History Corner: Major Ellicott’s Triangulation
Gregory Spies, PLS

Part 1 of this article appeared in the August issue.

The Triangulation
"On our arrival at the end of the compass line on the Mobile River, one serious difficulty presented itself, that was the continuation of the line through the swamp, which is at all times almost impenetrable; but at that season of the year absolutely so: being wholly inundated:- But, fortunately we found in the neighborhood of our camp a small hill, the summit of which was just elevated above the tops of the trees in the swamp. From the top of this hill, we could plainly discover the pine trees on the high land, on the East side.
Upon ascertaining this fact, we sent a party through to the other side, (along the water courses, by which the swamp is intersected in various directions), with orders to make a large fire in the night with light-wood; the same was likewise to be done on the hill before mentioned, to obtain nearly the direction from one place to the other. The atmosphere was much too filled with smoke, to discern a flag, or other signal, the woods being on fire on both sides of the swamp.- It happened unfortunately that the day before our fires were to be lighted, the fires in the woods had extended over almost the whole of the highlands, on both sides of the swamp; by which so many dead trees were set on fire, that there was no possibility of discriminating between them, and our fires. -It was then agreed that the parties should light up, and extinguish their fires a certain number of times; making stated intervals. -This succeeded so well, that we became certain of not taking a wrong fire in determining the angles. -Contrary to our expectation, a heavy rain fell on the same night, a short time after we had finished the experiment, and extinguished all the fires in the woods,- The storm cleared off with a strong north-west wind, which carried off all the smoke, and enabled us to determine the angles in the day, by erecting signals, which was accomplished on the second day of April. -This work was connected with the observatory in the following manner. At the observatory A (see Figure 1) a meridional line was traced, by taking the greatest elongations of Ursae Minoris, both east, and west, with the transit and equal altitude instrument: -equal distances were carefully measured in each direction, and a fine mark placed at the termination of each measurement, -the distance between those marks was accurately bisected, and a fine mark placed at the point of bisection for the meridian. The same operation was performed a second time, and although the difference in result, appeared too trifling to need any attention, it was nevertheless bisected, and that point of bisection taken for the meridian, -which designated by AE and terminated by a parallel of latitude drawn through B. -From the point A, a vista was opened to the summit of the hill at B: from B, to C, another vista was opened, which formed the base: the base was too short if it could have been avoided; but the hill would not admit of its being any longer. -D the signal on the east side of the swamp. -The angles were measured on the horizontal arc of the astronomical circle already mentioned. -This instrument by means of a vernier is graduated to 5," which by the help of a microscope may be easily subdivided by the eye, into 1 1/2 or 2 seconds. The measurements, and angles stand as below.

AB = 310.8 perches
BC = 70.356 perches
BAE = 37 58' 48"
ABD = 57 43' 21"
BCD = 139 23' 58"
DBC = 39 47' 1"
CDB = 0 49' 1"

From this data, AE is found to be equal to 244.9 perches, BE to 191.26 perches, BD to 3211.65 perches, EF to 2987.44 perches, and DF to 316.7 perches. DB being considered as an arc of a great circle, forming with the prime vertical an angle of 5 42' 9" to the north, being the excess of angles BAE and ABD above 90. From the result of the observations for the latitude, the observatory appeared to be too far north by 518.55 perches, which is designated by AH. It therefore follows, that the signal at D, was too far north by the sum of the distances DF, EA and AH, which is equal to 1080.15 perches: this distance was measured due south from the point D, and would intersect the parallel of 31, at the end of 215 miles and 169.6 perches from the high water mark on the Mississippi. From the termination of the abovementioned 1080.15 perches, another guide, or compass line was continued East, to the East side of the Coenecuch."

Thus Ellicott, by triangulation, prolonged the line of demarcation between the U.S. and Spanish West Florida across the flooded Mobile-Tensaw Delta into Creek territory. This mathematical projection of the boundary line across the swamp might have sufficed had not Ellicott been limited by his selection of the "small hill" as the site for triangulation baseline. Today this hill is named Andry Hill after Seymour Andry, a Creole who had obtained a Spanish land grant in the vicinity of Seymour's Bluff (c. 1790.) The summit of Andry Hill is less than one quarter of a mile in length. Ellicott's base line along the most elevated portion of the hill was 1160.87 feet long. This was the baseline of Ellicott's only triangle that was fully observed in his triangulation network. The other two legs of his triangle were roughly 10 miles in length which created an extremely acute angle on the other side of the delta at the signal point D. As anyone versed in the art of triangulation would attest, this was, indeed, a very weak triangle.
Ellicott's triangulation was perhaps the only way possible to project the boundary across the Mobile-Tensaw River Delta due to several factors. The spring floods had inundated the delta bottomlands with 18 feet of flood water. Topographical limitations caused Ellicott to select Andry Hill to establish his baseline. His scheduled meeting with U.S. Indian Agent Benjamin Hawkins and the Creek chief Mad Dog in Pensacola, on April 20th, left him scant time to devise another plan. It can be said, perhaps, that he did the best he could do under the circumstances.

The Accuracy of Ellicott's Triangulation
A modern analysis of the accuracy of Ellicott's methodology and calculations, however, reveals some gross errors introduced by the single, weak triangle he used to accomplish pushing the line into the Creek Nation. This, coupled with the inherent inaccuracies resulting from his astronomical instruments, observations, and calculations caused some major deviations from the stated goal of marking the 31st degree of north latitude on the ground. By using the known geographical positions of the Ellicott Stone and Mound No. 216 located in Stockton, Alabama on the east side of the Mobile-Tensaw River Delta, the accuracy of Ellicott's triangulation can be assessed. The Ellicott Stone is located at Point H in his Figure G. Mound 216 is located just to the east of the terminal point where line D-F (Figure 1) was prolonged "due south" to intersect Ellicott's 31st parallel. Therefore, according to Figure 1, Mound 216 should be due east on the "31st parallel" from the Ellicott Stone.
The NAD (83) geographical coordinates of the Ellicott Stone are 30º 59' 52.09384 north latitude and 88º 01 21.06684 west longitude. In retrospect, it can be seen that Ellicott's calculations derived from his observations at the astronomical station on Seymour's Bluff (his Point A) led him to erect the Ellicott Stone (his Point H) approximately 799 feet south of the actual 31st parallel. The triangulation did not factor into this calculation. However, the triangulation was tied to the astronomical station and the Ellicott Stone only through the traverse along his Line A-B.
The point of intersection of the prolongation of his line D-F and the line drawn between the Ellicott Stone and Mound 216 in Stockton is located at the NAD (83) geographical coordinates 30º 59' 46.13515 north latitude and 87º 51 38.36021 west longitude. This point of intersection is located approximately 1401 feet south of the actual 31st parallel and approximately 602 feet South of Ellicott's calculated 31st parallel prolong-ed east from the Ellicott Stone.

The Ellicott Stone Selected as the Initial Point
The line of demarcation as monumented by the boundary commission is consistently south of the actual 31st parallel and makes a dip to the south (approximately 602 feet) as it crosses the Mobile-Tensaw River Delta. In 1805, compounding the errors inherent in Ellicott's triangulation and astronomical determinations, Isaac Briggs, Surveyor General of Lands South of the State of Tennessee, instructed his deputy surveyors to select one of the "mile markers" on the U.S. southern boundary as an initial point to begin the first U.S. Public Land surveys in the Land District East of the Pearl River, Mississippi Territory in the tract where Indian title had been extinguished by the Choctaw Treaties of 1803 and 1805. His deputy surveyors selected the Ellicott Stone as the initial point for what today is known as the St. Stephens Meridian. The deputy surveyors were evidently unaware that the stone monument was not one of Ellicott's mile markers as they were post mounds numbered east from where the 31st parallel intersects the high water mark on the east bank of the Mississippi River. Rather, the Ellicott Stone was set 194 perches east of Mound 206.
The selection of the Ellicott Stone as the initial point caused numerable problems for the deputy surveyors who began the township surveys. None of the post mounds on the line of demarcation could be used as township or section corners. Later, in the 1820s, the St. Stephens Baseline was extended to the east and Township 1 North, Range 1 East and Township 1 North, Range 2 East were sectionalized through the Mobile-Tensaw Delta. When the baseline was run through the swamp and onto the high ground near Stockton, the deputy surveyors realized that Ellicott's boundary, the Mound Line, was some 850 feet south of their baseline.
To remedy this discrepancy between their survey and that of the U.S.- Spanish boundary commission, the deputy surveyors shifted Township 1 North, Range 3 East approximately 750 south. This correction on their part, to make the south boundary of the township coincide with the Mound Line, created a 750-foot jog in the St. Stephens Baseline. However, from Stockton east to the Chattahoochee River, the St. Stephens Baseline and the Mound Line remained separate lines, in part due to the initial confusion related to Ellicott's triangulation and the lack of knowledge among the U.S. Deputy surveyors concerning Ellicott's survey of the Spanish boundary.
The failed attempt, on the part of the U.S. deputy surveyors, to make the St. Stephens Baseline coterminous with Ellicott's line of demarcation has bred confusion ever since and actually caused the U.S. Congress (c. 1850) to call for a dependent resurvey of that portion of the Mound Line that forms the Alabama- Florida state line. Ellicott's triangulation, made in haste, under adverse conditions and the imminent threat of Indian reprisal, has to be recognized as one of the chief catalysts for all the confusion and seemingly unexplainable anomalies found along the old line of demarcation today which are witnessed by property boundary disputes, misplaced section lines, and, yes, even the continuing debate concerning the obliterated boundary between the States of Alabama and Florida. However, these stories must wait until another time.

Ellicott's Perception
But, we can take solace in the fact that Ellicott perceived that problems related to the work of the U.S.-Spanish boundary commission might come to light in the future. With the finesse of a true Federal bureaucrat he wrote for posterity:
"Measurements when accurately executed, in a known parallel of latitude, are generally preferable to observations of distances, not exceeding 100 miles; yet in this case, the measurement is not entitled to that weight, being done in haste, with a common chain, through thickets, swamps and ponds, where pins of more than ordinary length had to be made use of, which involved an insurmountable source of error: but not so considerable a degree as to justify its rejection."
Ellicott waxes further with his apologetic rhetoric in an effort to mitigate any adverse effects that errors propagated by the boundary commission might cause for future surveyors:
"It is presumed, that no apology will be necessary, for any small inaccuracies which may be discovered in the astronomical observations, when it is considered that they were made at temporary stations, and the apparatus frequently exposed to the weather, for want of tents, and other covering; and almost as frequently so injured by the transportation from one place, to another, through the wilderness, that if I had not been in the habit of constructing, and making instruments for my own use, our business must have been several times suspended, till repairs could have been made in Europe."
As for the commission's business in the vicinity of Mobile, it was finished and the instruments were taken down on April 10th and securely stored onboard the Sally. The next day, Ellicott ordered his surveyors to once again cross the flooded delta to begin the third compass line running east to the Conecuh River through the Creek Nation. The Sally then descended the storm swollen Mobile River headed toward Pensacola leaving behind the Ellicott Stone to stand in silent witness of Major Ellicott's visit to Mobile while monumenting his strange triangulation across the swampy river bottom that served as the boundary between the Choctaw and the Creek Nations.

Greg Spies is an adjunct professor at Troy State University in the Geomatics Program—Math and Physics Department. He authors a column titled "Retracing the Bounds" which appears in the ASPLS newsletter and he has also written for the ACSM Bulletin and Backsights, the biannual publication from the Surveyors Historical Society. He is also a Contributing Writer for the magazine.


Note: Most of the articles have images and/or graphics.
To view these, we suggest that you subscribe to the magazine.

HOMEPAGE ARCHIVES


Copyright 1995 - 2004 by GITC America, Inc, Inc. Articles cannot be reproduced,
in whole or in part, without prior authorization from GITC America, Inc, Inc.