The First Survey Through Hal's Lake Swamp

The First Survey Through Hal's Lake Swamp in 1809 Under the Treaty of Mt. Dexter

Contributed by: David A. Bagwell



On November 16, 1805, on Mount Dexter, in Pooshapukanuk, in the Choctaw country, two Commissioners of the United States -- Silas Dinsmoor of Alabama and James Robertson of Tennessee -- concluded the Treaty of Mt. Dexter with certain Great Medal Mingoes (major chiefs)(including "Pooshmattaha"), chiefs and warriors of the Choctaw nation. Under the treaty, the Choctaw nation ceded certain lands to the United States in what is now parts of Mississippi and Alabama, the easternmost boundary of which was:

the boundary between the Creeks and Chaktaws [sic] on the ridge dividing the waters running into the Alabama from those running into the Tombigbee, thence southwardly along the said ridge and boundary to the southern point of the Chaktaw claim.

The Treaty of Mt. Dexter set out how that line (among others) would be surveyed:

The two contracting parties covenant and agree that the boundary as described in the. . . [first] article shall be ascertained and plainly marked, in such way and manner as the President of the United States may direct, in the presence of three persons to be appointed by the said nation, one from each of the medal districts, each of whom shall receive for this service two dollars per day during his actual attendance, and the Chactaws shall have due and seasonable notice of the place where, and the time when, the operation shall commence.

The treaty was ultimately approved by the U.S. Senate and proclaimed on February 25, 1808.

One year later, on Monday, February 20, 1809, the survey began, in Mississippi.

Fortunately, the principal surveyors kept a journal and field notes, entitled on the first page:

Journal and Field Notes on The Boundary Lines Between The United States and The Chaktaw Nation of Indians surveyed pursuant to the Treaty of Mt. Dexter concluded on the 16th Nov. 1805

The survey was completed on November 28, 1809. The surveyors arrived at St. Stephens by "perogue" on December 1, 1809, settled accounts, and arrived at "Washington" (probably Washington, Mississippi) on Christmas eve of 1809; "the pack horses arrived on the 31st [and] the Indian chiefs leave me on the 15th of January, 1810". The survey was over.

The Journal was marked "received" by the U.S. Government on May 21, 1810.

On August 24, 1814 the British captured Washington, D.C. and burned the public buildings. Edward Tiffin, Commissioner of the Government land office ["GLO"] had anticipated that the wood GLO building would be burned, and he had his clerks and anybody else he could muster to stash the records -- probably including the Journal -- in private homes, which saved them.

Ultimately, the Journal was placed in the National Archives, from which we obtained a copy in 1994.

The Journal, hitherto little known and incredibly detailed, supplies nearly all we know about the survey. Cast of Characters. The three main people involved were surveyors Silas Dinsmoor and Levin Wailes, and local helper James Powell.

Silas Dinsmoor. Silas Dinsmoor was born September 26, 1766 at Windham, New Hampshire. When he reached eighteen, he moved to Bangor, Maine, where he worked a year on a farm for four dollars a month. He then prepared for college under the tutelage of a minister, graduated from Dartmouth in 1791, and taught for three years. Upon the reorganization of the Corps of Engineers in 1794, he was commissioned Lieutenant. President George Washington was discussing with Lt. Dinsmoor the best way to treat the Indian tribes; when Dinsmoor said it was to teach the Indians "civilization", Washington persuaded him to resign his Army commission and become Agent to the Cherokees. He spent five years among the Cherokees before becoming purser on a Navy ship, U.S.S. GEORGE WASHINGTON and, under Commodore Bainbridge, voyaged to Algiers and Constantinople. His success with the Cherokees, however, brought his appointment as Indian agent to the Choctaws. Marrying Miss May Gordon of Hampstead, N.H. in 1806, shortly an invalid, they travelled by land from New Hampshire over the Allegheny mountains to Pittsburgh, where Dinsmoor built a flatboat. They floated sixty-eight days downriver to the point on the Tennessee River nearest the Choctaws (probably Muscle Shoals), where he travelled to the Choctaw nation.

Dinsmoor as agent or factor to the Choctaws lived first on the Chickasahay River near present Quitman, MS, and then west of the Pearl River. As was written in 1910, before Political Correctness ensued, Dinsmoor "was enthusiastic at both his stations in improving the condition of the savages. He induced them to make larger truck patches, added cotton to their crops, and introduced poultry, hogs and horses. . . [W]ith Africans [he] operated a large plantation near the agency". In 1805, Dinsmoor was one of the Commissioners for the U.S. in the Treaty of Mount Dexter, which ceded to the U.S. the lands which caused the 1809 survey discussed in this paper.

Dinsmoor made an enemy of Andrew Jackson in at least two ways. First, he insisted on passports for the slaves of travellers. Second, "[t]he story goes that, when asked by the government authorities of Washington how far the Tombigbee ran up the country, he replied that it did not run up the country at all, but down".

In 1815, a period of Jackson ascendancy, Dinsmoor gave up the position of Choctaw agent, and moved to St. Stephens, where a son was born.

Dinsmoor did not officially become a surveyor until 1819, when he became principal deputy of the principal surveyor of the Public Lands of the United States south of Tennessee, by appointment from Thomas Freeman, principal surveyor.

Dinsmoor may not have been much of a surveyor. Peter Hamilton, real estate title lawyer extraordinaire as well as Mobile Historian, wrote in 1910 that when (in 1821) Fort Charlotte was abandoned by the military:

As provided by the Act, Silas Dinsmore [sic], United States Deputy Surveyor, was directed to plat it into squares, conforming as nearly as practicable to the general city plan. Dinsmore [sic] was, as we have seen, a man of marked individuality and made a plat more according to what he thought the city plan should be than what it was; for the streets were not continuations of those existing, and the lots were only thirty feet wide.

In 1824, Dinsmoor was principal Deputy Surveyor of Land of the U.S. General Land office at New Orleans.

Dinsmoor was chairman of the Reception Committee to welcome LaFayette to Mobile in 1826. Shortly afterward he took his family north for the summer and in the great fire of Mobile lost all his possessions. He is said to have surveyed Spring Hill in 1826 and lived there in 1828. Along with Henry Hitchcock (Ethan Allen's grandson and Mobile's first millionaire) Dinsmoor was a major contributor to the building of Barton Academy in Mobile, which still stands as a monument to those two men.

In the early to mid 1830's Dinsmoor moved to Cincinnati, then bought a farm at Bellevue, Kentucky, where he died on June 17, 1847 at age 81, and is buried, his grandson reported in 1910, "on a beautiful hilltop commanding a distant view of the Ohio River".

Levin Wailes. The journal lists both Dinsmoor and Wailes as "agents of the United States" in the survey under the Treaty of Mt. Dexter. Wailes kept most of the journal, seems to have the most actual surveying knowledge, and wrote with the clarity and pithiness and directness of Julius Caesar. Those who have most recently read the Journal have come away feeling almost a personal friendship with Wailes, though he never once makes a personal observation in the Journal.

Levin Wailes was born on March 9, 1768, in Prince George's County, Maryland, to a captain in the Revolutionary Army. When he was seventeen he embarked at sea with a relative, and made two voyages to England, which "taught him something about mathematics and about those instruments by which man charts his course whether on sea or land". Upon returning to shore, Levin became chief clerk to his brother Edward Lloyd Wailes, Sheriff of Prince George's County, helping to conduct the first presidential election. He read law.

Levin lived for a brief time in Philadelphia, where (as an Episcopalian) he saw President Washington worship, and met Robert Morris, "the patriot-financier of the revolution".

In 1792 Levin moved to Georgia, as agent of the North American Land Company, Morris' outfit. On October 28, 1796, Levin married Eleanor Davis of Prince George's County, but by 1797 Morris was broke and couldn't pay him.

Levin briefly ran a flour mill, but "[h]e was chiefly interested in surveying", a skill he had learned from a relative. A friend wrote of Levin that "he is the most correct and best surveyor in the State of Georgia, or perhaps, elsewhere", and "much of the time he was in the back country of Georgia with his compass and chain". Five of his nine children were born in Georgia.

In the spring of 1807 Levin moved his family to near Natchez, to Washington, in the Mississippi Territory, where he hoped to find great need for surveyors.

In September of 1808, Seth Pease, the surveyor of United States Lands south of Tennessee, offered Wailes a job to survey the Choctaw purchase "with all the perfection which the compass and chain is susceptible of and should be happy to avail myself of your service in this particular," at $4 per mile, the maximum.

At the end of the survey, Wailes presumably sent one copy of the Journal and Field Notes to the Land office.

In April of 1810 Wailes was named "Register of the Land Office of the United States in and for the Western part of the Territory of Orleans". During the War of 1812, he served as a Lieutenant in a Louisiana outfit, and moved to Opelousas until 1822.

On January 10, 1822 he was appointed Surveyor General of Mississippi and ten days later, also of Louisiana. By fall of 1822, Wailes was "surveyor of the United States Lands south of Tennessee". His correspondence in that post was directed to him at Washington, Mississippi, which is about six miles northeast of Natchez on Highway 61. From information about a grandson, we know that the family remained around Natchez for some time.

In 1822 Wailes hired John James Audubon for six weeks to teach art at the Elizabeth Female Academy of Washington, and invited Audubon to Wailes' plantation in Louisiana (across from Natchez), where Audubon drew the orchard oriole and tyrant flycatcher, plates 42 and 79 of Birds of America.

In his last years Levin lived with his son Benjamin at Washington, Mississippi, until his death on May 24, 1847, in his eightieth year. Presumably he is buried near his children in the family graveyard, which is probably at his son's plantation "Meadvilla", just south of Natchez.

The Wailes' family was close to the Butler family of Natchez, and across the river in Feliciana Parish in Louisiana, and it might well have been Wailes who introduced that family to land at Hal's Lake, and the John Foster who owned part of Hal's Lake might well have been the same John Foster who founded Washington, Mississippi.

James Powell. Both the Powell and Bates families were living below McIntosh's Bluff before 1791.

James Powell, almost undoubtedly from this family, was born in Georgia and his family settled at Black's Bend. James Powell married Sarah Bates, but apparently not until after 1810. Sarah Bates was a mixed blood, but it is not clear whether James Powell had Indian blood.

Before 1805, Washington County poll tax records show that James Powell had 800 acres of land on the upper end of McIntosh's Bluff. In 1805 he owned one slave, three horses, 120 cows, and 200 acres of unimproved 3rd quality land on the west side of the Tombigbee at McIntosh's Bluff. By 1807 he still owned 200 acres and a slave, but the land was now shown as "4th quality", all suggesting that (like Thomas Bassett and many others then) he more likely than not hunted for skins rather than farmed.

After the massacre at Ft. Mims, James Powell commanded sixty men of the 8th Regiment of Mississippi Militia at Ft. Hawn at Gullett's Bluff, six miles south of Jackson, on the line of travel to Mt. Vernon. Corps of Engineer River Charts still show "Powell's Landing" in Washington County, at Mile 88.2 on the Tombigbee, about a mile downriver of the Alabama Electric Co-op steam plant at LeRoy.

It is not clear when James Powell and Sarah Bates Powell died, but neither is shown in the 1820 Alabama census. They had ten children, who undoubtedly grew up hunting and fishing the Hal's Lake area at a time when the Tombigbee ran clear.

Mississippi Antecedents. Before the surveyors got to Clarke County, they spent the period between February and late September of 1809 surveying west of the Tombigbee, mostly in what is now Mississippi. That part of the Journal is truly fascinating, involving visits by Indian chiefs and their armed guards, intransigence by Indians, and such fascinating mischief, but not of immediate interest to Clarke County.

This paper will focus only on the portion of the survey between the cut-off in the south, going north to just below Choctaw Bluff (anybody else interested can take it north from there).

Initial Plans. On Tuesday, September 19, 1809, Levin Wailes noted in the Journal that Dinsmoor had written from New Orleans, and Wailes wrote "I find I agree", "that is this is the most favorable season for surveying the extensive cane swamp through which the boundary lines must pass north of the thoroughfare [cut-off] uniting Tombigbee and Alabama Rivers above Nannehubba Island".

Wailes also wrote of his plan:

as it is likely we should have less perplexity in ascertaining the dividing ridge between the waters of those two rivers in ascending than descending it, it will be advisable to commence the running of the ridge line at the lower extremity of the Chaktaw claim.

Preparation. So the effort began that day, September 19, 1809: "Accordingly, I set out this afternoon with my company in our boat down the [Tombigbee] River".

On September 20, 1809 Wailes arrived at St. Stephens in the morning and met Mr. Dinsmoor, took on supplies, and headed down the river to "Bilboe's Landing". Wailes had arranged to meet Silas Dinsmoor at "Mr. Slade's", and arrived nearby at Wilson's landing and camped in the afternoon of September 22. The next day, Wailes sent an "express" [messenger] to Mr. Slade's, after learning that Mr. Slade's was only two miles by land but ten miles by water from where they were. The messenger learned that Dinsmoor hadn't arrived yet, so Wailes headed south to Carson's Ferry at the west end of the cut-off, at Poll Bayou, arriving early in the afternoon on Saturday, September 23. Not finding Dinsmoor there, Wailes went down to Fort Stoddart to get supplies, arriving late at night, returning to Carson's Ferry on Sunday, September 23.

"Final" Plan. Dinsmoor and Wailes agreed to hire John Carson, the cut-off ferryman, to go upriver and inland east "to the terminating point of high land on the dividing ridge", where at prearranged times he would "kindle fires, and raise smoke, from which we hope to ascertain the nearest and most proper direction through the low grounds north of the thoroughfare and between the two rivers. . ." And, "in the meantime, we shall explore the thoroughfare and the lakes which form communication between it and either of the rivers", indicating that people in that area and period used Hal's Lake to travel.

Final Preparations. On Monday, September 25, John Carson with two surveyors "sets out in a Perogue up the Tombigbee River", while Dinsmoor and Wailes in another boat set out through the cut-off, generally surveying that area and Doctor's Lake. It was too late to retrace their steps through the cut-off to spend the night at Carson's Ferry, so they spent the night at Mrs. John Pierce's near the Alabama River end of the cut-off; the day was "clear and pleasant".

They spent September 26 moving their camp from Carson's Ferry at the west end of the cut-off to "William Mimm's field on Nannahubba Island", and spent the 27th determining magnetic deviation.

On Thursday, September 28th, bad news: John Carson came back and, though he had gone to high ground above the swamp, he had misunderstood the times for building fires, so no observer saw them. Wailes and Dinsmoor decided that, since Dinsmoor had to go to St. Stephens to pay the "Chaktaw Annuity" [annual payment for the lands in the purchase], the money for which he had picked up in New Orleans, Dinsmoor would himself go try to find the ridge north of the swamp and build the fires, since "we deem the ascertaining the proper course through the swamp by this means of so much importance. . ."

Actual Survey Begins: The South Cedar Post. On Friday, September 29, 1809, the actual survey began, when Wailes and his crew found a point about half way between the Tombigbee and Alabama ends of the cut-off. There, on the north bank, they put up a cedar post, marked "US" on the west, "MT" for "Muscogee Territory", on the south side, "SD 1809" [for Silas Dinsmoor and the year] on the east, and on the north side "LW" [for Levin Wailes], "round which we threw up a mound." This was the starting point, from which they began to run due north, returning in the evening to their camp on Nannahubba Island on the south bank of the cut-off. Second Failure at Sighting Fires in the North Makes the Job Look Tougher Than They Thought. From Saturday, September 30, to Monday, October 2, which had mild mornings and was generally hazy, Wailes posted "a regular look out", including several at the agreed hours "in the tops of tall trees", but they saw nothing, "except of a broad smoke about the horizon in the north from which no direction could be taken".

The next day, October 3, Wailes sent his boat to split the provisions, part at Mr. Wilsons, "and the residue at a Major Christmas', to be convenient to supply us on our way up the ridge."

The rest of the Wailes party took their own packs and crossed to the north bank of the cut-off to continue the survey line north and to move the camp to the north bank.

That evening William Thornton arrived, an "express" [messenger] from Mr. Dinsmoor, with Dinsmoor's letter of October 2, which advised that the fires were built at the proper times and made "a grand appearance"; Wailes wrote "our failing to discover them gives us reason to apprehend that the distance through the swamp is greater than it has generally been estimated at."

"The Crescent Lake". Silas Dinsmoor, who had been talking with and meeting people in the area, then broached the subject of "the crescent lake":

Mr. Dinsmoor also informs, that hearing of a Lake in the form of a crescent extending into the swamp from the Tombigbee River, which will be likely to intercept our north course, he has employed James Powell to accompany Martin Crane in a canoe up the lake to enable us more conveniently to explore it and judge of the course that will pass between it and the Alabama waters. Powell is to sound a horn to direct us to him.

The clearest "crescent" lake is what we now know as "Fishing Lake", mostly in T3N, R1E, 2, just east of the Tombigbee, due east of McIntosh, and northwest of the mouth of Hal's Lake on the Tombigbee. However, that lake -- however perfect a crescent shape it clearly is -- doesn't extend into the swamp "from the Tombigbee river", and it was not "likely to intercept [their] north course". James Powell, however, who lived there, was likely to know what he was talking about on both those points, thus indicating that the "crescent lake" was what we now know as Hal's Lake.

The next day:

We heard the sound of Powell and Crane's horn in a northwestwardly direction from us, answering them, they join us and become our pilots to their camp where we take breakfast and ascertain that the lake curves gradually to the N.W. and that our mile post list noted is about 30 chains east of its eastern most verge. We direct Powell to continue up the lake about one mile, and returning to the milepost, we continue our north course.

In all likelihood Powell was then canoeing on a lake not named on the maps, but is in T3N, R1E, 13, which connects to an unnamed creek or creeks which at both ends flows into the Tombigbee just across the river from Bilbo Island, and just across the river from Bates Lake in Washington County.

When this lake petered out to the west, they started looking more to the north, where [what we call] Hal's Lake lies, on October 5th writing in the Journal that:

Finding the course of this lake diverge from that of our line too much to make it useful to continue the canoe any longer in it, and learning from Mr. Powell that a creek or arm of Bates Lake, more north extends further to the East than this, we direct him to return to the river and ascend that arm of Bates Lake to the point which he may judge we shall approach it nearest by a continuation of the present course, and returning to. . . [the last mark] we continue our course due north. . . .

That "creek or arm of Bates Lake" they were seeking must have been Hal's Lake; nothing else fits. What we today call "Bates Lake", of course, is west of the Tombigbee, but it was not uncommon in those days for creeks with a single name to come into a river from both sides, witness Bassett's Creek.

Late the afternoon of Thursday, October 5th, the Wailes survey party reached the south bank of what was probably an offshoot of what we now call Hal's Lake:

To south bank of a creek or gut 4 feet deep and about 80 links wide slow current moving to the right probably the influx of the tide of Tombigbee River.

Hearing the sound of Powell's horn in a direction a little north of west we shape our course thither and arrive at his camp about the dusk of evening well drenched in consequence of a showery afternoon.

The next day, Friday, October 6, 1809, was a day not for surveying, but for exploring Hal's Lake in a canoe:

Finding ourselves encompassed by Lakes which we must avoid crossing to keep between the waters of the two rivers, set out with Powell and Crane in the canoe and explore several branches of Bates Lake, and the afternoon being rainy, we return to camp at night very wet.

The next day, Saturday, was back to surveying, sending Powell off to get food:

We attempt to ascend the creek on which we discontinued our line [ed. note: the small tributary] on Thursday the 5th and find it so obstructed by drifted wood as not to be even navigable in our small canoe. Despairing of any further advantages from continuing the canoe in this quarter we direct Powell and Crane to proceed up Bate's Lake to a bluff near Powell's whence we shall expect Crane to penetrate the swamp and endeavor to meet us with a supply of provisions.

They surveyed northward that day, Sunday, October 8th, and finally "camped on the creek about ten chains north of our line". At this point, they were doubtless having trouble finding anything much like a watershed in the swamp south of Hal's Lake, and on October 8-9, surveyed generally eastward, crisscrossing gullies and guts.

Now, time for another day exploring Hal's Lake, rather than surveying, Tuesday, October 10, 1809:

Our provisions being expended we despatch William Knight and William Thornton back on our line for a supply, myself and three others of our company employed until 2 PM in exploring the lake and several of its branches and having ascertained that it is fed by the Tombigbee River, we return to the cypress, post set last evening. . . .

That evening, Thornton came back at dark with some cornmeal and beef, and reported that Knight was waiting to procure more.

On Wednesday, October 11, Wailes and his group kept surveying, then back to camp, and "William Knight returned in the evening with some fresh pork, coffee and sugar", doubtless to a hero's welcome.

On Thursday, October 12, the surveyors continued on mostly north and east, until they were south of the eastern extremity of the Lake, and ultimately, to around what is now called "South Lake" at T4N, R2E, 34 or thereabouts, and returned to their last camp; the next day was similarly spent.

At this point they had surveyed far enough east that they were pretty well clear of Hal's Lake but they were out of provisions; having earlier heard Powell's horn, they headed up what we now call "South Lake Drain" toward the horn, the bluff on the north of the Lake, and food:

Having heard the horn on Thursday afternoon [three days earlier] and seeing a way recently cut through the cane, we conjecture that Crane has penetrated the swamp thus far and returned. And as we are again out of provisions and despair of any being brought us by him, we determine to leave the line, and pursue the marks which we suppose will direct our horse camp on the high land. . . we arrive at a Lake about two chains wide lying nearly north and south, and continuing along it about one mile, we come to its junction with Bates Lake, . . . about five chains [330 ft.] wide; we cross it on a raft of cane, about two hundred paces above or S.E. of a bluff on the N.E. side which is known by William Thornton (one of our company) to be not more than five miles from James Powells, whither directing our course, we meet Mr. Dinsmoor, the Choctaw deputation, and our pack horsemen. After travelling about two miles, and returning with them to the Lake we encamp and get refreshment. The day has been clear and pleasant.

On Sunday, October 15, 1809, Wailes left the survey in the hands of Mr. Dinsmoor, because of "domestic concerns requiring my absence for a few weeks".

For about two days the group surveyed and camped on the north side of the Lake.

On October 17th Dinsmoor, searching for water, first found the Alabama River, where:

I meet some Alabama's [Indians] and contract for a passage to David Tates a quadroon of these [illegible], educated in England where I arrive about 10 o'clock.

Spending the night, on the morning of October 18th -- "Foggy morning, clear day" -- he pumped Tate:

In a conversation with Mr. Tate and others I find he can give me very little information about the swamp, he says the Alabama is very crooked and its general course ascending is north and that the distance from our last post to Bates Place [sic: Lake?] cannot be great.

Here leaveth this paper from the survey. Anybody wanting to trace the survey team north of Hal's Lake is welcome to ask to use one of the copes which the author has donated to the Clarke County Historical Society, one of which is kept at the Clarke County Museum in Grove Hill.

Page last updated: 2012-02-14 00:59:03 PST

Search Site

Clarke County, Alabama

Clarke County, Alabama

Clarke County is a county of the U.S. state of Alabama. The county was created by the legislature of the Mississippi Territory in 1812. It is named in honor of General John Clarke of Georgia, who was later elected governor of that state.

Read More

Affiliated Sites

The USGenWeb Project


The ALGenWeb Project