Seed Bead Color Patterns from Colonial Period Sites in Texas and Louisiana

This article focuses on the seed beads recovered from the Spradley site (41 NA206), a possible Nacogdoche village site located south of Nacogdoches, Texas, and compares the seed bead color pattern to that of other colonial period sites in the region, including Deshazo (41 NA27), Stephens (41 NA202), Pearson (41RA5), Gilbert (41RA13), Roseborough Lake (41BW5), Vinson (41LTl), Womack (41LR1), 41H064, Atlanta State Park (41CS37), Ware Acres (4IGG31), and the shipwreck of La Belle in Texas; and Los Adaes (16NA16) and Colfax Ferry (16NA 15) in Louisiana. The possible meaning of different seed bead color patterns is briefly discussed.


INTRODUCTION
Sorting and inventorying 1116-inch screened material from the Spradley site (41NA206) resulted in the identification of a number of seed beads. Seed beads are glass beads less than 4 mm in diameter which were sewn onto clothing and other personal items. The traditional approach lo seed head analysis is to describe the beads according to recognized bead classification systems, most notably those of R. King Harris and lnus Marie Harris (1967), Kenneth E. Kidd and Martha Ann Kidd ( 1970), and Jeffrey P. Brain ( 1979). The results of most bead classification reports are used primarily for chronology and bead type distributions. Few studies have focused on seed bead color patterns-one notable exception is J. Gottfrcd's (1997) work on seed beads from several late 18th century sites in the northwestern United States. This goal of this article is simply to better understand the seed bead color paLLem of the Spradley site beads by oomparing it to other colonial period sites in Texas and Louisiana. The possible social ramifications of seed bead color patterns are briefly discussed.

THE SPRADLEY SITE (41NA206)
The Spradley site (41NA206) is the site of a probable historic period Nacogdoche Indian village habitation roughly three miles south of Nacogdoches. The site was recorded by Tom Middlebrook in 1998, and it is very important as it represents one of the more systematically tested historic period Caddo sites in the area (see Middlebrook 2007). (SFA) Introduction to Archaeology class. SFA student workers finished the sorting and inventory by working in the summer and fall of2007. All material from the three SFA field seasons at the Spradley site is curated at the Dr. Jim Corbin Archaeology Lab at SF A. To date, there have been no published reports of the archaeological investigations of the Spradley site, although several presentations have been made at professional meetings (Galan et al. 2004;Bibby 2006;Galan 2006).
A total of 65 seed beads have been recovered from the Spradley site. The colors include white, translucent, various shades of blue, green, and red ( Figure 1 ). By the end of spring 2007 it was clear that the seed head sample from Spradley was different from the seed bead collection from Los Adaes (16NA 16 ), a sample I was most familiar with. Black is the predominant color for seed beads in the Los Adacs collcl:lion (  Figure 3 shows the location of colonial period sites in Texas and Louisiana that have over 100 seed beads in their archaeological assemblages. Table 1 lists these sites, along with the references and bead color proportions. Most of the bead contexts arc burials; the beads from Spradley, Womack, Gilbert, Vinson, and Los Adacs are not from burials. Bead color in this study is Jumped-that is, all the various shades of blue are lumped together as simply "blue." There is no distinction made for compound beads in this study. Even though some beads with an overall white appeamnce actually have a clear exterior layer (e.g., Figure I a), these beads are lumped with white beads. Also, beads with a red exterior and yellowish core (e.g., Figure lg) arc described as red. The simplilled color categories include white, blue, black, red, clear, green, and amber. A pie chart generated in Excel and color-corrected in Adobe Photoshop was made for each site (Figure 4 ).
AL first glance, several patterns are noticeable. Only Los Adaes and Colfax Ferry have predominantly black seed beads. White seed beads occur in almost the same proportions al Spradley and Womack, but Spradley has more red, clear, and green beads. Roseborough Lake and Pearson are remarkably similar-Gilbert and Vinson are fairly similar, differing primarily in the proportions of red and black beads. Atlanta Stale Park is distinct by its almost total dominance of blue, while Ware Acres is distinct for ils total absence of blue. Stephens has substantially higher proportions of green, while Los Adaes has the highest proportion of clear beads.
Jeff Girard was kind enough both to suggest that correspondence analysis would help interpret the variation and also to perform Lhe analysis ( Figure  5). The correspondence plot shows that Deshazo, Stephens, Spradley, 41 H064, Roseborough Lake, and Pearson are fairly closely related. Surprisingly  The correspondence plot (see Figure 5) also shows the seed bead color pattern from the wreck of La Belle. Figure 6 shows the seed bead color pattern pie chart for Lo. Belle. The

DISCUSSION
So what might this all mean? Dillerences in head color preference among American Indian groups in North Ameril.:a have been noted, and for some groups-for example, the Plains Indiansthese differences are attributed to ethnic differences (see Stine et al. 1996:57). It is likely that at least one, if not more, of the sites in this study are associated with different tribal groups. Colfax Ferry is a site occupied by Pascagoula and Diloxi Indians, and as both the pie chans and correspondence analysis indicate, Colfax Ferry is very different from most of the other sites in this study. Colfax Ferry is most closely related to the seed bead color pattern for Los Adaes (see Figures 4-5). Womack, Gilbert, Pearson, and Vinson are described as "Nortefio" sites in their respective reports, allhough Tim Perttula (personal communication. 2007) suggests all but Vinson are Caddo sites. The remaining sites-Atlanta State Park, Roseborough Lake, Ware Acres, 41 H064. Deshazo, Stephens, and Spradley---<:an all be described as Caddo sites. Ware Acres is considered a Caddo

4.-----~~------~~------~~----~
sites, and the less remarkable, but still similar seed bead color patterns for the Gilbert and Vinson sites. It is possible that these four sites represent two groups being in two different places at different times, instead of four groups being in four different places. That is, given the earlier dates for Roseborough Lake and Gilbert, and later dates for Pearson and Vinson, it is possible that one group left the Gilbert site area prior to 1770 and went to the Vinson site area, and the other group left the Roseborough Lake area prior to 1780 and went to the Pearson site area. However, unpublished observations of gun parts and a religious medallion from Pearson suggest an earlier occupation for the Pearson site (Jay C. Blaine, personal communication 2008). Still, the potential in the future to use seed bead color patterns to track possible movement of individuals or groups through time is an intriguing possibility.  site, but the bead color pattern is very different from any other site in the study. Clearly, any step in associating bead color pattern with sociaVpolitical groupings, be they family groups or tribes, should be taken with caution. But, throwing caution to the wind for the moment, it is tempting to hazard an interpretation of the remarkable similarities between the seed head color patterns for the Roseborough Lake and Pearson

CONCLUSIONS
The intention of this article is in no way to diminish the importance of meticulous description of glass seed beads. This will always be a critical part of head analysis. The goal here was simply to investigate the variation in seed bead color patterns in colonial period sites. This is not a revolutionary idea. Bob Turner, the premier bead analyst in East Texas, mentioned to me that he had thought about it years ago. Seed beads were sewn onto clothing and other personal articles in patterns with varying colors being used. Sint:e different seed bead color patterns have been associated with different social/ political groupings, it is not too much of a leap to suggest that seed bead color patterns from art:haeological sites might have potential for providing so-ciaVpolitical information about the people who wore the beads. Future research might include counting the various colors of seed beads on historic Caddo examples of head work, determining the seed bead color patterns, and comparing this to the seed bead color pattems found on colonial period sites in Texas and Louisiana.