scholarly journals Analysis of a 1940 Caddo Sherd Assemblage from the Millsey Williamson Site (41RK3), Rusk County, Texas

Author(s):  
Timothy K. Perttula

The Millsey Williamson site (41RK3) is an 18th century Nadaco Caddo settlement and cemetery situated on an alluvial terrace on the east side of Martin Creek in the Sabine River basin. Some portions of the site are now covered by the waters of Martin Creek Lake, constructed in the 1970s. The site was first investigated in the 1930s, when at least 11 historic Caddo burials were excavated in the cemetery at the western end of the landform. In 1940, Jack Hughes, then an East Texas resident, but later a prominent Texas archaeologist, gathered a small collection of sherds from the Millsey Williamson site, and the analysis of these sherds is the subject of this article.

Author(s):  
Timothy K. Perttula

The Millsey Williamson site (41RK3) is an 18th century Nadaco Caddo settlement and cemetery situated on an alluvial terrace on the east side of Martin Creek in the Sabine River basin. Some portions of the site are now covered by the waters of Martin Creek Lake, constructed in the 1970s. The site was first investigated in the 1930s, when at least 11 historic Caddo burials were excavated in the cemetery at the western end of the landform. Buddy Calvin Jones excavated a disturbed historic burial at the site in 1955, and also occasionally collected glass beads from the surface of the site. The funerary offerings placed with this disturbed burial were not clearly enumerated by Jones, as his description of artifacts from the site included artifacts he examined in several other collections. He did note 275 sherds from the surface of the site and 12 whole or restored ceramic vessels from an unknown number of burials. Most of these sherds were recorded by Jones as being grog– (52 percent) or bone–tempered (43 percent), but 4 percent were tempered with shell. Perttula and Nelson recently documented 11 vessels from the Millsey Williamson site in the collections of the Gregg County Historical Museum (GCHM). These vessels include a Emory Punctated–Incised (shell– tempered) collared jar; a Maydelle Incised jar; a Bullard Brushed jar; a jar with brushing only on the body; a Ripley Engraved, var. unspecified carinated bowl; Simms Incised carinated bowl; two unidentified engraved carinated bowls with a continuous stepped rectilinear scroll design; a carinated bowl with a sprocket rim with a continuous negative scroll design; a carinated bowl with diagonal opposed and cross–hatched engraved lines on the rim; and a plain olla. The ceramic vessels are of diverse manufacture, form, and decorative methods. Most are carinated bowls and jars tempered with grog and bone, and fired in a reducing environment, and the former are decorated with engraved lines, while the latter are decorated incised, punctated, or brushed utility wares. On their own stylistic merits, none of these vessels in the GCHM collections is that of a recognizable Historic Caddo type, such as Natchitoches Engraved, Simms Engraved, var. Darco, or Keno Trailed, and in fact, most of these vessels cannot be identified as examples of specific types. The vessels that can be typed include Emory Punctated–Incised, Maydelle Incised, and Bullard Brushed jars and a Ripley Engraved, var. unspecified carinated bowl; one vessel has been dubbed Simms Incised because it is of a form and decorative style that matches Simms Engraved, except the motif is executed with incised lines. Also recovered from the site were clay and limonite pipes, ochre and vermillion, animal teeth, glass beads, metal gun parts, gun flints, iron knives, iron arrow points, and awls in the Millsey Williamson collection. There were also a variety of brass objects: a brass tinkler, coils, hawk bells, and unworked pieces of sheet brass.


Author(s):  
Timothy K. Perttula ◽  
Bo Nelson

There are collections of ceramic vessels and other artifacts from the Millsey Williamson (41RK3), Bead Burial, and L. N. Morwell sites in the Buddy Jones collection at the Gregg County Historical Museum. The purpose of this article is to put the documentation of these collections on record, as this documentation provides previously unavailable detailed information on the material content of probable 18th century Nadaco Caddo/Kinsloe phase historic sites in East Texas. Based on the limited available information from the Bead Burial and L. N. Morwell Farm sites, it is probable that all three sites are different names for the same Historic Caddo site situated along the Rusk and Panola County line in East Texas on Trammel’s Trace that was reported on by Jones. The Bead Burial site is reported to be ca. 5 miles south of Tatum along the Rusk-Panola County line, and the Millsey Williamson site is well known for the quantity of glass trade beads found there. The L. M. Morwell Farm site was excavated by C. W. Bailey in 1940, and a tag accompanying two ceramic vessels recovered from a Burial 4 at the site describe it as “Rusk Co. Martin Creek old trading post on Trammels trace.” Jones indicates that the Millsey Williamson site is an 18th century Nadaco Caddo settlement and cemetery situated on an alluvial terrace on the east side of Martin Creek, a northward-flowing tributary to the Sabine River. Some portions of the site are now covered by the waters of Martin Creek Lake, constructed in the 1970s. The site was first known in the 1930s, when at least 11 historic Caddo burials were excavated in the cemetery at the western end of the landform, and there was a habitation/village area on the highest part of the landform, east of the cemetery. Jones excavated a disturbed historic burial at the site in 1955, and also occasionally collected glass beads from the surface of the site. The funerary offerings placed with this disturbed burial are not clearly enumerated by Jones, as his description of artifacts from the site includes artifacts he examined in several other collections. He did note 275 sherds from the surface of the site and 12 whole or restored ceramic vessels from an unknown number of burials (Jones 1968:Table 1). Most of these sherds were grog- (52%) or bone-tempered (43%), but 4% were tempered with shell. There were also clay and limonite pipes, ochre and vermillion, animal teeth, glass beads, metal gun parts, gun flints, iron knives, iron arrow points and awls, and a variety of brass objects: a brass tinkler, coils, hawk bells, and unworked pieces of sheet brass.


Author(s):  
Timothy K. Perttula ◽  
Mark Walters

The Jonas Short site (41SA101) is one of a few known and investigated Woodland period mounds in the Trans–Mississippi south (i.e., East Texas, Northwest Louisiana, Southwest Arkansas, and Southeast Oklahoma). In fact, the site is one of only four identified mound sites of possible Woodland period age—and Mossy Grove cultural tradition—in the Neches–Angelina and Sabine river basins in East Texas and Northwest Louisiana: Coral Snake (16SA48), Anthony (16SA7), Jonas Short, and Westerman (41HO15). The Jonas Short site was located on an alluvial terrace of the Angelina River. It was investigated in 1956 by archaeologists from the University of Texas and the River Basin Survey prior to its inundation by the waters of Lake Sam Rayburn.


Author(s):  
Timothy Perttula

Site 41HS74 is an ancestral Caddo habitation site and cemetery on Hatley Creek, a southwardflowing tributary to the Sabine River, in the East Texas Pineywoods (Figure 1). The site was investigated in 1986 by Heartfield, Price and Greene, Inc. (1988). The re-analysis of the ceramic vessels recovered from nine burial features at the site are the subject of this article. The vessels are curated at the Texas Archeological Research Laboratory at The University of Texas at Austin (TARL).


Author(s):  
Timothy K. Perttula

Site 41AN28 is an extensive ancestral Caddo settlement on an alluvial terrace on the west side of Mound Prairie Creek in the Post Oak Savannah of East Texas. Mound Prairie Creek is an southward- and eastward-flowing tributary to the Neches River. The confluence of the two streams lies about 20 km to the east. Directly to the east of 41AN28 on the east side of Mound Prairie Creek lies the Pace McDonald site (41AN51). This site is an important Middle Caddo period (ca. A.D. 1200-1400) mound center with at least two earthen mounds and a settlement that covers more than 11 acres. Both investigated mounds at Pace McDonald were built to cover and bury special purpose structures where significant deposits of ash were accumulated, probably ash temples like those uncovered in the main mound at the nearby A. C. Saunders site (41AN19). Two calibrated radiocarbon dates from habitation areas at the Pace McDonald site range from A.D. 1200-1410.


Author(s):  
Timothy K. Perttula

The Womack site (41LR1) is an ancestral Caddo settlement situated on an alluvial terrace in a horseshoe bend of the Red River in north central Lamar County in East Texas. Harris completed the analysis and study of their 1938-mid-1960s investigations at the site, but the findings from the earlier archaeological investigations conducted at the site by the University of Texas (UT) in 1931 have not been previously published. In this article I discuss the 1931 investigations by UT at the Womack site, and also summarize the character of the artifact assemblage recovered at the site during this work. Lastly, I consider the occupational character and settlement history of the Womack site—particularly its history of settlement by ancestral Caddo peoples—and define a Womack phase for the ca. A.D. 1690-1750 archeological components at the Womack, Sanders (41LR2), Goss Farm (41FN12), and Harling (41FN1) sites along the upper Red River and the Gilbert (41RA13) and Pearson (41RA5) sites in the upper Sabine River basin.


Author(s):  
Mark Walters ◽  
Timothy K. Perttula

In February 1957, Sam Whiteside of Smith County, Texas, excavated a burial at 41SM53. This site was designated P-4 in Mr. Whiteside’s notes and it was one of several Caddo sites along Prairie Creek in the upper Sabine River basin that he investigated to varying degrees in the 1950s and 1960s. As an a vocational archeologist Mr. Whiteside made many important contributions to East Texas archeology. Dr. Dee Ann Story, of the Texas Archeological Research Laboratory at The University of Texas at Austin, who corresponded with Mr. Whiteside, later obtained the trinomial 41SM53 for the site.


Author(s):  
Timothy K. Perttula

The A. C. Gibson site (41WD1) is an ancestral Caddo site of probable Middle Caddo period (ca. A.D. 1200-1400) age in the Sabine River basin in the Post Oak Savannah of East Texas (Figure 1). The site is on a natural alluvial knoll in the floodplain of the Sabine River and Cottonwood Creek, just north of Cedar Lake, an old channel of the river. The site has been known since the early 1930s by collectors and site looters, early University of Texas (UT) archeologists, and then by later archaeologists from UT and Southern Methodist University, but it has heretofore not been scrutinized by Caddo archaeologists to any serious degree.


Author(s):  
Timothy K. Perttula

The Gilbert site (41RA13) is an important mid-18th century American Indian site on an alluvial terrace along Lake Fork Creek, adjacent to the upper part of Lake Fork Reservoir in Rains County, Texas. The site was first investigated in 1962 by the Dallas Archeological Society, and based on the findings from that work, the Texas Archeological Society (TAS) had a field school at the site in June and July 1962. There are several notable features of the Gilbert site. First, it contains 21 midden mounds about 6-9 m in diameter and ca. 1 m in height spread out over ca. 50 aces of the alluvial terrace landform. The middens do not represent habitation features, as the “only occupational features discovered besides the middens were two pits that were evidently used for storing grain or other products. No house floors, post-mold patterns, burials, hearths, or other such structural remains were found." Further investigation by Blaine identified other midden features (discussed further below, and the source of the ceramic sherds discussed in this article) and a well-preserved bell-shaped storage pit in Feature 20. The newly-discovered midden features were not mounded or had a clay cap, and neither contained any evidence of structural remains or features. Structural features are considered more likely to be found in inter-midden areas than in the middens themselves. A second notable feature of the Gilbert site is the abundance of mid-18th century European trade goods in the archeological deposits, much of it likely obtained from French traders. These goods include metal tools (knives, axes, wedges, hatchets, hoes, scrapers, awls, chisels, scissors, arrow points, and a Spanish sword hilt), gun parts, ornaments (especially glass trade beads), brass kettles, horse trappings, fl at and bottle glass, and chipped glass pieces. Third, there was a substantial aboriginal ceramic sherd assemblage from the Gilbert site. The analysis of the sherds suggested that they are from vessels “too stylistically and technologically diverse to represent only one locally-produced ceramic complex." Furthermore, “the majority, and perhaps even all, of the decorated ceramics [at the site] are derived ultimately from the Caddoan [sic], particularly Fulton Aspect [Late Caddo period], tradition. Many close parallels exist in the modes and styles of decoration, paste characteristics, and vessel forms." And lastly, the various results of the investigations suggested that the site was a village occupied by southern Wichita groups, possibly the Tawakoni, Kichai, or Yscani Indians. This conclusion is far from uniformly accepted, a point I will return to in the final section of this article. In the remainder of this article, I discuss the analysis of a small collection of previously unstudied ceramic sherds from two midden features (F-B3 and F-B4) excavated by Jay and Jerrylee Blaine from the Gilbert site. The focus of the analysis is to characterize the principal stylistic and technological characteristics of the ceramic sherd assemblages from these two middens, compare this assemblage in those aspects with the larger assemblage from numerous middens studied by Story, and then offer my own interpretation of the cultural affiliations of the Gilbert site occupants based on the ceramic sherd assemblage data.


Sign in / Sign up

Export Citation Format

Share Document