The History of Archaeological Investigations at The Jamestown Mound Site (41SM54), An Archaeological Conservancy Preserve in Smith County, Texas

Repository Citation Perttula, Timothy K. (2007) "The History of Archaeological Investigations at The Jamestown Mound Site (41SM54), An Archaeological Conservancy Preserve in Smith County, Texas," Index of Texas Archaeology: Open Access Gray Literature from the Lone Star State: Vol. 2007 , Article 15. https://doi.org/ 10.21112/.ita.2007.1.15 ISSN: 2475-9333 Available at: https://scholarworks.sfasu.edu/ita/vol2007/iss1/15


INTRODUCTION
The Jamestown Mound site (41SM54) is an Archaeological Conservancy (TAC) preserve in northern Smith Country, Texas in the northeastern part of the state. The Jamestown site is one of the largest Caddo mound centers in East Texas, with seven recorded mounds and an associated village area of unknown extent and internal complexity (Perttula 1989(Perttula , 1994. It is also one of the four premier mound centers in the Sabine River basin, the other three being Hudnall-Pirtle (41RK4, see Bruseth and Perttula 2006), a TAC preserve, Pine Tree Mounds (41HS15) (Gadus and Fields 2005), also a TAC preserve as of 2006, and Boxed Springs (41UR30, see Perttula et al. 2000), and was obviously an important civic and ceremonial center for the prehistoric Caddo peoples that lived there and in surrounding communities. Unfortunately, at the present time very little is known about the archaeological record preserved at the Jamestown site, or the exact locations of several of the smaller mounds on the preserve. Here, I summarize the history of archaeological research at the Jamestown site. This article is intended to be a companion piece to the report to be submitted to the TAC on the results of on-going remote sensing activities at the Jamestown preserve.
The Jamestown preserve covers approximately 18 acres of pasture ( Figure 1) divided into two tracts by a northsouth running fence (Figure 1, with the dividing fence removed). It is a large prehistoric Caddo mound center, with multiple mounds (Figure 2), roughly arranged in a circular pattern, with an open area (or plaza) between the mounds. The largest mound (Md. A) (Figure 3a-b), measuring ca. 43 m in diameter and 4 m in height, is situated in the southwestern side of the circle of mounds (see Figures 1 and 2). Mounds B-E (15-20 m in diameter and 40 cm-1 m in height) are probably mounds built over houses with a clay fl oor and a clay cap. The exact locations of Mounds D and E within the TAC preserve are not currently known. Md. A is known to have two levels of burned structural remains in the upper mound fi ll.

Previous Research
There has been very little archaeological attention given to the Jamestown Mound site before or after its establishment as an archaeological preserve by The Archaeological Conservancy.
The previous investigations that we are aware of that took place prior to our on-going 2005-2007 remote sensing effort include work by Sam Whiteside in the 1950s, who actually identifi ed and recorded the site a few years earlier. Sam Whiteside was a dedicated avocational archaeologist who lived east of Tyler in Smith County, of burned dirt associated with the pit, which led Whiteside to conclude that the pit represented a "fi replace." The large pit feature (see Figure 4) contained numerous artifacts in its pit fi ll. These included a plain body sherd from 0-15 cm bs, along with a possible bottle sherd that has been red-slipped and also has 2+ rows of fi ngernail punctates near the neck; this may be an example of a Maxey Noded Redware bottle sherd, examples of which are scattered on Middle Caddo period sites in the Red, Sulphur, and Sabine river basins in Northeast Texas. Deeper in the pit was a plain sherd from a carinated bowl along with a horizontal engraved bowl rim sherd. From 45-60 cm bs, Whiteside found a plain, polished body sherd, and from 60-75 cm, he recovered a plain bone-tempered rim. A large number of artifacts came from 15-30 cm in his excavations, among them a graywacke celt resharpening fl ake, two pieces of gray chert lithic debris, and eight sherds. One of these is plain, and one is a red-slipped bottle sherd. Other decorated sherds in the pit feature include an opposed incised utility   ware, two with tool punctated rows, one body sherd with a large single appliqued node, another with a large single node placed adjacent to an appliqued ridge, and an everted rim from a jar with tool punctated rows near the top of the rim, and a single appliqued node below the punctated row.
The second excavations by Whiteside were in Md. C, about 100 ft. southeast of Md. A (see Figures 1 and 2). This unit was about 2.5 x 4.6 ft. in dimension, and was dug to 3 ft in depth. In it, Whiteside uncovered a top 30 cm thick fi ll of sterile clay overlying a thin occupational lens, along with a single post hole that originated from the buried occupational lens ( Figure 5). The post hole was fi lled with clay, suggesting it had been pulled at the time the clay fi ll was used to cap the mound, thus fi lling the hole.
In the Md. C excavations, Whiteside recovered one incised grog-tempered sherd from under the clay cap (see Figure 5). The sherd has intersecting incised lines on the vessel sherd body.
The third Whiteside excavation was a 7 x 2 foot unit placed 45 feet northeast of Md. B. This showed in profi le a thin clay cap overlying a very dark occupational deposit that contained ceramic sherds ( Figure 6). This occupational deposit rested on the natural or original soil on the landform. Several post holes from a prehistoric Caddo structure were recognized and defi ned in the excavations, and they apparently originated from the occupational deposit in this area.
In Md. A, Whiteside excavated a unit (no data available on its total size) to expose a profi le of the fi ll in the upper mound. This work identifi ed two buried occupational zones, marked by concentrations of charcoal and ash at 27 cm bs, and also at 82-92 cm bs (Figure 7). These two zones were separated by a clayey mound fi ll, and there was clayey mound fi ll above the uppermost occupational zone. The concentrations of charcoal that marked both zones suggest that they represent the remnant of two different Caddo structures that stood on the mound, before they were burned and covered up by mound fi ll.
In another excavation 45 ft. north of Md. A, a 7 x 2 ft. unit, Whiteside exposed four post holes that may mark the corner of another prehistoric Caddo structure (Figure 8).
Most of the artifacts in the Sam Whiteside collection from the Jamestown site are from surface collections from prehistoric Caddo habitation deposits with an unknown provenience within the site. Other artifacts found in the surface collections are a number of decorated sherds, as well as plain body sherds (n=11), and plain grog-tempered rim sherds (n=7). There is one spindle whorl on a base sherd. A single piece of daub is in the Whiteside surface collection. Among the decorated sherds, red-slipped rims (n=2) and body sherds (n=26) are quite common, followed by tool punctated (n=7), opposed incised (n=3), cross-hatched incised (n=3), diagonal incised (n=3), and horizontal incised (n=2) decorative elements on body sherds. Other decorated utility ware sherds include fi ngernail punctated (n=2), zoned incised-punctated (n=6, including one The decorated fi ne wares are all engraved (n=10), and 10% of those have been red-slipped on exterior and interior vessel surfaces, indicating they are from bowls and carinated bowls. The engraving consists of rather simple decorative elements, among them sets of horizontal lines, sets of parallel engraved lines, diagonal lines, and opposed engraved lines. One body sherd has an equal cross-arm design, similar to that seen on a sherd from the Middle Caddo period Lake Clear site (41SM243) (Walters 2006: Figure 5).

Later Archaeological Investigations at Jamestown
In 1959, archaeologists from The University of Texas visited the Jamestown site (because of the information they obtained about it from Sam Whiteside), and they reported that the main mound (Md. A) was in good condition with the exception of one large hole near the crest of the mound (probably Whiteside's excavations). No artifacts were collected from the mound at the time, but during their reconnaissance of the site, a large amount of artifacts were collected from the plowed surface of the fi eld north and east of Md. A (see Figures 1 and 2), especially 100-200 m northeast of the mound in the areas suspected to contain possible midden deposits (see Figure 2). Their collection includes 204 sherds, six pieces of burned clay, two Alba arrow points, a portion of an abrader, a fi re-cracked rock, two pitted stones, and six pieces of lithic debris. Approximately 23% of the sherds were decorated, including incised, engraved, and red-slipped decorated sherds.
In the early 1970s, Mr. Robert Turbeville, an avocational archeologist living in Mineola, Texas, purchased a house lot that was situated at the south end of the Jamestown site, about 50 m south of Mound A. He apparently conducted limited excavations/explorations in a garden area north of his house (see Figure 2), but outside the TAC preserve, and recovered many ceramic sherds from this work (Perttula 1989:68). This collection has not been studied at this time. In the late 1970s, Dr. James Bruseth (then at Southern Methodist University) and Bob Skiles visited the site.
They excavated a small unit on Md. A at the site adjacent to a large pothole on the crest of the mound (this hole is probably the 1950s Whiteside excavations on the mound). They noted evidence of different fi ll zones, but only a few ceramic sherds were recovered or noted.
Finally, in the 1980s, archaeologists from the University of North Texas visited the site as part of an examination of Caddo mounds in the Sabine River basin. They excavated a number of shovel tests (Perttula 1989: Figure  24) on the eastern side of the site (the eventual TAC preserve) The 16 shovel tests were placed in fi ve of the mounds (Md. A-E) and then in areas around the mounds. Most of them contained prehistoric archaeological material. Mound fi lls were recognized in different shovel tests by a charcoal-streaked sandy loam A-horizon visible in shovel test profi les. Midden deposit was identifi ed in one shovel test by a dark brown to black sandy loam deposit with charcoal and ceramics between ca. 28-40 cm bs. The midden area-in the far northeastern part of the preserve (see Figure 2)-was demarcated as a slight rise (see Figure 1), but it seemingly represents an accumulation of refuse rather than a product of a deliberately constructed mound. Outside of the mound and the midden, the archaeological deposits in the shovel tests were thin, and they were primarily restricted to the plow zone.
Artifacts found in the shovel testing were concentrated in Md. B and the midden. The density is 3.5 artifacts per shovel test, which is about 35-40 artifacts per cubic meter in the archaeological deposits. Including artifacts from the surface, a total of 64 items were found, namely six fi re-cracked rocks, 10 pieces of lithic debris, a fl ake tool, fi ve burned clay pieces, and 42 ceramic sherds.
The sherds were tempered with grog and grog-bone, and some of the sherds had crushed pieces of hematite in the paste. The decorated sherds were few (n=6), and the small assemblage has a plain-decorated sherd ratio of 6.00 (i.e., six plain sherds to every one decorated sherd). Three of the sherds had a red-slip, one had a row of fi ngernail punctates, and two were cross-hatched incised body sherds.
During a fi rst effort at remote sensing at Jamestown in 2005 (to be reported separately), we had the opportunity to observe artifacts visible on the surface of the site across the TAC preserve. There were four areas where we noted prehistoric Caddo artifacts on the surface. The fi rst was at the entrance to the property (between Mds. A and B along FM 1253), where a diagonal engraved body sherd was noted. Two thin-walled body sherds were also noted in a gopher mound at the far eastern end of the preserve, and two more plain body sherds were noted northeast of Md. A. In a drainage/eroded area at the north-central boundary of the preserve (see Figure 1), a large number of prehistoric Caddo artifacts were noted. These included seven pieces of chert lithic debris, two chert cores (one being a single-platform core, and the another is a bipolar core), and 20 Caddo sherds. Sixteen of the sherds were plain body sherds, but four were body sherds that were decorated: one with parallel engraving; two with an exterior red slip; and one with curvilinear incised lines.

SUMMARY
Although there has been very little archaeological work done at the Jamestown Mound site (41SM54) since it was fi rst recorded in the 1950s, there is some information available from disparate sources (especially the excavations and artifact collections by Sam Whiteside) that provides a glimpse of the nature of the prehistoric Caddo occupation at this mound center. The mounds either capped important structures, or had structures that stood on them (Md. A) that were burned and capped with clay, probably as part of community-level rituals and ceremonies. There are Caddo habitation deposits in several areas at Jamestown, marked by midden deposits and features, including large pits and post holes from structures.
It will be important in coming years that archaeological work at the Jamestown Mound site focus on relocating all the mounds on the TAC preserve. The exact locations of Mds. D and E are not known with any precision, and there are no obvious surface manifestations of these two mounds. Also key for future TAC management and research needs at Jamestown will be to identify and defi ne the distribution of prehistoric habitation deposits, features, and middens at the site as these constitute the unstudied but important settlements of the Caddo people that lived and used the mound center. The available artifacts recovered from the site do suggest that its principal Caddo use was ca. A.D. 1200-1400, when it was one of the premier mound centers in the Sabine River basin in Northeast Texas.
The remote sensing effort-if successful-will be a good step in this direction, particularly if it can identify midden and structure locations across the site. A program of systematic shovel tests across the TAC preserve could-at very little expense and minimal impact to the archaeological deposits-serve the same purpose and help obtain information on the intra-site character of Caddo habitation deposits. This work would also have the added benefi t of recovering samples of artifacts from a controlled context to better understand the settlement of the Jamestown site. Furthermore, under the right circumstances, the fi nds from controlled shovel tests (artifacts and material remains) could be used to help establish the absolute age (from radiocarbon and thermoluminescence dating) of the prehistoric Caddo occupation at the Jamestown site.