A number of Mississippian period sites in the Middle Cumberland region are known to be located at natural springs. One of these is the famous mound site (40SU14) at Castalian Springs, Tennessee, and another is the equally well-known Gordontown site (40DV6) in south central Davidson County, Tennessee (Moore 1998:7). Archaeologists who work in the region have rightly determined that the presence of such springs was a major criterion in selection of these sites for settlement. The springs were sources of clear, clean, and potable water for domestic purposes. While highlighting the importance of these factors, the archaeological literature has been silent on the hydrogeological characteristics of the springs that served regional Mississippian sites and the amount of fresh water they were capable of delivering to their populations. This article examines the hydrogeological characteristics of one such spring at a little-known Mississippian site (40SU224) that was not formally identified, surveyed, and recorded until just 7 years ago. The author did a reconnaissance survey of the site in 2004-2005 and completed a state site survey form for it on May 7, 2005. The location of this site and more detailed information on it are available in the state site survey files at the Tennessee Division of Archaeology in Nashville, Tennessee. The ensuing description of 40SU224 is intentionally short on certain key details to avoid divulging its precise location to potential looters and the general public.
Site 40SU224 is located in Sumner County, Tennessee, on the second terrace (T-2) of a stream that has fairly good flow characteristics during the wet months of the year, but often goes almost bone dry in the late summer and early autumn dry season. No professional excavations have been conducted at this site. Because of various settlement pattern indicators and their distribution on the landscape, the type and size of this site are open to some debate. At the bare minimum, it was probably a Mississippian agricultural hamlet or perhaps a somewhat larger dispersed nucleated settlement. It might have been occupied during the Thruston phase (AD 1250-1450), as defined by Moore and Smith (2001:222), but this is speculative and has not been definitively determined from the results of excavations and radiocarbon dating. The natural spring that supported this settlement is on the floodplain (T-1) to the south of 40SU224 and is relatively close to the stream. Its proximity to the main occupation area of the site would have made it a convenient source of potable water for domestic activities.
The area immediately around the spring was subjected to landscape alterations at some point in the last 45 years. When the author first visited this spring in the 1960s, the water flowing from it during the summer months was cold to the touch and crystal clear. Its taste was quite similar to that of modern bottled water. Today the spring water bubbles out of the ground vertically and quite rapidly.
In the early 20th century, the U. S. Geological Survey (USGS) conducted a survey of wells and springs in Tennessee, and this spring was included in the survey. According to the survey information collected in 1927, the spring was located at 570 feet above mean sea level—an elevation that disagrees with a more recent topographic assessment. Its exact source was noted as being “concealed.” However, the spring was thought to originate in the Bigby limestone, which is the lower member of the Bigby-Cannon limestone that underlies the Catheys Formation.
On October 18, 1927, the yield of the spring was tested at 15 gallons per minute (gpm), but the rate of flow was noted as being seasonally variable. At a yield rate of 15 gpm, the spring would have produced 21,600 gallons of fresh water per day, and this rate was based on a yield test done at the height of the dry season in Middle Tennessee. Yields may have been greater during the spring wet season. The minimum annual yield of the spring in 1927 was approximately 7,884,000 gallons of fresh water.
These data indicate that the spring would have produced a very large amount of potable water for the Mississippian occupants of 40SU224, but this water may have been unavailable for brief periods of time. Considering the close proximity of the spring to the banks of the nearby stream, it is quite possible that human access to the spring may have been closed off by flooding during portions of the wet season and on certain summer days when intense thunderstorms caused flash flooding.
The spring at 40SU224 was also used during the Historic period. The first pioneer settler to occupy and intensively use the site was Peter Fisher, who apparently arrived with the second wave of settlers after 1795. The Fisher family undoubtedly used this spring as their primary source of potable water. Although unclear at this time, a map of Sumner County suggests that another family may have occupied the site and used the spring until at least 1878.
This spring was discussed with a local informant who was familiar with history and folklore in this part of Sumner County. According to folklore, this spring served as a reliable source of potable water for the surrounding area during times of drought in the 19th century. When other springs, wells, or creeks ran dry across the general area, this spring still produced water. As a result, local farmers would load empty wooden barrels onto their horse-drawn wagons, go to the spring for water, and haul it back to their farms for various uses, including watering livestock.
By 1927, the USGS records indicate that the spring at 40SU224 was used primarily to water livestock. Historically, this makes sense because a new home was constructed on the site in the late 1800s or very early 1900s. This home is known from historic era photographs. It was a large, one-story, balloon-frame home constructed in the National folk style, which is described and illustrated in McAlester and McAlester (1984:89-101). Either at the time this new home was built or shortly thereafter, the owner ceased using the spring for domestic potable water and installed a large, cylindrical, subsurface cistern immediately behind the house. This cistern was still present on the site as late as 2007.
In summary, the spring at 40SU224 produced a very large volume of potable water that was more than sufficient to meet the daily needs of the Mississippian population occupying the site. Based on a known rate of yield (15 gpm) determined during the dry season and historic-era folklore about widespread use of it during times of drought, this spring would have been a constant and reliable source of clean water throughout most of the year in ancient times. The only impingement on use during Mississippian times may have been wet season and occasional summer flooding of a nearby stream, which would have engulfed the spring and closed off human access to it for brief periods of time.
McAlester, Virginia and Lee McAlester 1984. A Field Guide to American Houses. 2003 reprint. New York: Alfred A. Knopf.
Moore, Michael C. 1998. “Environmental Setting” (Section I). In Gordontown: Salvage Archaeology at a Mississippian Town in Davidson County, Tennessee. Edited by Michael C. Moore and Emanuel Breitburg, pp. 1-5. Research Series No. 11. Tennessee Department of Environment and Conservation, Division of Archaeology, Nashville, Tennessee.
Moore, Michael C. and Kevin E. Smith 2001. “Summary Remarks” (Section XIII). In Archaeological Excavations at the Rutherford-Kizer Site: A Mississippian Mound Center in Sumner County, Tennessee. Edited by Michael C. Moore and Kevin E Smith, 215-248. Research Series No. 13. Tennessee Department of Environment and Conservation, Division of Archaeology, Nashville, Tennessee.