Keelboat: Barge or Boat?

The 'keeled boat'

Figure 1

Scale Model of the "Boat" or "Barge"

Conception by Richard C. Boss

scale model of a wooden, masted boat

James Wylder photo

Courtesy Lewis and Clark Interpretive Center, Great Falls, Montana

Working from William Clark's detailed plan and profile drawings and accompanying notes as well as references scattered throughout the journals, and with a thoughtful review of scant contemporary documentations and recent historical interpretations, nautical historian Richard Boss created models of the so-called "keelboat" and the "red" and "white" pirogues, as well as samples of two of their dugout canoes plus a Nootka Indian canoe that the captains acquired on the lower Columbia River. Boss built his models at a scale of one-half inch to the foot.

He gave all of them to the Fort Clatsop National Memorial, which eventually placed them on permanent loan to the Lewis and Clark Interpretive Center in Great Falls, Montana. With the Center's permission, photographs of Mr. Boss's models are used to illustrate discussions of those watercrafts on Discovering Lewis & Clark.1

Lewis's Detachment Orders

[26 May 1804]

The posts and duties of the Sergts. shall be as follows (viz)—   when the Batteaux is under way, one Sergt. shall be stationed at the helm, one in the center on the rear of the Starboard locker, and one at the bow. The Sergt. at the helm, shall steer the boat, and see that the baggage on the quarterdeck is properly arranged and stowed away in the most advantageous manner; to see that no cooking utensels or loos lumber of any kind is left on the deck to obstruct the passage between the burths—   he will also attend to the compas when necessary.—

The Sergt at the center will command the guard, manage the sails, see that the men at the oars do their duty; that they come on board at a proper season in the morning, and that the boat gets under way in due time; he will keep a good lookout for the mouths of all rivers, creeks, Islands and other remarkable places and shall immediately report the same to the commanding officers; he will attend to the issues of sperituous liquors; he shall regulate the halting of the batteaux through the day to give the men refreshment, and will also regulate the time of her departure taking care that not more time than is necessary shall be expended at each halt—   it shall be his duty also to post a centinel on the bank, near the boat whenever we come too and halt in the course of the day, at the same time he will (acompanied by two his guard) reconnoiter the forrest arround the place of landing to the distance of at least one hundred paces.    when we come too for the purpose of encamping at night, the Sergt of the guard shall post two centinels immediately on our landing; one of whom shal be posted near the boat, and the other at a convenient distance in rear of the encampment; at night the Sergt. must be always present with his guard, and he is positively forbidden to suffer any man of his guard to absent himself on any pretext whatever; he will at each relief through the night, accompanyed by the two men last off their posts, reconnoiter in every direction around the camp to the distance of at least one hundred and fifty paces, and also examine the situation of the boat and perogues, and see that they ly safe and free from the bank.

It shall be the duty of the sergt. at the bow to keep a good look out for all danger which may approach, either of the enimy, or obstructions which may present themselves to (the) passage of the boat; of the first he will notify the Sergt. at the center, who will communicate the information to the commanding officers, and of the second or obstructions to the boat he will notify the Sergt. at the helm; he will also report to the commanding officers through the Sergt. at the center all perogues boats canoes or other craft which he may discover in the river, and all hunting camps or parties of Indians in view of which we may pass.    he will at all times be provided with a seting pole and assist the bowsman in poling and managing the bow of the boat.    it will be his duty also to give and answer all signals, which may hereafter be established for the government of the perogues and parties on shore.

Meriwether Lewis listed a "Keeled Boat" in his pre-expedition shopping list, but after he finally got it, he and the other journalists of the Corps of Discovery simply called it "the boat" (190 times) or, less often, "the barge" (32 times).2 In fact, the term "keel boat"–spelled either as two words or one–appears but four times in the extant journals, and then only in references to the watercraft used by other travelers they met on the lower Missouri, coming or going. One that they saw in early September of 1806, for example, was a "large keel Boat roed with 12 oars," which suggests it was roughly half the size of their boat. Or, they may have been looking at what will be referred to here as a commercial keelboat.

John Melish (1771-1822), a Scotsman who moved to America early in the 1800s, and published the first important traveler's guide to the republic in 1812, described the new keelboat, "so called from being built upon a small keel." It was "remarkably well adapted to the navigation of these rivers, and as they are strongly manned, and ply both upward and downward, they are getting into general use, and are perhaps the best passage boats on the Ohio." The price, according to Melish, was "about two and a half or three dollars per foot." Barges, he continued, "are well known. They also sail up and down the river, but this species of vessel is principally used below Cincinnati and the [Louisville] falls."3

Other river boats

Another name in the journalists' nautical lexicon (4 times) was bateau–a French word for boat. On May 14, 1804, for example, when Clark and his crews embarked from "Camp River a Dubois," both Ordway and Gass referred to their big boat as a bateau. Furthermore, in his Detachment Orders of May 26, 1804, Lewis directed that "the messes of Sergts. Floyd, Ordway and Pryor shall untill further orders form the crew of the Batteaux." Literally speaking, Ordway, Gass and Lewis were correct, but by the latter half of the eighteenth century, bateaux (the plural of bateau) was commonly used to denote long, flat-bottomed open boats, tapered more or less to a point at both ends–which were often tilted up, or "raked"–and were either paddled or poled on lakes and rivers. They were frequently used in military operations during the French and Indian Wars and the American Revolution.4 Bateaux came in various sizes, described as 3-, 4-, or 5-handed, depending on the number of men needed to propel them. Otherwise, since they were usually built by eye rather than from plans, further details are vague. In the five times the journalists wrote of meeting bateaux on the lower Missouri without mentioning specific details, we cannot tell whether they were looking at small, lightweight skiffs, or larger boats like their own pirogues.

A relatively new craft of the time was the Schenectady boat, also known as a Durham boat or a Mohawk River batteau. The Schenectady boat was originally designed late in the 18th century for commercial and military traffic on the narrow, shallow rivers and canals of western New York and the upper St. Lawrence. Built of pine boards on oak frames, and usually having canvas awnings amidships, they were propelled by oars or poles, supplemented with small sails when the wind was right. Their capacity rarely exceeded two tons of cargo. Lewis and Clark encountered just such a craft only three days from the end of their expedition. When the Corps stopped at the village of Charrette, they saw seven trading boats moored there, all bound for the Osage and Oto villages upriver. Clark recognized the type. "Those boats," he wrote, were:

in the bateaux form and wide in perpotion to their length.    their length about 30 feet and the width 8 feet & pointed bow & Stern, flat bottom and rowing Six ores only the Skeneckeity form.    those Bottoms are prepared for the navigation of this river, I beleive them to be the best Calculated for the navigation of this river of any which I have Seen.    they are wide and flat not Subject to the dangers of the roleing Sands, which larger boats are on this river.5

Commercial keelboats

For our present purposes it is worthwhile to try to differentiate the keelboat from the barge. The classic keelboat of the type that began plying the Ohio and Mississippi Rivers sometime before 1780, and that served Missouri River commerce–especially the fur trade–until 1860. (For an illustration, see Figure 9 in W. Raymond Wood's "The Post-Expedition Fur Trade.") A commercial keelboat could be between forty and eighty feet long, with a beam of seven to ten feet, a draft6 of about two feet. It typically had a rather flat bottom, usually built on an interior rather than an exterior keel, to make it easier to drag or pole over shoals and sandbars. The hull was pointed at both bow and stern, which made for steadier control in the complex and often unpredictable currents of the wide Missouri. A hull with a squared-off stern tended to wheel out of control at the mercy of the current if headway was lost for any reason, even momentarily.

The hold of a commercial keelboat was partially covered by a boxlike structure that sheltered both the cargo and the crew's sleeping space. It could be rowed by four to twelve men seated on both sides of the bow, whose efforts could be supplemented by a square sail when the wind was right. In addition, where a river was shallow and its bottom firm, the oarsmen could pick up "setting poles" and push the boat upstream by walking or crawling astern on the cleated, 12- to 18-inch-wide footways or passavants, built inboard of both gunwhales. The craft was steered with a large, pivoted oar or "sweep" extending as much as ten or twelve feet beyond the stern.

Although the era of the steamboat may be said to have begun with the maiden voyage of Fulton and Livingston's New Orleans in 1811-12, more than fifty years elapsed before the great technological leap entirely displaced the commercial keelboat on the major American rivers.7


Owen's Dictionary of 1764, the 8-volume encyclopedia in the expedition's traveling library, defined barge in two senses. On the one hand it was, "in naval affairs, a boat of state . . . having bales and tilts, . . . seats covered with cushions and carpets, and benches for many oars." Bales were bundles or packs of cargo or supplies that could serve as seats. A "tilt" or "tile" was a shelter or awning made of tarpaulin (canvas, coated with tar for waterproofing) held up by wooden hoops of fir or ash. The term barge could also denote "a flat bottomed vessel employed for carrying goods in a navigable river, as those upon the river Thames, called west-country barges."8

Barges of that general description began to appear on the Ohio, Missouri and Mississippi Rivers about 1800. Some were about the same length as a commercial keelboat, but could be up to twice as wide, with a burden or capacity of up to 40 tons, and a draught of three or four feet of water.9 Owing to the relatively numerous sandbars and other obstacles encountered on the Missouri, barges used there were generally smaller, or at least narrower, with lower gunwales and flatter bottoms than those intended for use on the Ohio and Mississippi.10

The names and physical descriptions of watercraft of the early 19th century were difficult to clarify even then. Here is the way Timothy Flint (1780-1840), the Harvard-educated American author and missionary on the Mississippi between 1815 and 1825, opened his essay on the state of nautical design and nomenclature during that decade: "No form of water craft so whimsical, no shape so outlandish, can well be imagined, but what, on descending from Pittsburg to New Orleans, it may some where be seen lying to the shore, or floating on the river."11 In other words, the co-captains were free to call their flagship whatever they pleased. Nearly six times out of ten they opted for the generic "boat." The rest of the time, perhaps in a spirit of nautical pride, they referred to it as "the barge." (Sergeant Ordway once–on April 3 1805–referred to it as "the Big Barge.") Never, ever, after the early uses of bateau, did they call it anything but "the boat" or "the barge." Howbeit, journal editor Nicholas Biddle's single reference to the expedition's main watercraft as a "keel boat"12, undoubtedly prompted by Clark's response to his query, that "the party . . . had one large Keel boat of about [a] 55 [foot] Keel"13 closed the matter for the next 80 years, and ever since has borne a ring of truth that defies contradiction.

Lewis's boat


  • Single mast, footloose square sail
  • Twenty double-banked, 16-foot sweeps (oars)
  • Length, 55 feet
  • Beam, 8 feet, four inches
  • Draft, 3 feet
  • Burthen, 12 tons plus crew
  • Crew, minimum 22
  • Built at Pittsburgh, summer, 1803
  • Post-expedition history, unknown.

Construction of the barge in Pittsburgh proved to be a nightmare. Lewis arrived there early in the afternoon of July 15 after an uneventful 200-mile journey from Harpers Ferry. He immediately fired off a note to Jefferson which he closed with the remark that "the Ohio is quite low, but not so much so as to obstruct my passage altogether." At that moment he was still under the impression the boat would be finished by July 20, but he was far too optimistic. On July 22 he confided to Jefferson:

The person who contracted to build my boat . . . pleads his having been disappointed in procuring timber, but says he has now supplyed himself with the necessary materials, and that she shall be completed by the last of this month; however in this I am by no means sanguine, nor do I believe from the progress he makes that she will be ready before the 5th of August.; I visit him every day, and endeavour by every means in my power to hasten the completion of the work: I have prevailed on him to engage more hands, and he tells me that two others will join him in the morning, if so, he may probably finish the boat by the time he mentioned.14

The usual navigation season was nearing its end, and the river continued to drop daily, approaching a historic low. Nevertheless, I was determined to leave whenever the boat was finished, "though I should not be able to make a greater distance than a mile pr. day." He briefly considered substituting two or three pirogues for the barge to make the Ohio river descent, but "the best informed merchants" thereabouts assured him there was no place to procure a big-enough boat anywhere down river. Furthermore the well-meaning builder promised the barge would be done by the 13th. Lewis explains:

However, a few days after, according to his usual custom he got drunk, quarrelled with his workmen, and several of them left him, nor could they be prevailed on to return: I threatened him with the penalty of his contract, and exacted a promise of greater sobriety in future which, he took care to perform with as little good faith, as he had his previous promises with regard to the boat, continuing to be constantly either drunk or sick. I spent most of my time with the workmen, alternately presuading and threatening, but neither threats, presuasion or any other means which I could devise were sufficient to procure the completion of the work sooner than the 31st of August; by which time the water was so low that those who pretended to be acquainted with the navigation of the river declared it impracticable to descend it.15

Lewis finally began his 981-mile voyage down the Ohio River on August 31. William Clark joined him in October at the Falls of the Ohio near Louisville. Overall, it was a tedious eight-week journey. Owing to the exceptionally low streamflow that year, it was necessary on numerous occasions to manhandle the craft over rocky riffles, or hire farmers to drag it over with horses and oxen. Today, 21 dams with interlocking reservoirs guarantee clear sailing the year around for commercial as well as pleasure traffic on the Ohio River.

Upon the Corps' departure from their winter encampment at the mouth of the River Dubois, on 14 May 1804, the barge was the flagship, with two pirogues—large rowboats of six and seven oars, respectively—completing the captains' little fleet. In the spring of 1805, when the rest of the Corps of Discovery set out from Fort Mandan with the two pirogues and six dugout canoes, headed for shallower water on the upper Missouri, the barge was sent back downriver under the command of Corporal Warfington. Its fate after arrival in St. Louis is unknown.

Clark's sketches

Figure 2

Clark's First Sketch

To see labels, point to the image.

side view of barge

Beinecke Rare Book and Manuscript Library, Yale University

Figure 3

Thole pins and grommets


© 1997 by Richard C. Boss

Figures 2 and 4, which appear in Clark's Field Notes about 21 January 1804, illustrate his plans for improving the barge. The boat had a mainmast with a square sail, plus a spritsail (see the discussion of the white pirogue) when it left Pittsburgh, but broke twice under strong winds before reaching Wood River, and the sprit broke on the Ohio River in early September. The men were still working on Clark's hinged replacement at Camp Dubois in mid-April of 1804.

Rising from the stern in Clark's drawing is an ensign-staff which bears a naval pennant showing the national origin of the boat and its crew. (Note: The ensign-staff and flag are absent from Richard Boss's model–shown at the top of this page–only because they had accidentally been broken off and mislaid prior to the time the photograph was taken.)

Tholes and oars

The tholes, as Clark noted in his sketch, are "pins to row by." They hold the oars in position when being "pulled," provide the fulcrum by which the oarsman's leverage against the water is exerted, and help to insure equal power from all oars. Also, they allow the oars to be left to lie over the side in the event of "action," or emergency, such as the need for the two men at the bow and stern oars to quickly seize the halyards to handle the sail. Clark added the rectangular blocks of wood in which the thole-pins were mounted in order to raise the fulcrum several inches, allowing the oars to be raised enough to avoid high waves and large obstacles such as sawyers. Rowlocks were made with either one or two pins; Clark used one, which was the more common practice at that time. At the points where the oars are to rest in the oarlocks, the oars are attached to the thole-pins with grommets, which are made by wrapping a single strand of hemp rope around the loom, or inboard length, of the oar, twisting the bight into a loop and slipping it over the thole-pin, then wrapping it around the loom again and tying off the ends.16


The square sail hangs from a long cylindrical timber called a yard, and is hoisted up the mast with a block and tackle. The sail is loose-footed–that is, without a spar or "boom" to keep the bottom of it securely spread. The two rows of ropes faintly visible across the lower part of the sail are "reef points" by which the sail can be gathered up and tied in order to reduce the sail's area in a strong wind. On September 30, 1804, for example, Clark reported a hard wind which turned the boat and made it rock, which frightened the Sioux chief they had taken aboard. "We apoligised," the captain wrote, "& proceeded on under a Double reafed Sale."

Figure 4

Clark's First Sketch

To see labels, point to the image.

top view of barge

Beinecke Rare Book and Manuscript Library, Yale University

Figure 5

Interior of captains' cabin

A view inside the barge's cabin

© 1997 by Richard C. Boss

Figure 6

View of stern

quarterdeck and tiller

© 1997 by Richard C. Boss

Setting poles

six 18-foot-long setting poles are slung on each side, aft. Each has an iron tip on its lower end, for a secure purchase among the mud, sand and rocks of the river. Setting poles were especially useful for fending off drifting logs and snags, and avoiding rocks, shoals and sandbars.17

Tow line

Clark showed a rope attached to an eye-bolt at the bow on the water-line. It was probably used to moor the boat to a tree any time they "came to"—stopped to rest or encamp. Cordelling with that line would have been impractical because it would usually draw the bow toward shore. Instead, the line was attached to a pivot point on the mast-head, and led forward, according to Boss, "through the eye of a 10 to 12-foot adjustable pendant which was secured at the bow. By adjusting the length of the pendant, the tow line's effective fore and aft point of attachment relative to the vessel's center of lateral resistance could be controlled." That is, the bow could be kept in or near the main force of the current while the stern could have been drawn shoreward, which would have directed the current against the starboard side of the boat and thus helped to propel it upstream.18 (In those days a similar principle was used to power ferryboats across rivers.)

Storage lockers

The notes accompanying the sketch in Figure 4 pertain to the storage lockers Clark devised, including hinged lids that could be raised for defensive protection. The planking required was sawed and trimmed by a detail of soldiers under Sergeant Pryor in mid-February of 1804 at Camp Dubois.19

The most important element of design that is absent from Clark's sketch is the shape of the hull. Evidently he had no reason to include it, but it has elicited numerous conjectures for many years. "The principal question seemed to be," Boss wrote, "What underwater shape would the keelboat have had to be able to go both up and down a river having shoal water and numerous sand bars? Should it be flatter or rounder?"20 Judging from incidents recorded in the journals, it was too rounded for safety, especially on the broad, deep waters of the lower Missouri. Four times between May 24 and July 26, 1804, the barge nearly rolled over–surely a terrifying prospect at the time.

Captain's cabin

The captains' cabin (sometimes called the "cuddy cabin," although never by Lewis or Clark), situated on the after deck, contained two bunk-beds, a bench, a desk to hold a portable writing-desk with inkwell, shelves for the captains' library, and instruments such as the Hadley's quadrant (or octant) and sextant, Clark's two-pole chain, and the log line, reel and ship. The six cabin windows shown in Clark's sketch (Figure 2), three on each side, could be closed with sliding shutters on the inside. The three on the near side in the photo are obscured by the bundle of setting-poles. On June 21, 1804, the day that the men had to warp (See Rowing, poling, towing below the barge up some exceptionally rapid water, Clark reported that they accidentally broke one of the starboard windows (and lost some spare oars that were slung under the windows).

The cabin was topped by a quarterdeck shaded by an awning. This was the post of the helmsman, who was in control of the tiller, which was attached to the rudder. The rudder Clark showed in his sketch was rather small for a vessel of this size and design, if he drew it to scale—and Clark is believed to have been meticulous about such details. It might have been adequate when going against the current, but during the 43-day downriver voyage in the spring of 1805 it could have made steering sluggish.

Furthermore, it appears that the helmsman was originally stationed in the cabin, but during the shakedown cruise from Camp Dubois to St. Charles it became obvious that he could not see ahead clearly from that location, so the tiller was extended up to the quarterdeck, as shown in Boss's model.21

The small box on the bench at right would have held the boat's compass.


Slow going

Barges remained the principal river craft for military transport, and keelboats for river commerce, until steamboats completely took over those roles around 1860. Meanwhile, hired oarsmen were paid 80 cents a day—more than four times what an Army private earned in 1803. Major Amos Stoddard, who remarked that the Missouri's current was faster than the Mississippi's, described the incredible labor of working a keelboat up any river:

[Seeking Counter Currents]

Keel boats, however strongly manned, cannot possibly ascend to any great distance in the middle of the current; in some places, indeed, they cannot make head against it. They are obliged not only to ply along the shore, where the water is less rapid, and where counter currents or eddies frequently prevail, but they also find it necessary to keep on the side opposite to the bends. Hence they cross the river at the lower extremity of every bend, which can seldom be done without falling down with the current about half a mile. It is said by old boatmen, that they are obliged to cross the Mississippi three hundred and ninety times on ascending from New Orleans to St. Louis. If we admit the river to average three fourths of a mile in breadth, and the loss of half a mile at each traverse, on account of the velocity of the current, it is evident, that the track of the boat, between the two points just mentioned, exceeds in distance the direction of either shore, more than four hundred miles.


These traverses are also necessary on other principles. Greater and more numerous obstructions appear in the bends than opposite to them. The banks likewise along the bends are generally concave, and constantly giving way in large masses, sometimes by several acres at once, which render a passage near them dangerous; while the banks on the opposite side project with a sloping beach, usually covered or fringed next the land with willows, and therefore safe of approach. It is universally the case that, where the banks cave in and waste away on one side, those on the other increase by the deposition of new matter.

[Stemming the Current]

Boats usually ascend from fourteen to twenty miles in a day. The labor of propelling them is excessive; it requires great exertion to move them against the current; and boatmen find it necessary to rest every hour, at least at every traverse. The river is so winding, that the daily progress of boats to their destination, is very inconsiderable. In one instance they are obliged to stem the current for fifty four miles to gain five; in another thirty miles to gain one and a half; and similar instances, though of less magnitude, occur in the course of almost every fifteen or twenty miles.22

Whether on the barge or one of the two pirogues, the task of keeping the oarsmen working with one another like a synchronized machine was the responsibility either of the riverman at the bow, or the steersman at the stern. The technique might have involved merely a shouted cadence, but the French Canadian hands would have been more accustomed to rowing in time to their own singing, with the pilot or the steersman as song leader. John Bradbury, a British naturalist employed by the Botanical Society of Liverpool to study North American flora, left St. Charles in the spring of 1810 on a small keelboat of ten oars. The hired Canadian boatmen, he explained, measured the strokes of their oars by songs, "which were generally responsive betwixt the oarsmen at the bow and those at the stem; sometimes the steersman sung, and was chorused by the men."23 Veteran rivermen like Pierre Cruzatte and François Labiche, who signed on with the Corps at St. Charles in late May of 1804, probably coached the American soldiers in their own favorite their own favorite work songs, such as En roulant ma boule," which they could have used to set the pace or cadence of the boat's 20 oars, stroke by stroke.

Rowing, poling, towing

In those days there was not much clear sailing or routine rowing on rivers like the Missouri; life for a riverman was never as predictable as a helmsman's chant. Seldom could the crew settle for long into the comparatively easy routine of rowing. They met reality for the first time early in the morning of May 23, less than two days after leaving St. Charles, when they ran on a submerged log and were "detained" one hour while they redistributed the load in one of the boats so as to float a foot lower in the bow–"trimmed by the head," in sailors' lingo. On the very next day the crewmen met their first real test as sailors when they entered "a Verry bad part of the River Called the Deavels race ground." Lewis writes:

this is where the Current Sets against Some projecting rocks for half a mile on the Labd. Side. . . . we attempted to pass up under the Lbd. Bank which was falling in So fast that the evident danger obliged us to Cross between the Starbd. Side and a San bar in the middle of the river, we hove up near the head of the Sand bar, the Sand moveing & banking causes us to run on the Sand. The swiftness of the Current wheeled the boat, Broke our Toe rope, and was nearly over Setting the boat, all hand Jumped out on the upper side and bore on that Side untill the Sand washed from under the boat and wheeled on the next bank by the time She wheeled a 3rd Time got a rope fast to her Stern and by the means of Swimmers was Carred to Shore and when her Stern was down whilst in the act of Swinging a third time . . . into Deep water near the Shore, we returned, to the Island where we Set out and assended under the Bank which I have just mentioned, as falling in.24

That stretch, where they were "So nearly being lost," Clark called "the retragrade bend as we were obliged to fall back 2 miles."

There was little point in trying to make headway against the current in the deepest part of the channel, but near the river's edges, where the current was usually slower, if wading was practical or the bank was relatively free of trees and brush, the barge could be cordelled, or pulled by a small crew on shore with a rope a couple of hundred feet long. If the river was deep near a brushy bank, bushwacking was an option; either the cordelling crew or a crew aboard the boat could pull it ahead by grabbing branches and walking toward the stern. The only other practical strategy was warping, by which one end of a long rope was carried upriver in one of the pirogues, or walked up by an on-shore crew, and tied to a tree; the boat was then pulled up by a crew on the downstream end of the rope aboard the boat, who marched from stem to stern along the cleated lids of the storage lockers. Meanwhile, a second rope was carried farther up the river and tied to a tree, and so on. Only once is this technique mentioned in the journals–when Sergeant Gass reported: "On the 21st [of June, 1804] we had rapid water, and for about a mile had to warp up our boat by a rope."

'Frisian horses'

Figure 7

Cheval défriser25

cheval de frise

Library of Congress, Division of Prints & Photographs

Shortly after the Corps shoved off from their camp at the mouth of the Kansas River on 29 June 1804, at "a verry bad place of water," the stern of the keelboat collided with moving sand from a large sandbar, and the current swung her around within 6 Inches of a huge sawyer–a fallen tree whose root-wad has sunk deep into the muddy or sandy bottom, while its branches bob in the current, sometimes momentarily visible, sometimes not, but usually ready to wreak destruction. If the keelboat had struck that sawyer, Clark mused, "her Bow must have been Knocked off & in Course She must hav Sunk in the Deep water below." Close calls like that were seldom much more than the blink of an eye away.

Lewis himself, in a letter to his mother from Fort Mandan in March of 1805, described the geneses and potential threats of sawyers:

The base of the river banks being composed of a fine light sand, is easily removed by the water, it hapens that when this capricious and violent current, sets against it's banks, which are usually covered with heavy timber, it quickly undermines them, sometimes to the debth of 40 or 50 paces, and several miles in length. The banks being unable to support themselves longer, tumble into the river with tremendious force, distroying every thing within their reach. The timber thus precipitated into the water with large masses of earth about their roots, are seen drifting with the stream, their points above the water, while the roots more heavy are draged along the bottom until they become firmly fixed in the quicksands which form the bed of the river, where they remain for many years, forming an irregular, tho' dangerous chevauxdefrise to oppose the navigator.26

Lewis's metaphor was an apt choice. Properly cheval défriser (singular; plural, cheveux), the phrase denoted a "Frisian horse" (Figure 7), a nickname for a medieval defensive obstacle commonly employed against enemy cavalry ashore by the seagoing Frisian people, who occupied the North Sea Coast from the Netherlands to the border of Denmark. Variants of the cheval défriser were used in the American Revolutionary War, the Civil War, and in the defenses of some Pacific Islands during World War II.

Navigating downstream

Figure 8

"Snags (sunken trees) on the Missouri"

by Karl Bodmer (1809-1893)27

Bodmer, Snags

Rare Book and Special Collections Division, Library of Congress

The Swiss artist Karl Bodmer, who was employed by Prince Maximillian von Wied (1809-1893) to accompany him on a journey to the upper Missouri River, painted this primeval forest of grotesque limbs reaching up from the dark water to grapple the side-wheeler Yellow Stone into submission, and submersion. The artist's macabre sense of humor is reflected by that lumpy black spot poised on the tip of the snag at extreme left. A closer look reveals it to be a turkey vulture–hungrily awaiting the carrion to come.

Riding the current downriver held another chapter of demands and dangers. The pilot or riverman had to be unerringly observant, and capable of making quick decisions. The men had to follow his directions precisely, and ply their oars strongly and steadily to keep moving slightly ahead of the current, especially when they had to cross ("break") an eddy line, where the circling currents could swing the boat out of control. It required a sharp eye to read the river and avoid snags, "dead men" and "sawyers" that could puncture the hull.

It cannot be said that there was a hierarchy of hazards, but shoals and sandbars where they could run aground and lose control exposed them in turn to potentially deadly drifts or log jams (embarasses) against which a free-floating boat could be helplessly pinned by the powerful current, or propelled against falling banks such as Amos Stoddard described (above), which could capsize or swamp a craft that blundered too near the falling dirt, rocks, and trees.28

It was nothing short of miraculous that the barge's return trip from Fort Mandan to St. Louis was completed in 43 days at the height of the Missouri River's spring freshet of 1805, without any reportable accidents that we know of. It is all the more remarkable because it is doubtful that Corporal Warfington had a full complement of oarsmen. What saved him from terminal disaster, beyond his own vigilance and skill as a ship's captain, may well have been that his pilot and boatman was the French trader Joseph Gravelines, and his crew included at least five of the engagés who had helped the expedition reach the Knife River villages in 1804–Deschamps, La Jeunesse, Malboeuf, Pineau (Pinaut) and Rivet.29 Their priceless cargo of scientific specimens, reports, journals, and maps, from which we are privileged to reconstruct many of the early achievements of the Corps of Discovery, is the everlasting–but frequently overlooked–monument to their navigational skill and personal bravery.


1. Richard C. Boss, "Keelboat, Pirogue, and Canoe: Vessels Used by the Lewis and Clark Corps of Discovery," Nautical Research Journal, Vol. 38, No. 2 (June 1993), 68-87. There have been other reconstructions of the expedition's flagship all of which drew international attention during the Bicentennial observance: Butch Bouvier, L&C Replicas in Onawa, Iowa; the Lewis & Clark Exploratory Center of Charlottesville, Virginia; the Lewis and Clark State Historical Site, Hartford, Illinois; the Lewis & Clark Boat House and Nature Center, St. Charles, Missouri; and the Missouri Valley Legacy Center, Bismarck, North Dakota.

2. Chapelle emphasized the lack of precision in nomenclature for watercraft. For example, "'barge' . . . may be a lighter, a long, narrow rowing boat, a ship's boat, a galley, or even a galley-ship—depending wholly upon the date of its usage and the context in which it stands." Howard I. Chapelle, American Small Sailing Craft: Their Design, Development, and Construction (New York, Norton, 1951), p. 9.

3. John Melish, Travels through the United States of America, in the Years 1806 & 1807, and 1809, 1810, & 1811 (Philadelphia: Thomas & George Palmer, 1812), 334. In 1816 he published his Map of the United States with the contiguous British & Spanish Possessions, Compiled from the latest & best Authorities—the first large-scale map of North America between the British (now Canadian) boundary and the southern tip of Baja California. Along the middle and upper Missouri, and the Columbia River to the Pacific, it included many details taken from Clark's map, which had been published in 1814 to illustrate Nicholas Biddle's paraphrase of the captains' journals. Melish's chart also served as a working plan for mid-century migrations to the Pacific Coast and the southwest. He followed it with numerous regional maps of considerable use to travelers, and with contributions to several important atlases. By the end of his life, Melish had earned a well-deserved reputation as the first great American cartographer.

4. ibid., 80-84.

5. Moulton, Journals, 8:367.

6. Draft, also spelled draught, denotes the depth of water a boat displaces, and thus indicates the depth required to float her. W. H. Smyth, The Sailor's Word-Book: an Alphabetical Digest of Nautical Terms (London: Blackie and Son, 1867), s.v. Draft.

7. Erik F. Haites, James Mak, and Gary M. Walton, Western River Transportation: The Era of Early Internal Development, 1810-1860 (Baltimore: Johns Hopkins University Press, 1975), 15-18.

8. A New and Complete Dictionary of Arts and Sciences; comprehending all the branches of useful knowledge (London: Printed for W. Owen, 1764), s.v. barge. See also William Henry Smyth, The Sailor's Word-Book: An Alphabetical Digest of Nautical Terms, (London: Blackie and Son, 1867), wherein a barge is "a boat of a long, slight, and spacious construction, generally carvel-built, double-banked, for the use of . . . captains of ships of war. . . . Also, a flat-bottomed vessel of burden, used on rivers for conveying goods from one place to another." "Carvel-built" means that the planks of the sides are all "flush and smooth, the edges laid close to each other, and caulked to make them water-tight; in contradistinction to clinker-built, where they overlap each other." "Double-banked" means that two opposite oars are pulled by rowers seated on the same thwart or bench.

In Noah Webster's first dictionary, the Compendious Dictionary of the English Language (1806), "barge" was vaguely defined as "a row boat for landing or pleasure." By the 1880s, "barge" gained another usage, according to the Dictionary of American Regional English: "a horse-drawn wagon or sleigh used for transporting a large number of people; now also a bus, especially a school bus." See also Note 10, below.

9. Leland D. Baldwin, The Keelboat Age on Western Waters (Pittsburgh: University of Pittsburgh Press, 1941), 46.

10. The barges seen today on major rivers such as the Ohio, Mississippi, Missouri, and the Columbia–Snake River System, represent a new type that originated with the rise of the steamboat. First made of wood in the early days of steamboat history, but of steel since the late 1880s, these are large, shallow, flatbottomed, square-ended, unpowered and unmanned scows that are "towed"—actually pushed by diesel-powered tugboats. They carry immense quantities of bulk freight, such as grain, coal, and certain chemicals. The capacity of a modern bulk-freight barge is approximately 1,500 tons, so a typical 15-barge tow carries perhaps 22,500 tons. For comparison, the capacity of the largest railroad hopper car is 100 tons; of a 100-car unit train, 10,000 tons; of a large truck trailer, 26 tons. Ingram Barge Company, at Accessed May 19, 2007.

11. Timothy Flint defined "barge" as being "of the size of an Atlantic schooner, with a raised and outlandish looking deck. It has sails, masts and rigging not unlike a sea vessel, and carried from fifty to an hundred tons." A keel boat of that time was "of a long, slender and elegant form and generally carries from fifteen to thirty tons. Its advantage is its small draft of water, and the lightness of its construction." Its propelling power was by oars, sails, setting poles, and cordelles (ropes used to drag the boat upstream by crews of men walking on shore). In northern latitudes during spring thaw, when the river is too high to permit the use of setting poles but the banks are lined by riparian shrubbery, cordellers wading in the shallows can resort to "bush-whacking," or pulling themselves upstream by the bushes. Timothy Flint, "No Craft So Whimsical—No Shape So Outlandish," in John Francis McDermott, ed., Before Mark Twain: A Sampler of Old, Old Times on the Mississippi (Carbondale: Southern Illinois University Press, 1968), 3-4.

12. "The party was to embark on board of three boats: the first was a keel-boat 55 feet long, drawing three feet of water, carrying one large square-sail and 22 oars. A deck of ten feet in the bow and stern formed a forecastle and cabin, while the middle was covered by lockers, which might be raised so as to form a breast-work in case of attack." History of the Expedition under the command of Captains Lewis and Clark, to the Sources of the Missouri, across the Rocky Mountains and down the River Columbia to the Pacific Ocean. Performed during the years 1804-5-6. By order of the Government of the United States. Prepared for the Press by Paul Allen, Esquire, in two volumes. (Philadelphia: Bradford and Inskeep, 1814), 1:2-3.

13. "The Nicholas Biddle Notes," in Jackson, Letters, 2:534.

14. Ibid., 1:112.

15. Ibid., 121-22. Historians of Elizabeth, Pennsylvania, one of the oldest towns in the Monongahala Valley, insist that Lewis's boat was built there at the boatyard of Captain John Walker. In Pittsburgh, however, students of the expedition point out that Lewis could hardly have rowed, poled, dragged or sailed the 26 miles from Elizabeth to Brunot's Island by midmorning on August 31, 1803, which strongly suggests that it was built at William Greenough's boatyard, located near the north end of today's Liberty Bridge, less than five miles from Brunot's Island.

16. The thole-and-grommet was replaced after about 1820 by a pivoted cast-iron, crutch-shaped or forked thole-pin, which is still the standard on most rowboats. Moulton, Journals, 2:162. Smyth, s.v. "thole" and related terms. The etymology of thole is unknown.

17. Sergeant Gass reported on September 22, 1804, that at an old Indian camp they found some abandoned dog-travois poles that would "answer for setting poles." The implication is that they appropriated some for their own use.

18. "Tho' frequently said that the boat was with difficulty got over bars &c.," Clark explained to Nicholas Biddle, "it is to be understood that in the channel of the river there was also water enough for much deeper vessels, but the rapidity of the current obliged us to go near the shore & creep up along the sides." Nicholas Biddle Notes, c. April 1810. Jackson, Letters, 2:534.

19. First Detachment Orders, February 20, 1804. Moulton, Journals, 2:175.

20. Boss, 71.

21. Ibid., 71-72.

22. Amos Stoddard (1761-1813), Sketches, Historical and Descriptive of Louisiana (Philadelphia: Mathew Carey, 1812; reprint, New York: AMS Press, 1973), 374-75.

23. John Bradbury, Travels in the Interior of America, in the Years 1809, 1810, and 1811, London, 1819 (reprint, Lincoln: University of Nebraska Press, 1986), 39.

24. Moulton, Journals, 2:250-51.

25. Petersburg, VA; sections of chevaux-de-frise before Confederate main works. LC-B811-3206; Reproduction Number LC-DIG-cwpb-01597. Two places form left and right halves of a stereograph pair (LC-B811-3206A, LC-B811-3206B).

26. Jackson, Letters, 1:223. Properly, cheveux défriser (literally, "to straighten the hair"), a French-Canadian riverman's expression meaning hair-raising or frightening.

27. Voyage dans l'intérieur de l'Amérique du Nord, exécuté pendant les années 1832-1833 et 1834, par le prince Maximillien de Wied-Neuwied. (Paris: A. Bertrand, 1840-43), Plate 6.

28. One of the most eloquent explanations of what it meant to be able to read "the face of the water" like a book is to be found in the last three long paragraphs of Mark Twain's memoir, Life on the Mississippi.

29. Moulton, Journals, 3:227n.

Funded in part by a grant from the National Park Service's Challenge Cost Share Program

Journal excerpts are from The Journals of the Lewis and Clark Expedition, edited by Gary E. Moulton
13 vols. (Lincoln: University of Nebraska Press, 1983-2001)