Reexamining the Evidence for Bear Ceremonialism in the Lower Mississippi Valley

This chapter reexamines the archaeological evidence for black bear (Ursus americanus) ceremonialism in the Lower Mississippi Valley (Arkansas, Louisiana, and Mississippi). While the environment of the Lower Mississippi Valley was certainly favorable for black bear, we find that not only do ethnohistorical records indicate the importance of black bear but there is a particularly high amount of bear remains at sites dating to the Late Woodland period (AD 700–1200). Feltus, a Coles Creek mound center in southwest Mississippi, is then presented as a case study of black bear ceremonialism. Extensive excavations show that black bear remained a consistent focus throughout the site’s use. Although the specific nature of bear use differs when comparing off-mound and pre-mound contexts to summit-related activities, black bear is consistently found in association with feasting events and ritual activities related to the setting of freestanding posts.

The Cahokia Connection at the Lake Providence Mounds, Louisiana

Several hundred potentially exotic sherds were recovered from late Coles Creek period (ca. A.D. 1150 to 1260) contexts at the Lake Providence Mounds in northeast Louisiana. A number of these sherds appeared to represent the remains of vessels from Cahokia or sites in the American Bottom region of southern Illinois, while others likely were local copies of Cahokia vessels. A selected sample of these “exotic” sherds, along with recognized local specimens, were subjected to petrographic thin sectioning (petrographic analysis) and instrumental neutron activation analysis (INAA). These analyses suggested that some of the potentially exotic sherds were, indeed, from the American Bottom area, while others could not be sources to a specific region. Possible reasons for the exotic vessels at Lake Providence are discussed.

Another Look at the Snipes Site (41CS8) on the Sulphur River, Cass County, Texas

The Snipes site (41CS8) was excavated by Jelks in 1952 as part of the River Basins Surveys (RBS) program administered by the Smithsonian Institution in cooperation with the National Park Service, the U.S. Army Corps of Engineers, and the Bureau of Reclamation. Snipes was one of three sites excavated by the RBS prior to the inundation of a large part of the lower Sulphur River valley by Texarkana Reservoir, now Lake Wright Patman. The Snipes site was apparently occupied during at least some part of the Woodland period (ca. 500 B.C.- A.D. 800), mainly during the latter part of the period, and can be considered a component of the Fourche Maline Culture on the basis of the artifacts recovered from both habitation archaeological deposits and burial features. Other artifacts in the collection attest to the use of the Snipes site during Late Archaic and post-A.D. 900 ancestral Caddo times, as we will discuss below. The site was estimated to cover ca. 6-7 acres of an upland landform about 1.6 km south of the Sulphur River, and was marked by a scatter of lithic artifacts, mussel shell, animal bones, charcoal, pottery sherds, etc. Excavations were done in a series of 5-ft. squares dispersed across a 300 x 200 ft. area; according to Jelks, “additional squares were opened adjacent to the most productive test squares.” The archaeological deposits ranged from ca. 8-38 cm in thickness from the surface, and had been well disturbed by plowing. Apparent midden deposits marked by “a great deal of carbon and grease” were identified in several parts of the excavations. During the work at the site by Jelks, nine prehistoric burials were excavated there, including three (Burials 7-9) that were discovered an unknown distance northeast of Burials 1-6 during the last round of work at the site. Jelks reported that traces of human remains from adults of unknown sex were present in Burials 6 and 7, but failed to mention if such was the case for Burials 8 and 9. However, since Burial 8 was reported to have contained two individuals, and the orientation of the heads was recorded, human remains (again, probably from adults, although not noted) were obviously preserved in this burial as well, but apparently not recovered. For Burial 9, Jelks simply noted that preservation “was poor.” Burials 6, 7, and 8 contained individuals that were placed in graves in an extended supine (i.e., on their back) position. Burial 8 had two individuals in extended supine position. The position of the deceased individual in Burial 9 was not recorded. Burial 6 had a Coles Creek Incised, var. Stoner bowl by the right shoulder of the deceased, and one small bowl each had been placed as a funerary offering in Burials 7-9; in two instances, the small bowls were by the left shoulder of the deceased. Two other vessels were funerary offerings in Burial 1 and a separate burial feature excavated by I. B. (Bogey) Price after the main RBS investigations.

Comparing Caddo and Coles Creek Pottery Using Petrographic Analysis

Pottery classified as “Coles Creek Incised” is common both to the earliest Caddo sites along the Red River and to contemporary sites in the Lower Mississippi Valley. Although it often is suggested that Coles Creek pottery from the two regions can be distinguished by differences in paste, no detailed comparative studies have been carried out. An initial attempt to identify variation through the use of petrographic analysis was carried out by comparing 50 samples drawn from sites in northwest and central Louisiana. Although no sharp dichotomy was noted between the regions, the study identified distinctions that support the notion that most Coles Creek pottery was made locally and different technological traditions may be represented.

The Harling Site (41FN1): An Ancestral Caddo Mound Site on the Red River in Fannin County, Texas

The Harling site (41FN1), also earlier known as the Morgan Place, is a little-known ancestral Caddo mound site located on the first alluvial terrace of the Red River in the northeastern corner of Fannin County in East Texas. The only professional archaeological investigations at the Harling site took place in November-December 1960 by a University of Texas crew led by Dr. E. Mott Davis, in advance of proposed mound leveling by the landowner. Other than short summary articles by Davis, the results of the excavations and analyses of the recovered artifacts from the Harling site have not been previously published. The mound at the site was leveled in 1963 by the landowner, Mr. R. A. Harling. The single mound at the site was approximately 70 x 52 x 2.1 m in length, width, and height. There was a borrow pit area at the southern end of the mound. The mound at the Harling site appears to be the westernmost known of the more than 100 Caddo mounds that have been reported in East Texas. According to Davis (1996:463), the site is on the western frontier of Caddo communities in the Red River valley, and Caddo settlements are found at most only a few miles to the west of the site along the river, but are common to the east of the Harling mound. Based on the 1960 excavations of the mound and an examination at that time of the surrounding alluvial landforms—which were plowed—there was no substantial Caddo settlement at the Harling site, or any associated settlement cluster within ca. 2.5 km of the mound, although there were scattered artifacts from the surface dispersed both east and west of the mound. When R. L. Stephenson, E. O. Miller, and Lester Wilson visited the Harling site in August 1950, however, they commented that artifacts were abundant in the plowed fields around the mound. In particular they noted that the ceramic sherds were mostly plain and grog-tempered, and some of the sherds had a red slip (i.e., Sanders Plain). R. King Harris also collected artifacts from the site, primarily from an area to the west of the mound and near the edge of the alluvial terrace . He collected from this area W Gary dart points, Alba arrow points, plain sherds, and one Coles Creek Incised rim with an incised lip line. In the fields east of the mound, Harris collected a number of small triangular arrow points, suggesting that this area was where the latest Caddo occupation had taken place.

Notes on the Hudnall-Pirtle Site (41RK4) in the Buddy Calvin Jones Collection at the Gregg County Historical Museum

The Hudnall-Pirtle site (41RK4) is an important Early Caddo (ca. A.D. 900-1200) period multiple mound center and large village situated on an alluvial terrace of the Sabine River in East Texas. Although best known through the archaeological investigations conducted by the Texas Historical Commission (on behalf of the Archaeological Conservancy) at the site in 1989 and 1990, Buddy Calvin Jones, then of Longview, completed his own investigations at the site in the 1950s and early 1960s, although he never published any of the archaeological findings from his work, and is has not been clear from the available records and anecdotal information where he conducted his excavations. At the 1963 Caddo Conference, Jones described the Hudnall-Pirtle site (which he called the Bivins Farm site) as follows: there is one big mound site (Bivins farm) located in the Sabine bottomlands of northeast Rusk County. It is composed of five mounds arranged around a plaza; they consist of two large rectangular temple mounds and three large circular mounds. I would guess this site to be primarily of Alto origin although tests in the village area revealed Alto and Coles Creek sherds. Since the Hudnall-Pirtle site is now known to actually have eight mounds, it would be useful to present-day Caddo archaeologists to know which mounds Jones was referring to in his description, and how their layout and placement compared to current maps of the site. It would also be helpful if information was available on where Buddy Jones conducted excavations at Hudnall-Pirtle, and what he may have found in that work. Until just a few months ago, all that was known or conjectured about the Buddy Calvin Jones excavations at the Hudnall-Pirtle site consisted of information gleaned in conversations with his mother, who mentioned trench excavations in Mound C, thought to be a burial mound, and other trenches excavated in village areas and at least one mound in 1958 and 1959. Fortunately, however, Buddy Calvin Jones maps and profiles from his work at the Hudnall-Pirtle site have been recently donated to the Gregg County Historical Museum (GCHM), although it is not known if all the notes, maps, and profiles compiled by Jones are now in the GCHM collections. This article discusses the Buddy Calvin Jones notes on his 1956-1961 work at the Hudnall-Pirtle site.