Boston to Fulton Road
That there was a pre-Civil War route from Boston (now “Old Boston”) to Fulton is indicated by surveyor calls in two head right surveys first noted by Gary Pinkerton (see TrammelsTrace.org). These were an 1859 survey (Bowie County Abstract #423) and and an 1861 survey (#419). A third, the 1855 Allen Brazzeal Survey near Boston, says the NW corner of the Survey is “on the Fulton Road.” The route as portrayed here came east from Boston across what is now the Red River Army Depot and likely met or crossed the route of Trammel’s Trace near today’s Nash. Several early maps of Texas, including Jesus de Cordova’s 1853 map from the Library of Congress, show a route that looks much like that in our Google map and are consistent with the two head right surveys. How much earlier this segment of a Boston to Fulton route was used is uncertain.
The route shown in our Google map goes through these three survey points and otherwise follows roads shown on the 1918 Bowie County Soil Map and also the 1908 USGS Texarkana and New Boston quadrangle maps. This route is still called Old Boston Road in current maps as it cuts across Red River Army Depot (although it has been obscured in places by RRAD roads and operations) and east of RRAD until it reaches Nash where it meets the current New Boston Road, US 82. It would also have crossed the route of Trammel’s Trace near this point. Whether the “Boston to Fulton Road” as called in the surveys then went up Trammel’s Trace is an open question as Trammel’s Trace had likely fallen into disuse at the time of the surveys. The “Boston to Fulton Road” as called in the 1850s surveys more likely would have gone more straight east, following the route of the current US 82 more or less, towards Rondo then up to Fulton, as the currently designated historic Southwest Trail route goes..
It does seem very likely that the route shown in our Google map, or a very similar one, at least to the crossing of the Trammel’s Trace route near Nash, is that of the Boston to Fulton Road as called in the survey notes and shown in contemporary and later maps, but there may also have been another route, the Boston to Mooresville Road, south of the one drawn here and perhaps used earlier. The Republic of Texas established a new post route in late 1841 to “commence at the county seat of Bowie County, and go via Colonel Charles Morris’s and David Lane’s to Fulton, Arkansas.” The De Cordova map and others show the route as shown in our Google map going well north of Mooresville (the homestead of Charles Moores–the “Morris” is in error), which is where a post office is shown in other maps, so this implies that the “Moores” route was no longer in use by 1860. The “David Lane’s” location indicated in the post route designation is curious, but the 1840 survey notes for the George Brinlee head right mention that the David Lane head right is on the eastern boundary of the Brinlee head right, although that is not the name on this head right as of 1840. It is quite possible that David Lane (a surveyor for Bowie County) did occupy this head right in 1840 but was not able to patent it. So, the “David Lane’s” noted in the post route designation might well have been on this head right.
The 1841 John Arrowsmith map of Texas from the Library of Congress shows a road going west across to Clarksville from Lost Prairie south of Fulton and could be interpreted as going through what was later called Mooresville. It is possible that the 1841 post route used a trail suitable only for horses, and this would not have been cleared for wagons and fell into disuse later. The Arrowsmith map does not show Trammel’s Trace coming across from Fulton as it is thought to have gone, so this adds some uncertainty to this maps accuracy. However, other maps show a road going across to Lost Prairie, which is shown as having a Post Office, then crossing the Red River at Dooley’s Ferry and either up the north side close to the River or up to Springhill and, in either case, then to Fulton.
The route east of the “Boston to Fulton” Road from where it crossed Trammel’s Trace is uncertain. Trammel’s Trace was beginning to fall into disuse by the early 1840s (see Pinkerton, Trammel’s Trace, pp. 195-198). Also, the Boston to Fulton road as drawn in our Google map was probably not used until after this time, as no settlements or cemeteries earlier than 1861 along this route are known and the two head right surveys calling the “Boston to Fulton Road” are from 1859 and 1860.
It seems that the more likely route east, at least by 1860, would have gone across to Rondo and then up to Fulton from there. How this route went through what became Texarkana in 1873 is also uncertain, but what seems probable is that it followed closely to what is now US 82, a route, at least east of State Line Avenue, that has been designated as part of the Historic Southwest Trail. A hint of “Old Boston Road” is also found in a Texarkana street by that name between New Boston Road and Robison Road. So, we have drawn a conjectural route (in blue) as shown in the Google map excerpt below that follows US 82 more or less through Texarkana until reaching US 67. The actual route across what is now the airport is unknown, but our map speculates that it would have followed 19th Street around the bottom of the airport then the “Old Post Road” (CR 25) until reaching the vicinity of Rondo, which was in existence well before 1840 as the Old Rondo Cemetery was established at least as early as 1839, the year of the earliest burial known. From here the assumed route goes up the Old Post Road until reaching Dooley’s Ferry Road and then on east to Lost Prairie and Dooley’s Ferry, then up the north side of Red River until reaching Fulton. The route could just as well have continued up what is still called the Old Post Road until reaching Trammel’s Trace during the period that route was still in use.
Below is an excerpt from the 1864 Red River Campaign map from the Library of Congress. The routes shown from Fulton to Dooley’s Ferry and from Dooley’s Ferry to what must have been the Rondo area match well what is shown above.
Our Google map of the route from Dooley’s Ferry uses the “River Road” on the north side of Red River, but it seems likely the route would not have been usable in very wet weather and the route up to Spring Hill and then west to Fulton would have been better. In any case, the Boston to Fulton Road, by whatever name, could have followed several different routes over the years, and probably did.
The Boston to Mooresville Road
It does seem very likely that the “Boston to Fulton” route shown in our Google map, or a very similar one, at least to the crossing of the Trammel’s Trace route near Nash, is that of the Boston to Fulton Road as called in survey notes and shown in contemporary and later maps, but there was another, earlier route, the Boston to Mooresville Road, south of the later Boston to Fulton Road. The Republic of Texas established a new post route in late 1841 to “commence at the county seat of Bowie County, and go via Colonel Charles Morris’s and David Lane’s to Fulton, Arkansas.” De Cordova maps and others show the later Boston to Fulton route as discussed above going well north of Mooresville (the homestead of Charles Moores–the “Morris” is in error), which is where a post office is shown in other maps, so this implies that the “Moores” route was no longer in use by 1860. The “David Lane’s” location indicated in the post route designation is obscure, but the 1840 survey notes for the nearby George Brinlee head right mention that the David Lane head right is on the eastern boundary of this Brinlee head right, although that is not the name on this head right as of 1840. It is quite possible that David Lane (a surveyor for Bowie County) did occupy this head right in 1840 but was not able to patent it. So, the “David Lane’s” noted in the post route designation might well have been on this head right.
A route very similar to the one below follows current roads also shown on the 1906 USGS New Boston Quadrangle Map. Part of it is also shown on the plat map in the 1854 S. D. Elliott Survey, with two crossings of the Survey boundary called, as shown here where the road crosses Elliott Reservoir on Red River Army Depot. This route or something similar to it is likely the 1841 postal route from Boston to Fulton through Charles Moores’ post stop, which connected to Trammel’s Trace near what came to be called Mooresville. Going west the route connected to the slightly later Boston to Fulton Road that went across on a more northerly course, connecting to Trammel’s Trace near today’s Nash and going east to Rondo and from there up to Fulton. This route goes through the community of Rock Creek, whose cemetery has an earliest recorded death year of 1836.
Clarksville to Dalby Springs Road
This route, which also went on south to Stephenson’s Ferry and Trammel’s Trace, or, alternatively east to Boston, is shown on an 1870 Red River County Cadastral map from the Library of Congress website. We have shown the route as following existing roads for the most part, although the route shown on the 1870 map seems to vary only slightly from this. The roads as shown here seem to be very old roads for the most part when driven currently. How this route was different, if at all, from the earlier Clarksville to Stephenson’s Ferry Road is not understood at present.
Clarksville to Mill Creek (Berlin) Road
A Clarksville to Mill Creek Road is mentioned in several sources. A road is shown on an 1857 military map available at the Library of Congress website that apparently went from Clarksville up to the vicinity of Mill Creek and a crossing to Laynesport on the Arkansas side of Red River. Both Pine Springs and Spring Hill in Bowie County were stops on a stage line between Clarksville and Mill Creek (also known earlier as Burkham’s Settlement and very near the Berlin landing and town site on Red River) . The road would also likely have gone through the English community. Otherwise, there is not much evidence, such as head right surveys, as to the exact route. The route here is drawn to follow the early roads shown on the 1918 Bowie County Soil Map and the 1923 Red River County Soil Map. Highway 114 from Clarksville to US Highway 259 is a very good paved road and a beautiful drive, following a likely 1840s route, through prairies and pastures, passing through the English community (see separate entry), established in the 1840s. Also, the road passes by the Pine Springs and Spring Hill communities and cemeteries. Spring Hill cemetery was established in 1845 as the sign there says, and is worth a stop. East from 259 on CR 2735 towards where the Mill Creek Settlement and Berlin would have been becomes a good gravel road but ends up on private land and can not be followed to the end.
“New” Clarksville to Mt. Pleasant Road (1853)
This old road is first mentioned in survey notes as being “new” in 1853. The few Red River County survey notes found to this time that mention the road generally show it as being directly on or close to current FM 990 and CR 1412 (the latter is a particularly old road). The route shown here follows these old roads except near the Sulphur River crossing, where the old road is known to have been to the west of the current bridge. The old iron bridge (the “Meadows Curve Bridge”) still in existence halfway across the Sulphur and a few hundred yards upstream of the existing modern bridge was reportedly built in the 1930s. The route shown here in Titus County follows the existing Titus CR 1905 and is consistent with the sparse known evidence. The actual original route likely varied from this current road slightly but in general must have been in relatively close alignment. For those wishing to follow this old route all the way through from Clarksville to Mt. Pleasant, a high-clearance, preferably four-wheel drive, vehicle is recommended, particularly in wet weather as the road can be very rough. It could also be flooded in some areas following heavy rains. However, it is well worth the trip as it is one of the few old roads in this area that for the most part follows the route of a pre-Civil War road and is probably not all that much changed.
DeKalb to Clarksville Road
That there was a road or at least a route from Boston (Old Boston now) to DeKalb and on to Clarksville beginning around 1840, if not earlier, is very likely. The Republic of Texas identified (Old) Boston to DeKalb and Clarksville through Savannah as a post route in 1841. Such a road is shown on some of the earliest more-accurate maps of Texas (see, for example, the 1849 Jesus de Cordova map (and later De Cordova maps) or the 1857 Pressler map, (both available at the UNT Portal) even though little detail can be discerned from these maps. It is reasonable to believe that this route is what is sometimes called “The Ridge Road,” as shown on some maps, such as the 1945 map by James Dawson that is available at the Southwest Arkansas Regional Archives at Washington, Arkansas. If so, this would also have likely been the route known as the Chihuahua Trail followed by the Chihuahua traders with Henry Connally in 1840.
Here we have chosen to follow roads shown on the 1923 Red River County Soil Map, except for one notable gap south of Annona where no old road is known. Although the general shape of the route is reasonably well known from pre-1860 maps, the details are not, and we have found no other confirming evidence, such as head right surveys, to this point. Thus the details of the route shown here are uncertain; however, the old roads shown from the 1923 Soil Map certainly appear to be reasonable candidates for the early 1840s route. The route as shown can be driven east from Clarksville (high-clearance vehicle recommended) until reaching the point where CR 4205 turns north. Although the Google map shows CR 4217 continuing east from here, this is no longer possible to drive as there is a gate and private road going east at this point. CR 4205 can be driven north from here to US Highway 82. There is also a stretch past this point where there are no known roads, even on the 1923 Soil Map, that continue.
Jonesborough to Nacogdoches Road
Although it was in use much earlier, perhaps at least by the early 1820s, one of the earliest known mentions of a Nacogdoches to Jonesborough Road is in the “Laws of (the Republic of) Texas.” In 1837 the Republic’s Postmaster-general was asked to establish a number of mail routes, including a route “from Nacogdoches, via Epperson’s Ferry on the Sulphur Fork, to the county seat of Red River County, Jonesborough….” (See W. L. Newsom, “The Postal System of the Republic of Texas,” Southwestern Historical Quarterly 20 (October 1916)). Whether this postal route as authorized was ever used is unknown, as no other references to this particular route nor roads from Epperson’s Ferry to Jonesborough are known. although there was a road from Stephenson’s Ferry to Clarksville (and on to Jonesborough) at least by the 1850s and probably much earlier.
Eventually, there were likely at least two roads by the “Jonesborough to Nacogdoches” name, the earlier one going west of Mt. Pleasant and the second going through Mt. Pleasant, then going on to meet Trammel’s Trace, probably near what is now Hughes Springs and/or Jefferson. The route shown in our Google map approximates what we know about the “original” Jonesborough to Nacogdoches Road. That this was known by this name is shown in a number of notes for land grant (head right) surveys along the route as shown.
In Red River County this, or a similar route, apparently was also sometimes called the “Spanish Trace” or the “Spanish Trail” by some of the early Anglo settlers. This identification may have come about from stories that the route was used by “Mexican” traders in the 1830s and probably earlier. One such report was given by Martin Ragsdale who at the age of four came with his family to settle “three miles below Jonesborough.” Ragsdale, testifying in 1893 in the Greer County Supreme Court case, told of how Mexican traders followed one of two “Santa Fe to Natchitoches” roads going west to east (and returning). These were a river road and a prairie road, according to Ragsdale, with the prairie route going west towards what was later Paris. He also mentioned the “Trammell trace”, which “ran from Jonesborough nearly due south, and was traveled by a good many wagons.” It is unclear as to whether he means that Mexican traders used this route, although he said he “saw Mexicans at Jonesborough trading, who said they had traveled the prairie route.”
J. W. Williams, in a Southwestern Historical Quarterly article (January 1944) on the National Road of the Republic of Texas, states that the “first settlers of Jonesboro had no travel facilities except by a few Indian trails and a dim path known to them as the Spanish Trace.” Here he references A. W. Neville, The History of Lamar County. It is unclear as to how this relates to the possible route of a “Spanish Trace” from the Stephenson’s Ferry area up to Pecan Point as shown in, for example, the James Dawson 1945 maps available at the Southwest Arkansas Regional Archives in Washington, Arkansas. Skipper Steely does not mention this Pecan Point “Spanish Trace” when he writes that “trails went due south from the Kiomatia Plantation and Jonesboro, merging northwest of Clarksville and running just west of the present town. The main route went on south, passing just east of modern day Johntown….”
Farther south the route shown here was also often referred to as the “Cherokee Trace,” which seems to have split off from the ongoing road to Nacogdoches somewhere south of Gilmer, where the Nacogdoches road is once again called the “Jonesborough Road.” The John Arrowsmith 1841 map, available at the Library of Congress website, shows a road running from Jonesborough all the way south to Nacogdoches, passing what appears to be several miles west of Clarksville. One of the other stories about this route is that in 1832, Sam Houston, coming across the Red River at Jonesborough, spent several days with the Clark family, then, according to Strickland in “Red River Pioneers” rode south to Nacogdoches. Strickland believed he followed the usual route south to Nacogdoches but cut over to Trammel’s Trace on ‘the Choctaw Trail,” also called the Caddo Trace. In any case, the Jonesborough to Nacogdoches Road seems to have been well-known and well-travelled, particularly as an immigrant route, and used as an alternative (combined with the Fort Towson Road from Fulton) to Trammel’s Trace when the latter was unusable due to weather and mud.
Trammel’s Trace from Fulton to Old Unionville
The forging by Nicholas Trammel, in the 1820s, of Trammel’s Trace, the “first road to Texas from the north,” as the title of Gary Pinkerton’s book states, was a history-making accomplishment in that it connected, with a road passable by wagons, the Southwest Trail from St. Louis with the Spanish/Mexican world beginning at Nacogdoches, Texas. This provided a continuous trade and immigration route along an almost continental reach, coming by the Ohio river from the American east to St. Louis, then by the Southwest Trail and Trammel’s Trace to Nacogdoches, and then west along the Camino Real to San Antonio and down to Mexico. The eastern route from Fulton, Arkansas to the intersection, near Old Unionville south of the Sulphur River, with the western route, sometimes called the Spanish Trace, as shown on this map is taken directly from Gary Pinkerton’s map here (and thanks much to Gary for making this available!). This map relies on a number of head right survey notes available by searching the Texas General Land Office website). Information from these survey notes as provided by Gary for Trammel’s Trace is indicated on this map by brown square icons. Although the survey notes provide the most accurate and precise documentation of Trace crossings, the route between known survey points must still be considered as an approximation, and additional information is always being sought. One segment of the route from just east of the Texas state line to a few miles west is shown on some early head right (cadastral) maps, which helps provide some verification for this stretch. Other confirming evidence can be provided by Trace ruts that are still visible in a few areas, including at the Historical Marker on Highway 77 east of Douglasville.