, local dive instructor Ron Plouff and Dr. Robert Neyland from the Navy Historical Center assisted with some of the dredging and recording at different stages in the project. The Lake Champlain Maritime Museum provided logistical support, and also housed a conservation lab for the treatment of artifacts. The project additionally received INA support.

One of the primary objectives of the 1995 fieldwork was to excavate Allen's starboard ceiling and frames which were buried beneath the Poultney River's heavy clay banks. A grid was constructed during the first week of the project and placed over the wreck. This structure was twenty feet long by five feet wide (6.6 m x 1.6 m) and was divided into 5-foot-square excavation units. Transversely, the grid extended from the starboard edge of the keelson nearly to the top of the surviving starboard frames. The units within the grid were numbered consecutively from 1 to 4, with Unit 1 closest to the bow and Unit 4 to the stern. A second grid (10 feet [3.3m] long) was installed, but left un-excavated.

Dredge work began during the first week of excavation and continued until July 27th. The excavators started with Unit 1 and proceeded from stem to stern. Heavy clay (up to 8 inches [20.2 cm] deep in certain areas) was removed by systematically peeling off the sediments in 4-inch (10 cm) layers and hand-feeding them into the dredge system. Tight-meshed dredge bags were secured to the dredge exhaust and examined after each level was completed and between each frame. Many iron nails, wood chips and fragments, and pieces of pine ceiling were found mixed into the clay.


Allen's keel was estimated to be between 68 ft. and 70 ft. in length (22.4-23.1 m) and probably consisted of two timbers flat-scarfed together. Analysis of the keel was restricted to a five-foot section exposed in the bow, a three-foot section in the stern, and measurements of its top surface taken at fifteen frame positions. Samples of wood from the keel could only be obtained forward of amidships. The samples revealed that the forward keel timber was fashioned from white oak.

A rabbet was cut along the top of the keel to fit the outer planking. Its two grooves were at 40 degree angles and were roughly 1 to 11/2 inches (2.5-4 cm) deep. They ran from stem to stern approximately 1/2 inch (1.5 cm) down from the top of the keel. The rabbet started just forward of the after-end of the apron and could be recorded up to the point where it was difficult to discern due to the outer planking.

Stem and Stern Assemblies

The stem was flat-scarfed to the keel. This timber measured 2 feet, 7 1/2 inches (83.5cm) in length, and extended up from the horizontal plane of the keel at approximately a 58 degree angle. The keel extended about 10 inches (25.5 cm) forward of the stem's base. The lower edge of the stem was considerably worn and deteriorated. This suggested that the space was not intended to hold an outer timber. Instead, the stem's original dimensions would have filled that space. However, the wood samples taken during the excavation disagree with this theory. They show the existence of three different wood types: the section of the stem closest to the forwardmost end of the keel was fashioned from red oak. The timber fastened between this outer stem, or "false stem," and the apron was of American elm; and the apron was of white ash. Two plugs, or "stopwaters," were found in the scarf table where the stem and keel were attached. The apron was fastened to the top of the stem with four 3/4 inch (2 cm) diameter iron through bolts.

William Doughty's row galley plan of 1813- from which Allen was believed to have been built, showed a double-ended vessel with its rudder hung from a curved sternpost. Allen's stern bore little resemblance to that of Doughty's plan. There was a straight sternpost which was fastened to the top of the keel by a long flat scarf and secured to the keelson and a frame by means of two 3/4 inch (2 cm) diameter iron drift bolts. An iron eye bolt with a 1-1/2 inch (3.6cm) diameter hole was driven into the after end of the sternpost. This makeshift gudgeon was a fast and easy way to hold the lower rudder pintle.


The galley's keelson was nearly intact. It consisted of two timbers flat-scarfed together 24 feet (7.92 m) abaft the stem. The overall length of the keelson was 67 feet, 3 inches (22.3 m) and was accessible for documentation at all points. The forwardmost end was considerably deteriorated. The top surface had stanchion mortises as well as two mast steps.

At 51 feet (16.8 m) abaft the stem a series of planks extended transversely across the top of the keelson. This transverse decking, or planking, was above the level of the ceiling, and was perhaps intended to provide seamen with a surface to walk on that was not cluttered with ballast.

An interesting bolt pattern along the keelson provided evidence that may help confirm the existence of two keel timbers. At the stem, amidships, and the stern were a series of three 3/4 inch (2cm) diameter bolts that had been driven through the keelson and keel within a foot of each other. This provided additional rigidity to Allen's backbone at these potential weak sites. Similar drift bolt patterns were found 20 feet, 4 inches (6.2 m) from the stem and 30 feet, 10 inches (9.1 m). One of these locations coincided with the keelson scarf, and the bolts clearly were intended to secure the scarf. The second grouping of bolts may define the location of the keel scarf.

Starboard Ceiling and Frames

The ceiling on the starboard side of Allen was generally well-preserved. Some strakes reached more than 14 feet (4.6 m). Their widths ranged between 7 inches (17.7cm) and 15 inches (38 cm), and thicknesses between 3/4 inch (2 cm) and 1 inch (2.5 cm). Some of the notable features found on the ceiling included small wooden battens extending laterally from the starboard edge of the keelson toward the bulwarks. It is possible that these battens were designed to fix bulkheads in the hull's storage areas. Samples of both ceiling and battens were removed for documentation on shore. This also enabled divers to record the frames underneath.

Irregularities existed in Allen's frame construction and composition. The frames appeared to have been assembled with little regard for size and spacing. Allen's shipwrights were building in a hurry in 1814, and may have allowed considerable leeway for error. For example, one starboard first-futtock was attached to the wrong side of the floor (it did not follow the pattern of the other frames forward of amidships), drift bolts missed frames and simply passed through the keel and keelson, there were no limber holes cut into the underside of the frames, and some of the frames may have been produced from unseasoned wood. One frame in particular was abnormally large and soft. It was the largest frame in the entire vessel and was fashioned from white pine, an unusually weak wood to choose for a floor timber.

Preliminary Reconstruction

Reconstruction of Allen required close examination of the archaeological data from the summer of 1995 and educated conjecture. The vessel's starboard side was best preserved and thus yielded the most information. The missing portions of the hull were put together based on comparisons with other contemporary vessels of a similar class and information obtained from historical documents.

Brown mass-produced Allen and its sister galleys, and it appeared as though he used Doughty's plan only as a reference point during his work. The 1995 data revealed a variety of deviations from Doughty's original design. By looking at what construction details Brown chose to omit, we can begin to understand the problems he faced while building Allen, as well as his solutions.

Allen's hull shape was considerably different from Doughty's plan. Doughty showed a series of sections with a full turn of the bilge. Allen had a more pronounced deadrise (fig. 7). Macdonough needed his galleys to be sail-worthy. The sharper angle of Allen's hull would have allowed it to cut through the water more easily and provided better lateral resistance. Doughty's galley was more barge-like. It would readily have supported its guns, but would not have been easy to handle under sail.


Archaeological study of the ship during the 1995 Whitehall Project yielded a wealth of information about row galley construction on Lake Champlain during the War of 1812. Once the hull analysis and artifact documentation have been completed, we will have a better understanding of life aboard a Lake Champlain galley, and the ability to identify the construction details unique to Noah Brown-built vessels. Allen is the only known craft built by this important 19th-century shipwright that is available for study. Brown built simple, strong, and serviceable hulls capable of satisfying the U.S. Navy's needs. Allen is a testimony to the skills--and compromises--necessary to construct and outfit an inland flotilla within a period of less than two months.


The 1995 Whitehall Project was supported by a Legacy Grant administered by the U.S. Navy Historical Center, the Lake Champlain Maritime Museum, the Institute of Nautical Archaeology at Texas A&M University, INA Director Harlan Crow. Additional support was provided by the New York State Education Department and the Vermont Division for Historic Preservation. The knowledge and leadership of Dr. Kevin Crisman and Arthur Cohn brought the project into being and kept it organized, informative, and safe. Special thanks to Dr. Robert Neyland from the U.S. Navy Historical Center, and students Steve Bilicki, Steve Butler, Erich Heinold, Pierre Larocque, Scott McLaughlin, Scott Padeni, Cheryl Quinn, Erika Washburn, and Rob Wilczynski.

25 May 2001