<%BANNER%>

Identifying Social Structure Through Spatial Patterning at Bayview, a Weeden Island Coastal Community in Northwest Florida

Permanent Link: http://ufdc.ufl.edu/UFE0024190/00001

Material Information

Title: Identifying Social Structure Through Spatial Patterning at Bayview, a Weeden Island Coastal Community in Northwest Florida
Physical Description: 1 online resource (127 p.)
Language: english
Creator: Ellison, Tria
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: archaeology, ceramic, community, gis, southeast, spatial, weeden, weedon, woodlands, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: While much of the archaeological literature on Weeden Island has centered on burial practices and ceramics artifacts, relatively little is known about the social organization of their communities. Many Weeden Island communities are circular or arcuate in shape, and recent literature on the topic has shown that there are significant differences in artifact distributions between the opposing sides of the ring or arc, indicating the potential presence of dual social organization in Weeden Island culture. This study investigates whether dual social organization is observable through spatial analysis of ceramic artifacts and faunal remains at the Bayview site (8BY137), a circular Weeden Island community located on the East Bay near Panama City, Florida. Shovel testing reveals a disproportional distribution of two ceramic types across the site, and may indicate that there is a level of dual social organization at the site. Conversely, the faunal analysis of two column samples from opposite sides of the ring shows little differentiation, indicating a lower degree of duality at the site. Although seemingly conflicting, the results from these two datasets reveal that the types of material remains that offer insight towards social organization in Weeden Island communities are highly contextual in nature, and that duality may not be visible in all forms of material culture.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Tria Ellison.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Sassaman, Kenneth E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024190:00001

Permanent Link: http://ufdc.ufl.edu/UFE0024190/00001

Material Information

Title: Identifying Social Structure Through Spatial Patterning at Bayview, a Weeden Island Coastal Community in Northwest Florida
Physical Description: 1 online resource (127 p.)
Language: english
Creator: Ellison, Tria
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: archaeology, ceramic, community, gis, southeast, spatial, weeden, weedon, woodlands, zooarchaeology
Anthropology -- Dissertations, Academic -- UF
Genre: Anthropology thesis, M.A.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: While much of the archaeological literature on Weeden Island has centered on burial practices and ceramics artifacts, relatively little is known about the social organization of their communities. Many Weeden Island communities are circular or arcuate in shape, and recent literature on the topic has shown that there are significant differences in artifact distributions between the opposing sides of the ring or arc, indicating the potential presence of dual social organization in Weeden Island culture. This study investigates whether dual social organization is observable through spatial analysis of ceramic artifacts and faunal remains at the Bayview site (8BY137), a circular Weeden Island community located on the East Bay near Panama City, Florida. Shovel testing reveals a disproportional distribution of two ceramic types across the site, and may indicate that there is a level of dual social organization at the site. Conversely, the faunal analysis of two column samples from opposite sides of the ring shows little differentiation, indicating a lower degree of duality at the site. Although seemingly conflicting, the results from these two datasets reveal that the types of material remains that offer insight towards social organization in Weeden Island communities are highly contextual in nature, and that duality may not be visible in all forms of material culture.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Tria Ellison.
Thesis: Thesis (M.A.)--University of Florida, 2009.
Local: Adviser: Sassaman, Kenneth E.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0024190:00001


This item has the following downloads:


Full Text

PAGE 1

1 IDENTIFYING SOCIAL STRUCTURE THROUGH SPATIAL PATTERNING AT BAYVIEW, A WEEDEN ISLAND COASTAL COMMUNITY IN NORTHWEST FLORIDA By TRIA MARIE ELLISON A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PA RTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Tria Marie Ellison

PAGE 3

3 To Andy, my l ove, my life and my best friend This could not have been accomplis hed without your limitless patience and support

PAGE 4

4 ACKNOWLEDGMENTS This thesis, while considered a sole authorship, is actually the product of many hardworking and dedicated people who have helped along the way. Initiall y, this project was conceived b y committee member Michael Russo, of the Southeast Archaeological Center. I t hank him for his guidance and inspiration that initially set the stage for this research. I would also like to thank my chair, Kenneth Sassaman, for accepting me as a student when my interests changed to Southeastern Archaeology and for his patience with me as I transitioned into this region of study I would also like to thank committee member Kitty Emery, who has contributed a tremendous amount of support and rigor to the faunal analysis portion of this thesis. Without her guidance, the faunal chapter might only be a few pages long. In addition to my committee, few others have offered as much support as Ann Cordell from the Florida Museum of Natural History. The dedication of her time and knowledge to this project in teaching me first -hand the nuances of Weeden Island ceramic types, was instrum ental to my ability to identify and ana lyze the ceramic assemblage. I would also like to thank my amazing team of volunteers: Andy Belcou rt, Michael Kay, Gen Dick and Katie Morrison, who donated a week of their lives to help excavate and who tolerated camping in the oppressive heat of a Florida July out at Tyndall AFB. I cant tell you how much I appreciate your hard work and attention to d etail in recovering artifacts and fieldnotes I would also like to thank Westley Westphall III, previously the Chief Environmental Officer at Tyndall AFB, for arranging site access, security and camping accommodations for the excavation of the Bayview sit e, as well as organizing the funding for this research through the National Park Service. Special thanks to the National Park Service as well, for funding the radiocarbon dating at Beta Analytic.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................................... 4 LIST OF TABLES ................................................................................................................................ 8 LIST OF FIGURES .............................................................................................................................. 9 ABSTRACT ........................................................................................................................................ 10 CHAPTERS 1 INTRODUCTION ....................................................................................................................... 11 2 THE ARCHAEOLOGY OF CIRCULAR COMMUNITIES .................................................. 17 The Concept of Community ....................................................................................................... 17 The Archaeological Community ................................................................................................ 18 Identifying Community Social Structure through Spatial Analysis......................................... 19 Circular Communities ................................................................................................................. 19 Circular Weeden Island Communities ....................................................................................... 22 Summary ...................................................................................................................................... 23 3 THE CONCEPT OF WEEDEN ISLAND ................................................................................. 24 Introduction ................................................................................................................................. 24 Geographic Extent of Weeden Island ........................................................................................ 24 Mound Ceremonialism ................................................................................................................ 25 Social Organization ..................................................................................................................... 26 Previous Research ....................................................................................................................... 28 Early Investigations at Weedon Island ............................................................................... 29 The Development of the Weeden Island Cerami c Typology ............................................ 29 Excavations at Kolomoki Mounds ...................................................................................... 30 McKeithen Weeden Island .................................................................................................. 32 Wakulla Weeden Island ....................................................................................................... 33 Coastal Weeden Island Sites ............................................................................................... 34 Old Homestead, Walton County, Florida ................................................................... 35 Plantation West, Santa Rosa County, Florida ............................................................ 35 Bird Hammock, Wakulla County, Florida .................................................................. 35 Summary ...................................................................................................................................... 36 4 FIELD METHODS AND DATA RECOVERY ....................................................................... 38 The Bayview Site (8BY137) ...................................................................................................... 38 Causes and Potential Impacts of Modern Disturbance at Bayview ......................................... 39 Establishment of Systematic Grid for Shovel Testing .............................................................. 40

PAGE 6

6 Field Excavation and Processing of Shovel Tests ..................................................................... 4 1 Excavation of Column Sample 2 ................................................................................................ 42 Summary ...................................................................................................................................... 42 5 CERAMIC TYPES AT BAYVIEW .......................................................................................... 45 Introduction ................................................................................................................................. 45 Ceramic Types found at Bayview .............................................................................................. 46 Weeden Island Plain ............................................................................................................ 46 Wakulla Check Stamped ..................................................................................................... 46 Swift Creek Complicate d Stamped ..................................................................................... 47 Carrabelle Punctated ............................................................................................................ 47 Carrabelle Incised ................................................................................................................ 47 Weeden Island Incised ......................................................................................................... 48 Weeden Island Punctated .................................................................................................... 48 Mercier Red on Buff ............................................................................................................ 49 Weeden Island Red .............................................................................................................. 49 Keith Incised ........................................................................................................................ 49 Ruskin Linear Punctated ..................................................................................................... 49 Crooked River Complicated Stamped ................................................................................ 50 Tucker Ridge Pinched ......................................................................................................... 50 Indian Pass Incised .............................................................................................................. 50 Dunlap Fabric Impressed .................................................................................................... 50 Laboratory Processing and Identification of Sherds ................................................................. 51 Results .......................................................................................................................................... 51 Plain Sherds .......................................................................................................................... 52 Decorated Sherds ................................................................................................................. 52 Early Weeden Island .................................................................................................... 53 Late Weeden Island ...................................................................................................... 53 Summary ...................................................................................................................................... 53 6 SPATIAL ANALYSIS OF CERAMIC ARTIFACTS ............................................................. 56 Introduction ................................................................................................................................. 56 Dividing Bayview into Spatially Discrete Clusters .................................................................. 56 Testing for Spatial Contemporeity ............................................................................................. 57 Radiocarbon Dating ............................................................................................................. 57 Proportions of Early and Late Weeden Island Ceramics .................................................. 58 Testing for Dual Social Organization through Spatial Distribution of Ceramic Types .......... 59 Spatial Comparison of Decorated and Plain Ceramics at Bayview ......................................... 60 Spatial Comparison of Specific Types of Decorated Ceramics at Bayview ........................... 61 Summary ...................................................................................................................................... 63 7 FAUNAL ANALYSIS ................................................................................................................ 71 Introduction ................................................................................................................................. 71 Interpreting Sociality through Faunal Remains ......................................................................... 71

PAGE 7

7 Laboratory Methods .................................................................................................................... 73 Preliminary Results of Faunal Identification and Quantification ............................................. 76 Column Sample 1, Level 4 .................................................................................................. 76 Column Sample 2 ................................................................................................................ 77 Level 2 .......................................................................................................................... 78 Level 4 .......................................................................................................................... 79 Data Analysis of Faunal Remains .............................................................................................. 79 Season of Occupation .......................................................................................................... 80 Habitat Utilization ............................................................................................................... 81 Ritual Use of Animals ......................................................................................................... 83 Relative Density ................................................................................................................... 84 Summary ...................................................................................................................................... 85 8 DISCUSSION .............................................................................................................................. 92 Introduction ................................................................................................................................. 92 Inferring Social Structure through Spatial Analysis ................................................................. 92 Contrasting Site Areas through Faunal Analysis ...................................................................... 94 9 CONCLUSION AND SUGGESTIONS FOR FUTURE RESEARCH .................................. 96 APPENDI X A SHOVEL TEST INFORMATION ............................................................................................ 98 B A RTIFACT DATABASE ........................................................................................................ 101 C CERAMIC DATA .................................................................................................................... 104 D COLUMN SAMPLE 1 LEVEL 4 ............................................................................................ 113 E COLUMN SAMPLE 2 LEVEL 2 ............................................................................................ 114 F COLUMN SAMPLE 2 LEVEL 4 ............................................................................................ 116 LIST OF REFERENCES ................................................................................................................. 120 BIOGRAPHICAL S KETCH ........................................................................................................... 127

PAGE 8

8 LIST OF TABLES Table s page 5 1 Total Ceramic Data (2007) .................................................................................................... 55 6 1 Combined Ceramic Datasets ................................................................................................. 69 6 2 Field Sample Numbers in Each Quadrant ............................................................................ 70 6 3 Proportions of Early and Late Ceramics between East and West Sid es of Site ................. 70 6 4 Proportions of Plain and Decorated Ceramics between East and West Sides of Site........ 70 6 5 Proportions of Wakull a Check Stamped (grams) ................................................................. 70 6 6 Proportions of Carabelle Types (grams) ............................................................................... 70 7 1 Summary of Faunal Material by Class .................................................................................. 88 7 2 Relative frequency of Mullet ( Mugil sp.) per Column Sample Level ................................ 88 7 3 Habitat Information on Species from Columns 1 and 2 ...................................................... 89 7 4 Number of Species Present by Habitat ................................................................................. 90 7 5 Relative Frequency of NISP, MNI and Weight of Species by Habitat ............................... 90 7 6 Relative Density of Column Sample Levels ......................................................................... 91

PAGE 9

9 LIST OF FIGURES Figure s page 1 1 Weedon Island Si tes Referred to in the Text ........................................................................ 15 1 2 Locations of Shovel Tests ...................................................................................................... 16 4 1 Location of Bayview on East Bay. ........................................................................................ 44 6 1 Field Sample Numbers used in comparative analyses ......................................................... 64 6 2 Distribution of Plain Sherds .................................................................................................. 65 6 3 Distribution of Decorated Sherds .......................................................................................... 66 6 4 Distribution of Carrabelle Sherds .......................................................................................... 67 6 5 Distribution of Wakulla Check S tamped Sherds .................................................................. 68

PAGE 10

10 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Master of Arts IDENTIFYING SOCIAL STRUCTURE THROUGH SPA TIAL PATTERNING AT BAYVIEW, A WEEDEN ISLAND COASTAL COMMUNITY IN NORTHWEST FLORIDA By Tria Marie Ellison May 2009 Chair: Kenneth E. Sassaman Major: Anthropology While much of the archaeological literature on Weeden Island has centered on burial practi ces and ceramics artifacts, relatively little is known about the social organization of their communities. M any Weeden Island communities ar e circular or arcuate in shape, and recent literature on the topic h as shown that there are significant differences in artifact distributions between the opposing sides of the ring or arc, indicating the potential presence of dual social organization in Weeden Island culture. This study investigates whether dual social organization is observable through spatial analysis of ceramic artifacts and faunal remains at the Bayview site (8BY137), a circular Weeden Island community located on the East Bay near Panama City, Florida. Shovel testing reveals a disproportional distribution of two ceramic types across the site, and may indicate that there is a level of dual social organization at the site. Conversely the faunal analysis of two column samples from opposite sides of the ring show s little differentiation, indicating a lower degree of duality at the site. Although seemingl y conflicting, the results from these two datasets reveal that the types of material remains that offer insight towards social organization in Weeden Island communities are highly contextual in nature, and that duality may not be visible in all forms of ma terial culture.

PAGE 11

11 CHAPTER 1 INTRODUCTION Ring -shaped middens are found throughout the Southeastern United States, as well as in other areas across the globe and span many different time periods. The presence of ring -shaped communities from such dive rse cultural settings leads me to ask what significance this pattern may have in understanding social complexity. Many researchers have recently been interested in ring -shaped middens and the meanings that they hold for understanding the sociality of pre Columbian cultures in North America ( Heckenberg er 2004; Means 2007; Russo 2004; Russo and Heide 2002; Ru sso et al 2006; Saunders 2004; Thompson 2006, among others). One theoretical model that has frequently been utilized in recent discussions is the dual s ociety model, originally brought forward by Lvi -Str auss (1968:133163). This model describes that many societies living in circular villages group themselves, both cognitively and spatially into two distinct, yet interlinked halves. By looking at societie s such as the Bororo, Winnebago and the Omarak ana, Lvi Strauss found that the s e divisions were intrinsically tied to the individuals perception of his role in society. For example, in the minds of the elite, the circular Winnebago village plan was diametrically divided, but to the less prestigious individuals the village was divided into two concentric circles (Lvi Strauss 1968: 134135). L ike many other social models, the dual society model has the potential to be of great use in interpreting archaeological sites, but also warrants critique of being overly broad or generalizing if used without considering the intersection of theory with the actual community it is trying to explain. In his discussion, Lvi -Strauss examined cultures that were in the prese nt, or not so distant past, and therefore approached the problem through ethnographic examples. This allowed him to see that there are potential caveats to its application, even when describing sociality of peo ple in the present.

PAGE 12

12 Lvi -Strauss realized that these complicated sets of dichotomies are sometimes related, yet at times can act as mediators for other sets of dichotomies. Multiple factors are present in cultures that act as dualities or frameworks for the organization of social patterning. However elegant his model may be, the point is not to use ideas such as these as strict templates and blindly plug in data from any given site, but rather to use the data as a means of testing ideas about social organization, in order to be able to ask more nuance d questions about the site or culture at hand. Bayview (8BY137) is a ring-shaped shell midden located directly on the East Bay near Panama City, Florida (Figure 1 1) This midden represents the domestic archaeological remains of a community, and most of the artifacts from this site are asso ciated with Weeden Island types, which date from between 2001200 CE (see Cordell 1984; Milanich 1994; Sears 1954; Willey 1949). As such, Bayview is generally considered a Weeden Island site. The material found at Bayvie w presents the opportunity to examine this model of social organization, and to offer critical data on the site to future researchers in this region as well as other areas. In order to explore and test this model using material from Bayview, I expanded on previous work at the site (Russo et al. 2006). Russos previous excavation involved collecting data from shovel tests at the site on a 20 -meter interval grid, primarily to delineate the site and to determine what the midden contained It was from this exc avation that it became clear that Bayview was a ring -shaped midden with distinct areas of artifact clustering near est to the shore of East Bay. The differences in spatial distribution of artifacts at the site inspired me to further explore whether the use of space at Bayview was ordered in such a way as to reveal evidence of repeated activity areas and social structure I doubled the number of shovel tests and mapped the findings from both excavations (Figure 1 2 ). When the data were mapped in ArcMap, t he

PAGE 13

13 p atterns showed that not only are there differences between the eastern, coastal half of the site and the western, more inland, half, but there are also two significant clusters of artifact types within the northern and southern arcs of the ring. The spatia l clustering of two discrete artifact types on opposing sides of the circular midden suggests that Bayview shows signs of dual social organization. T o test this observation further, I split the site into four equally sized and equally spaced quadrants, and aggregated data from the shovel tests from each. I compared the data from each of these quadrants in order to s ee if the visual observation coincided with the proportion of artifacts from each quadrant I predicted that they would reveal similar results, yet there are some interesting points of departure between the results of these two methods. In addition to studying the spatial distribution of artifacts at the site, I have also taken two column samples from opposite sides of the site to compare faunal data. One line of evidence cannot make the case for social organization, because sociality occurs on many different levels simultaneously. It may or may not be that social patterns found reflected in the distribution of pottery or shell tools are the same as what is reflected in the distribution of food refuse at Bayview. One potential hypothesis concerning habitation at the site is that the area near the water was occupied year round, while the remainder of the site was seasonally occupied by visitors, le ading to the greater quantities of artifacts by the shore compared to the inland side of the site. From the faunal material, I had hope d to be able to determine whether the two areas of the site are similar or different with respect to diversity and distribution of species, and whether there are any indications that the site was occupied year round. The r esults from these data do not suggest

PAGE 14

14 s ignificant differences between the two areas of the site, contrasting with the results from the ceramic artifacts. In terms of the struct ure of this thesis, I start in C hapter 2 by discussing the theoretical foundation of this thesis. In this chapter, I address the concept of the community in archaeology and various structural and practice approaches taken in interpre ting sociality within communities, such as the work by Lvi -Strauss, among others. Previous work using these theories is also discussed, from other regions as well as within the Southeast. Chapter 3 examines the history of Weeden Island studies, and th e cultural and temporal context of Weeden Island in the larger sphere of Southeastern United States archaeology. I also outline the basic understanding of Weeden Island culture up to now, including discussion of previous research at the more prominent site s. Chapter 4 introduces the site, and describes my field methods and results. Chapter 5 explains the analysis of ceramic artifacts recovered from both my excavation and previous work done by Russo et al. (2006). Chapter 6 transforms the ceramic dat a into a spatial anal ysis of the site using GIS, dividing the site into quadrants and comparing relative frequencies of artifact types across two separate axes. This chapter also explains the methods used in the development of this approach. Chapter 7 addresses the faunal analysis of the two column samples, one obtained during my excavation and one from 2004. Chapter 8 discusses the findings and examines Lvi -Strauss and similar models, concerning the materialist construction of communities that show signs of being dual societies. It compares the patterns observed from this project to make the argument that Bayview does suggests some signs of duality that could translate to the larger picture of Weeden Island as a whole. Chapter 9 concludes the t hesis with recommendations for future research at Bayview a s well as Weeden Island sites in general. Various appendices and tables follow the text.

PAGE 15

15 Figure 1 1. Weedon Island Sites Referred to in the Text

PAGE 16

16 Figure 1 2 L ocations of Shovel T ests

PAGE 17

17 CHAPTER 2 THE ARCHAEOLOGY OF CIRCULAR COMMUNITIES The Concept of Community The first step in determining how communities use space is to define what communities are, and as simple as this might seem, there are many different opinions as to what a community represe nts. There are three basic and central elements to a community: people, places and the time at which they come together. Yaeger and Canuto (2000:5) suggest that a community is an ever emergent social institution that generates and is generated by supra -household interactions that are structured and synchronized by a set of places within a particular span of time. This definition highlights the regional political economy of communities, emphasizing that they are situated within a spatio -temporal context wh ich first must be defined if they are to be studied. This is particularly important at Bayview, because if different groups are represented at the site, the question would be if they are contemporaneous. The emergent quality of communities, according to Ca nuto and Yaeger, is grounded in practice theory and the work of Bourdieu (1977). This stance clearly separates the concept of community from a traditional evolutionary framework of scale in settlement or household archaeology, and begins to ask the questions of how communities are conceived and are perpetuated through time. Isbell (2000:243) emphasizes that traditional definitions of community are grounded in ideas of (1) shared residence or space, and (2) shared life experiences, knowledge, goals, and sentiments. This view notes that not only are communities the merging of people and place at a particular time, but they are also the gathering of people through mutual senses of relationship to one another. These definitions point to the existence of a community as an event, or even an instance of community. Furthermore, they point to concepts or ideas of togetherness in the minds of

PAGE 18

18 participants, an emic perspective that is not fully conceivable from an etic point of view. Is there a unit of measure that reflects the interactions and experi ences that make up a community? Archaeology does not study time or individual interactions or events, but material culture and the remains of past events. Is it possible to take elements of these definitions and transfe r them to archaeological investigation? One of the most important and ubiquitous elements of the concept of community is the spatial component From this, it may be possible to infer the manifestation of community in its various forms by examining t he way people use space within an archaeological site The Archaeological Community Means (2007:6) introduces the theoretical portion of his study of circular villages in the northeast with the observation that households in village settlements intentionally ch ose to be part of a large r social entity the community This statement points out two factors that show the importance of studying the use of space within a community by archaeologists. First, there is a disparity between studying a singular household a t an archaeological site, and studying a region through settlement archaeology. How communities are formed and existed in the past is just as important as how many communities exist in a region or how a particular household is laid out. Secondly, Means poi nts out the notion of choice on the part of the individuals in the community to become part of that group. The reasons various choices are made as a community are reflected through the material culture of a site, and therefore the interactions and structur e of a community should be visible vis -vis the distribution of artifacts at a site. Pluckhahn (2003:11) uses this concept to interpret the different activity areas at Kolomoki. Using practice theory, he explains that where these practices persisted lo ng enough to be visible archaeologically as community and regional settlement patterns, they reflect traditions and structures (Pluckhahn 2003:11). These patterns, visible archaeologically, may not

PAGE 19

19 offer as much in fine grain data as ethnographic studies, but can still indicate areas where more research can be beneficial to understanding particular cultures. Identifying Community Social Structure through Spatial Analysis Various social institutions are visible through archaeological investigations of c ommunities. Some of the earliest work in archaeology that explores this approach is Longacres (1970) study at the Carter Ranch Longacre explored the concept of community structure by studying the use of space and different types of ceramic artifacts. He states that the material remains of an archaeological site are patterned, and that this is the result of the patterned behavior of the members of an extinct society (Longacre 1970:2). Once the link between artifacts and past behavior was made, Longacre f ormed hypotheses concerning the structure of the society. He was primarily interested in discovering whether there were different types of ceramics found within the site that could indicate that there are different residential groups (Longacre 1970:28) l iving within a single village. Using ethnographic corr elation as well, Longacre argued that not only are distinct groups visible at Carter Ranch pueblo, they were organized through m atrilineal kinship ties. Women we re more likely to have been creating pott ery in the pueblos because in the 1960s they were the primary potters. Although I see problems with this interpretation, mainly the direct correlation between ethnography and archaeology, studies such as these point to the possibility that discrete social groupings are identifiable within a community through the distribution of artifacts (see also Deetz [1967], Hill [1966 and 1968]) Circular Communities The study of circular communities has been approached by many different theoretical viewpoints but mos t often these theories are linked either to structural representations of social organization, or to different int erpretations of the functions of the ring shape. These sites have

PAGE 20

20 been widely interpreted as centers of power, places for dances, sports and games, fishing stations, and domestic structures (Saunders 2004:249). Lvi -Strauss (1963), in his study of moiety divisions in circular villages is largely concerned with the division of space within the ring. Heckenberger (2004:293) suggests that the parti tioning of space reproduces the social organization of the group, through tangible, although often subtle, manipulation of the body t hrough space. Lvi Strauss (1963:139) likens these spatial divisions to the result of a balanced and symmetrical dichoto my between social groups, between aspects of the physical world, or between moral or metaphysical attributes. Russo (2004:3637) proposes that we might find the roots for circular organization in individual interaction, household configuration and reflect ed through these social structures in a larger sense through the configuration of settlements. The study of shell remains at Archaic middens in the southeastern United States serve as a good example of how circular middens are able to reveal layers of social differentiation and inequality. Great quantities of whole oyster shell were deposited at the Rollins shell ring site in three major episodes, visible in trench profiles of the site and separated by thin layers of earth (Saunders 2004). At the Joseph R eed Shell Ring, shell was also found to have been deposited in large amounts in a short amount of time (Russo and Heide 2002:74). At both the Rollins site and the Joseph Reed site, large amounts of small fish were also found, although it is primarily oyste r that creates the greatest visual impact at the site. Russo notes that if shell is deposited quickly, as opposed to the gradual accumulation of daily meals discarded underfoot, relatively less evidence should be found of crushing, wind borne sand, surfac e fires, artifacts, fauna drawn to exposed shell (e.g. land snails) and other subaerial indicators of human or natural activity (Russo 2004:43). This points to the great speed at which these Archaic shell ring sites are formed through feasting activities compared to accumulation from daily refuse patterns.

PAGE 21

21 Russo has also suggested that the intentional construction of Archaic shell ring s has been interpreted as memorialization or monument to the feasting or ceremonies that occurred at the rings (Russo e t al. 2006: 69). Yet research at the Sapelo shell ring complex ( Thompson 2007; Thompson, et al. 2004) has shown that much of the ring was built gradually, as people occupied the site S eas onality studies at the site suggest that people were occupying the si te year round. This could indicate the beginnings of sedentism which could be associated with cultural changes associated with people who occupy space together in a more permanent way. The faunal data, when combined with the seasonality of shell and the more frequent presence of decorated pottery at the Rollins and Joseph Reed sites points to the interpretation that these shell rings were more likely created as special purpose sites (Saunders 2004:261). The smaller lenses containing more gradual accumula tion, diverse species and artifacts at the Joseph Reed site could indicate that this shell ring was used as a habitation site as well (Russo and Heide 2002:75). Russo sees no reason why a site such as Joseph Reed could not have been used for more than one purpose (Russo 2002). He cites the work of Heckenberger, whose description of Kuik uru ring villages is an example of how ring villages can represent habitation areas, political stages, and ritual practice, and that all of these different aspects of Kuik ur u culture are associated in a dialectic manner to comprise the physical utilization of space (Heckenberger 2004). The sociality of communities occupying circular middens has also been examined in the northeastern United States. Means (2007) has taken the theoretical standpoint of Lvi -Strauss, among others, to test whether the circular villages of the Monongahela tradition show signs of social organization on the level of moieties or dual organization models. In this study, architectural features, non arc hitectural features and artifact concentrations at multiple sites were examined to see if there were significant differences between areas of the site. In terms of the

PAGE 22

22 ceramics Means examined both functional as well as stylistic differences, and whether t here were differential distributions throughout any of the sites. The results showed that the clustering he had expected to discover was not obvious from the material culture, but certain factors may have hindered the potential for this type of analysis wi th the Monongahela culture. The analysis of artifacts on a spatial distributional level demands that a large enough sample is brought together in order to effectively observe clustering, and very few artifacts were taken into consideration. Also, t h ere wer e very few types of ceramics found at the Monongahela sites, and many of these do not have specific stylistic attributes, aside from cordmarking. Stylistic differences in this case may be too subtle to view from the limited amount of data on hand. Furtherm ore ceramic artifacts are evenly distributed throughout the sites According to Means (2007), t he even distribution of artifacts may be the most important argument for little social differentiation between family groups at the site. This is not to say tha t there was not social differentiation present within the Monongahela culture but that the archaeological record in terms of material culture did not reveal these elements at the sites in question. Circular W eeden Island Communities In addition to showi ng signs of more complex social organization, it is possible that the layout of coastal Weeden Island ring middens represents both egalitarian and ranked aspects of culture simultaneously. Russo notes that in transegalitarian societies, egalitarian ethics often linger and displays of unequal empowerment among members may be proscribed or tolerated only under specific social circumstances such as at feasts or in other rituals (2004:30). Pluckhahn (2004) has noted that at Kolomoki the earliest habitation at the site shows signs of egalitarian social structure. The egalitarian aspect of culture is evidenced in collective group construction of a built environment, where culture lies in opposition to the natural world, yet within this construction of society li mited imbalances in power are allowed, as evidenced by the

PAGE 23

2 3 unequal distribution of the artifacts and faunal refuse evidencing feasting. In terms of what we know about Weeden Island culture, the maintenance of social bonds and kinship ties appears to be one of the most important aspects of social organization. Summary The central question of this thesis is whether Bayview reflects patterning in social organization through the distribution of artifacts and material culture within its ring -shaped midden. In order to counteract the problems associated with inference and generalizing interpretations, multi ple lines of investigation must be considered, and it is just as important to explore how these data work together as a system and affect the culture as a who le. As Kus has pointed out, The most obvious challenge of the theoretical handling of a concept of social representation is avoidance of both a subjective idealism and a psychological reductionism (1983:282). Social ranking in Weeden Island sites may not be overly obvious within domestic contexts, and different levels of social organization may not be visible in all forms of material cultu re. The types of artifacts that show these types of structure may indicate which aspects of culture are utilized as ex pressions of identity or power or, conversely, unity and egalitarianism. This is why Bayview is investigated from both the spatial distribution of ceramic types and the faunal data which are all theoretically linked through the concept of sociality in arc haeological communities. As each aspect of this thesis is addressed, theory relating to that specific area of specialization will be acknowledged, while simultaneously connecting each to the broader question of how we can view soc iality through the use of space.

PAGE 24

24 CHAPTER 3 THE CONCEPT OF WEEDE N ISLAND Introduction A Weeden Island archaeological culture is a concept that is difficult to generalize as a simple framework of ubiquitous cultural attributes, due to multiple aspects of regional variation and lo calized changes over time. As Milanich (1994:39) has pointed out, it can represent the ceramic series laid out by Willey in 1949, the related temporal period in cultural evolution, the ceremonial complex (i.e. complex burial mounds with caches of specializ ed mortuary pottery), or a level of socio -political and economic integration. In addition to these conceptual problems, there is substantial evidence of regional interpretation of the culture through different modes of subsistence, settlement patterns, pre ceding cultures and transitions to later cultures However, it is useful to outline some of the common ground in Weeden Island before examining the various points of departure. Geographic E xtent of Weeden Island Weeden Island sites are distributed across a large portion of the eastern Gulf Coastal Plain in the southeastern United States Willey (1945: 226) identifies the Weeden Island culture area as extending from Baldwin County, Alabama in the we st, along the Gulf to the Little Manatee River in the south Milanich (1994:160162) describes that the Weeden Island heartland region encompasses northwest and north Florida, and adjacent areas in Alabama and Georgia leaving room for comparative study between these areas and the periphery. Weeden Island develops out of the Swift Creek, Deptford and Yent Green Point complexes during the Middle Woodlands period. Radiocarbon dates are strikingly consistent throughout the region considering the extent and diverse nature of geographic terrain. Some of the various non calibrated C14 dates are offered from different researchers based on particular sites or regions :

PAGE 25

25 A.D. 300 to 600 through 1200 (Bense 1994:170), A.D. 500 through 1000 (Percy and Brose 1974:8), and A.D. 200 through 900 (Kohler 1978, Milanich et al. 1984). M ound Ceremonialism Ceremonialism in terms of mound construction and highly elaborate burial ritual is found at many of the larger Weeden Island sites, spanning the entire geographic and temporal extent of Weeden Island culture. Weeden Island mound ceremon ialism varies among location s as well as time period s ranging from complex mound centers such as Kolomoki with multiple interments and large plazas, to singular small burial mounds at village sites along the Gulf coast Ritual activities took place at the se mounds, at the time of burial as well as afterwards. The ceremonialism associated with Weeden Island, despite differences in scale and numbers of prestige items, retains a common theme, with specialized pottery recalling much of the bird and animal symb olism of earlier Hopewell cultures to the north, east facing caches of burial goods, as well as multiple interments a nd secondary burial practices. The eastern alignment of burial goods and mound structures may also be reflected in the construction of vill age sites. Although domestic and elite ceramic types show significant variation throughout the region in midden contexts, the mortuary or ritual ceramics demonstrate a striking repetition within mound contexts, at sites such as McKeithen in northcentral Florida, Kolomoki in southern Georgia and at the Weedon Island site in Tampa. This pattern signifies a pronounced geographic continuity of a religious or ritual ideology and a widespread interaction among groups separated by great distances. Milanich (1994:164) notes, McKeithen Weeden Island ceremonialism is indistinguishable from early Weeden Island ceremonialism in the panhandle. And radiocarbon dates from the McKeithen site indicate that the period of this ceremonialism is the same in the two regions.

PAGE 26

26 William Sears first excavated the Kolomoki Mound s site between 1948 and 1953. As it stands today, the complex consists of eight mounds, the largest of which measures 17 meters in height, and seven smaller dome -shaped mounds measuring between two and six meters tall (Sears 1953:223). I tems from excavations at Mound D produced 18 extended burials, 11 crematory burials, five bone bundles with indeterminable numbers of individuals (secondary burials), 40 eastward facing skulls (without associated bodies, sec ondary burials), many ornamental goods and artifacts, and a large pottery cache on the eastern side of the mound containing over 64 complete vessels. These vessels were very similar in style to the ones found at the Weeden Island site, and consist of human and animal effigy vessels and jars with cut -out geometric designs (which would be later called killed vessels) (Sears 1953:224228). In Mound E, a similar assemblage was uncovered, but on a smaller scale. Mound E contained an undisclosed number of buria ls, along with fifty -four vessels in an eastern cache. There were less perforated vessels here than in Mound D, but they still represented the majority of the ceramic assemblage (Sears 1953:225228). Social Organization The understanding of social organi zation of Weeden Island cultures is related to the interpretation of many aspects of social behavior, including the expression of a particul ar decorative style in ceramics elaborate burial mound construction and ritual practice, settlement patterns, and e xchange. Ceramics found in mound contexts were the first Weeden Island artifacts to be identified and included in C. B. Moores observations (Moore 1999 [1918]), although midden ceramics were soon included in various typologies as well (e.g. Willey 1949). Sears (1971:54) would be the first to propose a sacred-secular argument for differing ceramic assemblages in burial mounds and middens, noting that, the different cultural phases were the sacred and secular components of a single who le, not two distinct, entities.

PAGE 27

27 Cordell (1984) also analyzed the ceramic assemblage to test Sears hypothesis regarding the sacred -secular affiliation of types of ceramics at McKeithen. She based her analysis on surface decoration, vessel form, site context and manufacturin g technology, to see if there was any difference in the level of craftsmanship between the sacred and secular pieces. What she realized was that instead of a dichotomy between the two types, a third type, prestige ware, was distinguishable (Cordell 1984; M ilanich 1994; Milanich et al. 1984:130). Brenda J. Sigler Lavelle posed a model of social organization for Weeden Island cultures based on work in north Florida (Milanich et al. 1984:40), where villages are thought to have been interacting segments of kinship systems. As the population of the first village outgrew a comfortable population size, a group would bud off and start a new settlement, yet still retain ties to the previous settlement. This model explains the varying sizes of settlements in the north Florida region, at least, and could have applications in other areas of Weeden Island as well. Mound burials with multiple interments were often found associated with the larger villages, and Sigler Lavelle attributes this to a representation of kinshi p ties. A villager who may have left to live in a recently budded-off village may even return to his previous location to be buried in the mound, as a permanent confirmation of lineage membership and it would allow the physical centralization of ritual ob ligations for lineage descendant s (Milanich 1994:169170). Villages or certain kin groups, within this model, could achieve higher status from a broader base of resource availability and possibly even have greater access to large groups of people for community activities. Not only would the original villages have been used for burial and ceremonies, but they could have also been the site of larger aggregations of people for trading and feasting. This could, in turn, cause certain individuals to gain power within the central

PAGE 28

28 village, although this power would be limited according to the abilities of the individual, not passed down to fut ure generations (Milanich 1994: 169170; Milanich et al. 1984:4041). Steponaitis (1986:383) agrees that the burial of cer tain individuals with exotic and prestige goods leads to the interpretation of these persons holding limited amounts of power, related to either trade, ritual knowledge, or the ability to organize labor for large community projects such as the building of mounds. Brose (1979:149) has further suggested that the mortuary complex is predominantly the material expression of a series of local exchange systems structured by li neage leaders. According to Broses model, trade items were often buried with prominent individuals, in order to retain the value of the objects as rare items. Essentially, the social organization posed by these models suggests that within certain cultural and economic constraints of the Weeden Island culture, individuals were able to achie ve rank by organization of trade or specialized ritual. However, these positions of power were tempered by a largely egalitarian social system, primarily evidenced by burial of all ranks of pe ople in the mounds (Brose 1979: 149150). Previous Research So me of the first extensive archaeological work done at Weeden Island sites was the excavation of burial mounds b y C. B. Moore in the early 1900s. Part of the initial draw was the ease in which mounds were located, compared to other more subtle sites, as not ed by numerous scholars prior to 1900 (Willey 1949:17). With the beginning of excavation in the early 1900s, it became clear that mounds usually contained associated grave goods in the form of prestige ceramics and personal ornamentation. V illage contexts were routinely overlooked in the early days of archaeological investigations.

PAGE 29

29 Early Investigations at Weedon Island J. Walter Fewkes (1924) was the first to excavate the Weedon Island site over the winter of 19231924. The Smithsonian Institution, which f unded the initial excavation, has since made efforts to conserve the site. Today a state of the art research facility is located at the site, along with a museum and educational programs for teachers and children. The work done by Fewkes at this site first noted the existence of two types of ceramic styles, separated by context in the burial mounds as compared to habitation areas. Fewkes treated the two assemblages separately and did not make the connection that the same culture was responsible for both. Wi th the advent of radiocarbon dating and further research, it became possible to assess that these two assemblages did fit into the same temporal framework. It is important to note, however, is that the two assemblages were so different from each other that he considered them to be from the result of different groups of people (Fewkes 1924). The Development of the Weeden Island Ceramic Typology The work of Gordon Willey (1945) in the 1940s built on Fewkes research in the Weeden Island area. Willey concent rated his excavations in the Manatee Bay area and revisited the previous work of archaeologist Clarence Moore. Clarence Moore surveyed and excavated over one hundred sites along the Florida Gulf coast between 1896 and 1914, and published his findings throu gh the Journal of the Academy of Natural Sciences. The material gained from these excavations is useful as a starting point for creating typologies, but the drawback of this work is that Moore limited his excavation to burial mounds, and he only made minim al notation of any ceramic items other than what he considered the most spectacular pieces (Steponaitis 1986:363364). Consequently, the collections consist of a small percentage of the available site material, and only from a fraction of the available sit es. Despite these shortcomings, Willeys typology is the mainstay of ceramic analysis pertaining to Weeden Island, even sixty years after

PAGE 30

30 its conception. Willey also discusses the existence of separate contexts for utilitarian ware in the middens and mortu ary vessels in burial mounds: Wakulla Check -Stamped is the most abundant type in the later Weeden Island midden sites. In contrast to the village middens, the burial mounds show very little check stamped pottery [] adorno or affixed effigy fragments, sherds of the large human figurine vessels, and pieces from the special perforated vessels common to the mounds are rarely encountered in midden excavation [Willey 1945: 241]. This distinction between village context and burial mounds would prove to be a groundbreaking discovery that would lead future research on Weeden Island for decades to come. Excavations at Kolomoki Mounds Sears (1953) excavations at Kolomoki produced artifacts that were very similar to those found at the Weeden Island site, yet the architecture and settlement plan were much more complex and imposing. It was because of this that Sears originally placed Kolomoki in the Mississippian period, characterized by large pyramid structures and great plazas. He designated the pottery types as K olomoki ware, which is actually identical to the Swift Creek ceramics found elsewhere in the Weeden Island region. Despite the initial attribution of the site to the Mississippi period, the work done at the Kolomoki site by Sears continues to influence arc haeologists to this day. Pluckhahn (2003: 9 11) notes that neither processual nor actor -based interpretations of ceremonialism are appropriate to understanding ritual processes, and that a historic approach is more informative. In his interpretation of cer emonial activities, however, Pluckhahn still uses ethnographic and historic references to gain inference to interpret these activities. His point is not to erase these lines of evidence, but to involve a multi -scalar level of investigation and the types of interactions that these different aspects of culture have on each other in a dialectic relationship.

PAGE 31

31 Pluckhahn describes that during the establishment of Kolomoki as a large regional center of activity and trade, people initially existed as a basically e galitarian group. He further identifies a suite of ceremonial activities that did not evidence a division in status between people. Instead, these activities reinforced community ties and allegiances, including the process of the formation of the mounds over long periods of time involving the work of many people, the inclusion of a cross -section of the cultural representation in the burial mounds, and a lack of social stratification in the households excavated. The construction of the enclosure that wraps a round the entire site, while it is not the single most impressive structure at the site, would have been the most labor intensive, and would have required the cooperation of a large number of people to complete (Pluckhahn 2003:192195). At the height of mound construction, during the Kolomoki II phase, tension grew between the apparent egalitarian ethos of the people at the site, and a possible inclination of certain groups or individuals towards aggrandizement. At this time, burial ritual also reached th e peak of complexity, and Mound D was constructed, the largest mound at the site (Pluckhahn 2003:198). However complex Kolomoki may have become, evidence of group status and solidarity appear to be more evident than single individual leaders, through the c ontinuity of possibly culturally enforced egalitarian practice. If this was the case, then there is no need for separation between ceremony and village life at Kolomoki, and this possibly could be true for other Weeden Island sites as well. However, cer tain sites do show signs of nonegalitarian distribution of goods such as ceramics, personal items and midden deposits. As our knowledge of Weeden Island culture grows, future investigations will need to continue in this vein to see if this type of approach ca n identify similar or opposing interpretations in other regions.

PAGE 32

32 McKeithen Weeden Island Jerald Milanich excavated the McKeithen site over the course of six field seasons in the 1970s. The resulting report was a culmination of the collaborative efforts of many archaeologists, studying the site from such differing aspects such as faunal analysis, ceramic analysis, site layout, and the examination of mortuary practices from the artifacts found in the burial mounds. McKeithen is different from both the Weeden Island site and Kolomoki in that it falls between the two in s ize, and its location in north -central Florida places it geographically between Kolomoki and Weedon Island. The site consists of three large platform mounds, with evidence of a structure placed on top of one of the mounds, a plaza, interment of a person with associated artifacts pertaining to bird symbolism, and large amounts of ceramic material (Milanich 1997:119). Similar to the other two sites previously discussed, McKeithens ceramic assemb lage initially showed evidence of two specific types of pottery, secular and sacred, according to style and site context in the village middens or in the burial mounds. What Milanich had expected to find in the site assemblages, after testing the area in 1976, was that McKeithen would prove to be similar to Kolomoki, yet on a smaller scale (1997:117). He had initially thought that each of the three mounds present at the site would represent one individual, whose interment dated to different eras of site occupation, and would contain material culture from each of those periods (1997:118). However, the data from the next six field seasons suggested that the three mounds were all from the same period, and that village settlement began before mound constructio n and lasted well after ritual activity at the mounds had ceased. This suggests that McKeithen i s an anomaly compared to other Weeden Island communities from the same region and period, which had either only one mound or none at all. Ann Cordell conducted three separate ceramic analyses of the vessels and objects found at McKeithen, and achieved new discoveries about Weeden Island ceramic techniques and

PAGE 33

33 technologies. One of the marked differences she discovered in the McKeithen assemblage was the absence o f Swift Creek pottery. Additionally, through the analysis of trace elements found in the paste of the ceramics, Cordell was able to assess that while some of the elite and mortuary vessels were from nonlocal sources, some were certainly made at the McKeith en site (Milanich 1997:123124). This leads to the good possibility that the mortuary ceramics were not all made at one location, but had multiple production areas that were connected at least by iconographic style. Cordell (1984) also notes that differ ing types of paste were used for the production of cult ware as compared to utilitarian ware, and these may be the result of what type of clay was considered most appropriate for each type of vessel. She argues that the thicker -walled, lower temperature fi red utilitarian vessels are more durable for cooking use, and that the higher temperature firing of the cult wares makes them much more friable and less functional as cooking vessels. One issue that is not address ed is that if objects were made with diffe rent firing temperatures, they would need to be fired separately, and perhaps by different people. This could infer that special firings would be solely for ritual vessels, and possibly even that only certain individuals were allowed to create these pieces of pottery. Wakulla Weeden Island The inland northwestern Florida Wakulla Weeden Island culture probably shows the most dramatic break from Weeden Island as described. At this time, the hunting-gatheringfishing society began using maize agriculture in addition to the previous economic pursuits (Milanich 1994:194), and settlements became much more widespread in order to make room for shifting agricultural plots (Percy and Brose 1974:2022). Mound ceremonialism and circular villages wer e no longer in use at this time in this area. This period in Weeden Island culture is considered the latest, occurring immediately before the switch to Fort Walton culture in northern

PAGE 34

34 Florida. This most likely indicates a shift in political structure in the interior Wakulla area, stemming from the shift in economic strategy, that led to a marked decrease in community ceremonial activities and the lack of specialized pottery classes (Percy and Brose 1974:22). This development would characterize the end of Weeden Island culture as it had been known for centuries and the transformation towards intensive agricultural societies in much of the region. However, there is evidence of Wakulla ceramics at coastal sites in northwest Florida (Russo et al. 2006), and no signs of agriculture or changes in settlement structure. Coastal groups would persist in hunting -gathering -fishing subsistence practices until European contact. Coastal Weeden Island Sites The argument for a coastal variant of Weeden Island sites is called for by the striking differences that coastal sites have from inland sites. By looking at various aspects of Weeden Island material culture, patterns emerge that illuminate the fact that although people from the coast were similar to inland groups in many ways, there was a di stinct regional expression that was certainly influenced by the experience of coastal living. In addition to previous approaches centering on ceramic analysis, there is enough data available to lead to an understanding of the contrasting elements of inland and coastal settlements, in terms of subsistence, site function, social organization and size of settlement. Coastal Weeden Island sites are mostly comprised of smaller ring middens, which are oftentimes associated with a singular burial mound. As mentio ned earlier, a heavy reliance on maritime resources is present in the faunal material obtained from midden assemblages. Little excavation has been done in this region on these types of sites, in order to establish habitation area boundaries and depth of de posits.

PAGE 35

35 Old Homestead, Walton County, Florida This Santa Rosa Swift Creek/Weeden Island site is located on the northern side of Chocktahatchee Bay, and represents what Bense (1992) considers a classic example of a circular Santa Rosa -Swift Creek ring mi dden, although large amounts of Weeden Island material are located within its assemblages as well. The ring itself measures approximately 70 meters in diameter today, but erosion tests have shown that the site may have been closer to 100 meters at the time of its occupation. Old Homestead is a very shallow site, with deposits reaching to depths of 30 centimeters. Layers of oyster and clam shell represent the two components, the oyster is associated with Weeden Island material while the clam is associated wi th Santa Rosa/Swift Creek assemblages. A large amount of evidence from marine resource seasonality point to the village being occupied year round, and subsisting mainly from estuary fish and shellfish, and limited amounts of terrestrial animal, represented by a few white tailed deer ( Odocoileus virginianus ) and various species of turtle. Numerous post molds were also discovered at the site indicating domiciliary structures in a circular formation, surrounding an open plaza (Bense 1992). Plantation West, S anta Rosa County, Florida The site of Plantation West is most notable for the systematic shovel testing done by Doran and Piatek (1985), which identified the presence of twelve discreet middens, most likely affiliated with households that were situated in a circular formation with a sterile central plaza area. These households also exhibit differential distributions of artifacts, which may indicate some level of social inequality at the site (Russo et al. 2006:8687). Bird Hammock, Wakulla County, Florida Bird Hammock is located within the St. Marks National Wildlife Refuge, about two miles from the Gulf Coast in Northwestern Florida. Bird Hammock has a long history of

PAGE 36

36 archaeological investigation, beginning with the excavation of the two small burial mounds by C.B. Moore (1918), in which he discovered as many as eleven burials in one mound and also the typical east side cache of ceremonial pottery. After subsequent excavations by Willey in 1948, which identified the circular formation of a refuse midden, Bense excavated the site in 1968, and wrote her thesis in 1969 on material from this site. The midden measures 240 feet in diameter, or 73 meters, and many faunal remains were recovered. At that time, comparative collections for the region were unavailabl e, and there was a lack of qualified zooarchaeologists to analyze the material. Subsequently, faunal analysis would not be completed until almost 40 years later. Nanfro (2004) examined the faunal material from four of the features at the site, and found a mix of both inland and coastal faunal remains. What is interesting, in contrast to Milanichs (1974) interpretation of the inland site of Sycamore as a seasonal encampment, is that Bird Hammock appears to have been occupied year round, evidenced by the presence of migratory species of birds and sturgeon (suggesting November through April occupation) and seasonality of scallops (indicating June through November settlement) (Nanfro 2004:61). The faunal analysis, along with the presence of more than one bur ial mound and a large settlement size, places the Bird Hammock site at what would best be described as a transitional locale, fitting between the coastal and inland models that I have described. These types of sites are to be expected, as people occupy regions with more than one ecological zone Summary The concept of Weeden Island consists of many similar identifying characteristics that connect sites to each other as culture, despite great distances. However, the differences described above also allow f or regional variations. Inland mound complexes such as Kolomoki and McKeithen reflect labor -intensive efforts, which involve more complex social organization and leadership, intrinsically connected in the case of Weeden Island to ideology and ritual activi ty.

PAGE 37

37 Evidence of these activities, at this scale, has not been found at any of the coastal sites, and indicates that there may have been a different type of political organization on the coast as there was at inland sites. Does this indicate that in order to better understand Weeden Island coastal sites, a different methodology needs to be employed at coastal sites ? Studying mounds is not possible at many coastal sites, since they have either been destroyed or may not have ever existed. Spatial analysis of villag e sites such as the work presented in this thesis, can be helpful in understanding social organization Field methods and data recovery at Bayview are discussed in C hapter 4.

PAGE 38

38 CHAPTER 4 FIELD METHOD S AND DATA RECOVERY The Bayview Site (8BY137) Bayview is located on Tyndall Air Force Base, on the northern side of the peninsula directly on the East Bay (Figure 4 1) East Bay is a small bay connected to the Gulf of Mexico, just southwest of Panama City, Florida. Bayview is named for, and is located directly underneath, the Bayview military housing subdivision, just off Tyndall Parkway T hree previous excavations at Bayview have been filed with the Florida Master Site File. The first excavation was done by Knudsen (1979), which gave the site its numerical designation at the Florida Master Site File In 1985, New World Research did some minor testing at the site, and suggested that it was an arc -shaped midden that followed the shoreline of East Bay. In 2004, the Southeast Archaeological Researc h Center (SEAC), a branch of the National Park Service, initiated an intense survey of the site, and discovered that it was actually a much larger ring -shaped shell midden (Russo et al. 2006). In order to determine the extent of Bayview, Russo et al. (20 06) excavated shovel tests at Bayview on a systematic metric grid, with test points placed 20 meters from each other in all four cardinal directions. The grid was placed 10 degrees east of magnetic north. The shovel tests excavated in 2004 were 30-by30 cm square units, which were dug until the excavator reached sterile soil. According to the report, tests were excavated to a minimum depth of 75 cm unless standing water and/or culturally sterile subsoil was reached prior to 75 cm (Russo et al. 2006:37). A rchaeological materials present at Bayview included ceramics, lithics, bone, shell and shell tools. In 2004, Russo et al. (2006) extracted a 50 by50 cm column sample in the field at grid coordinates 849N 480E. This sample was taken from the eastern part of the site, which consisted of the deepest and most dense shell midden. Heavy disturbance was noted in the 2004

PAGE 39

39 excavation (Russo et al. 2006:3233), as evidenced by the combination of modern construction materials with prehistoric artifacts in some of the test units. During my excavation, I also noticed certain areas of the site were heavily disturbed, due to road and housing construction, and the demolition of houses. The results of Russos excavation showed that Bayview was a circular midden, with heav y concentrations of artifacts on the eastern side of the site. This observation called for further testing to look into the specifics of these differences in artifact distribution, and to test the hypotheses that these differences were related to different ial social status or ethnicity at the site. In order to do this, I double d the shovel tests at the site to see if a clearer image of the site was possible, and to see if the clusters could be delineated to be differential activity areas. Causes and Potent ial Impacts of Modern Disturbance at Bayview Bayview is the largest coastal Weeden Island ring midden found to date, and although much of the site has been disturbed, the site still holds great potential for research. Bayview is currently still threatened due to its location beneath a military housing development that is slated for partial demolition. Bayview was first impacted by the construction of military housing and paved roads, but has been even more significantly impacted by the demolition of four houses at the site. Russo et al. (2006) noted that the more sparse areas of artifact concentrations at the site could possibly be due to the level of impact from demolition. Their recommendation for the site was that demolition cease because of the impend ing impact this would have to the integrity of the site (Russo et al. 2006:109). Bayview, despite its problems due to construction and demolition, is still a very important site to examine. In many cases during my excavation in 2007, disturbance was limi ted to the top 10 cm of subsurface, with many shovel tests holding virtually undisturbed midden. The general formation of the site is still intact, and some areas impacted by construction and demolition still have very

PAGE 40

40 deep deposits of cultural material. The type of construction may have something to do with the level of preservation, being concrete block homes, which necessitate little excavation of subsurface in order to construct. The high frequency of shovel tests near houses containing intact midden d eposits evidences the minimal level of disturbance from this type of construction. In fact, one shovel test placed less than one meter from a house contained intact midden material to over a meter below surface. Eighteen of the shovel tests revealed intact shell midden. Some of these tests revealed that the shell midden had been buried by fill dirt or possibly grading. It is possible that some of the material at the site had been sheared off the top of the midden as the site was graded in preparation for th e construction of military housing at the site. Grading may have also pushed the midden around a bit in order to get the ground flattened, but that not all of the material was removed. It is only in the areas of demolition and in the construction of paved roads that impact has seriously compromised the integrity of the site. Because of the potential bias these areas of disturbance could make to the interpretation of spatial distributions of artifacts, note d all areas of observed subsurface disturbance, as well as the depth of the disturbance whenever possible. Despite the potential problems the disturbance may have on certain areas of the site, much of Bayview is largely intact, and it is unlikely that the construction and demolition of houses in this case holds much weight in negating the value of examining the spatial distribution of artifacts at the site. For this thesis, I assume that the patterning and clustering of artifacts at Bayview are now the result of post depositional processes, unless I otherwi se acknowledge particular disturbance effects. Establishment of Systematic Grid for Shovel Testing In order to be able to combine the data from my excavation with the data from the 2004 excavation, I followed the same basic field methods for excavation, w ith the exception that I weighed and noted the depth and extent of shell midden deposits from the shovel tests. In my

PAGE 41

41 excavation, I doubled the number of shovel tests by staggering my units between the original grid points. This was done by locating one of the previous excavations original points, and orienteering a north -south line from that point, following their original grid. From the baseline, tests were placed at 10 meters east and west initially, then 20 meters from there on. As far as grid point co ordinates, I followed the same numerical system as previously established at the site, in order to maintain consi stency, as well as creating a way to locate shovel tests from both years Field Excavation and Processing of Shovel Tests I chose to follow these same guidelines in order to achieve a comparable dataset. I wanted to make certain that the sample s taken in my excavation would be as similar as possible to Russos, in order to look at the spatial patterning of the site. In all, 98 shovel tests were excavated at Bayview in 2007, 78 of which contained archaeological materials. All material from shov el tests was screened through 1/4 inch mesh, again keeping with the same method s as the previous excavation. Artifacts were removed from the screens and pl aced into labeled bags indicating which point on the grid the test was located, as well as a field sample number. When a test was negative, no field sample number was assigned to that test. In tests where shell midden was still intact, shell was weighed in the field in a bucket with a hanging scale. Shell was not collected from shovel tests, and as excavation was completed for each unit, the shell was placed back in the unit along with the backfill. Modern artifacts (such as glass, bottle caps, housing mate rials, etc.) were not ed, but not collected. Archaeological materials present at Bayview included ceramics, lithics, bone, shell and shell tools. Ceramics are the most prevalent material at the site, and relatively low amounts of lithics were encountered. This thesis does not address the presence of lithics or shell tools, but these artifacts are curated along with the rest of the materials found at the site discussed herein.

PAGE 42

42 A full listing of shovel test locations, depths and disturbance (if present) is a ttached as Appendix A. Appendix B consists of a list of all archaeological materials present in the shovel tests, separated into very basic classes of ceramics, lithics, shell tools, and shell. As tests were completed, I recorded the GPS coordinates of each test location. These points were taken with a Magellan eXplorist 210, which is WAAS enabled and has a metric accuracy of less than 2 meters. I recorded both the coordinates a nd the accuracy of the readings. Certain fluctuations in accuracy are due to the canopy of magnolia trees and possible multipath interference from the East Bay in some areas of the site Excavation of Column Sample 2 In addition to the ceramic data from the site, a second, independent dataset was examined, resulting from the exca vation of a column sample from the e astern side of the site. In 2006, Russo et al. extracted one column sample from the western side of the site. The excavation of the second column allowed for the comparison of faunal material from bo th sides of the site I excavated a 25 -by 25 cm column at 870N 371E in 10 cm levels to 50 cmbs and bagged each level in its entirety for processing in the lab. The column sample I excavated was taken from the field in its entirety and was screened in the lab with nested box s creens of 1/4, 1/8 and 1/16 inch mesh. Screening techniques are of specific interest to coastal sites, where nearly 40% of material is lost through 1/4 mesh (Russo 2006:57). Summary T he field methodology outlined in this chapter fulfilled two main pur poses: to collect a representative sample of artifacts from Bayview, and to retain spatial information in order to form hypotheses regarding social organization vis -vis spatial variation and artifact distribution across the site. Ceramics are the most f requently encountered artifact class at Bayview. Shovel testing revealed that in all positive test units, ceramics were encountered. This creates an

PAGE 43

43 opportunity to use the ceramics as means of testing whether there are different areas of activity or artifa ct types throughout the site. The presence of many different ceramic types at Bayview allows for an analysis of the sherds from a formal stylistic perspective. The initial research question pertaining to the ceramics is which types are most commonly enco untered at Bayview, and once this has been accomplished, a spatial analysis of these types can be performed. Chapter 5 will address the types of ceramics found at Bayview, and their relative frequency of occurrence. By combining the ceramic and faun al material from Russo et al.s (2006) excavation with my data from 2007, a much larger dataset of archaeological material and spatial information is now available to look at the distribution of artifacts th roughout the site. It is to the ceramic portion o f the analysis that I will now turn.

PAGE 44

44 Figure 4 1. Location of Bayview on East Bay.

PAGE 45

45 CHAPTER 5 CERAMIC TYPES AT BAYVIEW Introduction The purpose of this thesis is to examine whether social organization is evidenced through the spatial distribution of artifacts throughout the Bayview site. The initial step in such an exercise is to identify whether there are distinct types of ceramics to analyze. The use of style in ceramic analysis was chosen as an approach, theoretically based on the work of Wobst (19 77), who suggests that style is an important tool used to express identity (among other things) through material culture. Style can be used in order to communicate social allegiance to a certain group, as well as to define mutually expectable behavior pat terns and makes subsequent interaction more predictable and less stressful (Wobst 1977:327). Willey (1949) divides Weeden Island ceramics into two categories Weeden Island I and Weeden Island II. The most commonly encountered Weeden Island I types are Weeden Island Plain, Carrabelle Incised and Carrabelle Punctated Weeden Island II is distinguished by high percentages of Wakulla Check Stamped sherds (Willey 1949:397). Although this resource is still the standard in identifying Weeden Island types, the legitimacy of Willeys chronology and time periods is questionable. The paucity of radiocarbon dates available from Weeden Island sites is one of the problems, but with the resurge nce of current interest in Weeden Island studies, these issues are starting to be resolved. For the purposes of this thesis, I have used Willeys (1949) typology, as well as agreeing with his rough classification of time periods or phases (Weeden Island I, Weeden Island II, etc.) but do not assign specific dates, due to the lack o f consensus on this matter. I will offer my own dates, based on radiocarbon assays, in order to add to the information we have on proposed dates for the various Weeden Island divisions

PAGE 46

46 The following section outlines the distinguishing features of the ty pes of pottery found during the 2007 excavation at Bayview, based on Willeys (1949) descriptions. Table 5 1 shows all sherds from the 2007 excavation by type. Because all sherds come from shovel test units, there are not many cross -mendable pieces, and no complete vessels. Therefore, identification was based primarily on surface decoration. The various types found at Bayview are described in detail below in order of which types were most commonly found, based on weight of sherds Ceramic Types found at Ba yview Weeden Island Plai n Weeden Island Plain sherds are distinguished by a folded rim, most often on the exterior side of the vessel, and may have an additional incised line around the fold of the rim. The incised line can also be used in place of a fold ed rim. Because this type is identifiable only from its rim sherds, deriving accurate counts of this type in archaeological assemblages is problematic. It is certainly possible that many of the other plain sherds in the assemblage can be included with Weed en Island Plain. Plain sherds could also be undecorated portions of decorated type s such as Swift Creek, which have similar paste and surface treatment as Weeden Island Plain. Wakulla Check Stamped The paste and temper of this type vary greatly, but are broken into three main categories by Willey (1949:437438). These three types include one similar to Weeden Island types, using fine sand as a temper, and range to using much more coarse pieces of sand and to using hardly any temper at all. Wakulla Check S tamped is much different from other Weeden Island types in that the vessels are mostly covered with stamping, and that there is no curvilinear motif to the design. The stamps are created by carved paddles. This type is mainly affiliated with the later Weed en Island II, extending into Fort Walton.

PAGE 47

47 Swift Creek Complicated Stamped Willey (1949) describes the stamping of late variety Swift Creek pottery as consisting of mostly curvilinear and complex designs. The most common design motifs are hatched teardr op or snowshoe elements; concentric spirals; concentric circles as part of a greater design; interlocking scrolls and rectilinear elements; intertwined meander; simple concentric curved lines ( Willey 1949:431). The stamping was done with a carved woode n paddle, and is usually limited to the top portion of the vessel. The main difference between the late variety associated with Weeden Island sites and the earlier Swift Creek variety of predominantly Swift Creek sites is that the execution is done more c arelessly and boldly (Willey 1949:431) in the later period, resulting in a less refined appearance of the design motifs. Carrabelle Punctated Carrabelle Punctated has the widest variation in its design of all these types. The paste and temper are simila r to Weeden Island Plain, and decoration is limited to the upper portion of the vessel. The variation comes mainly from the tool used to make the impressions, and Willey lists eight possible shapes and sizes of punctuations. This type is associated with bo th Weeden Island I and Weeden Island II Carrabelle Incised Carrabelle Incised also has a similar paste and temper to Weeden Island Plain, and has similar design qualities to the Crooked River Complicated Stamped variety, only the chevron style markings a re lined up into rows or into what Willey (1949:422) calls herringbone style patterns, and the designs are incised rather than stamped. The placement of these incised decorations are most often within zones near the tops of vessels, bordered by incised l ines as well. The main difference between the Carrabelle and Weeden Island Incised varieties is that Carrabelle does not appear to contain representational depiction of symbols or objects, while

PAGE 48

48 Weeden Island is very expressive in this realm. Carrabelle In cised is associated with both Weeden Island I and Weeden Island II Weeden Island Incised The paste and temper of Weeden Island Incised are similar if not identical to that used for Weeden Island Plain wares. The incised lines created by cross hatching or parallel hatching comprise the background for the central design image, which is left plain. Backgrounds are also created by punctuation by a blunt instrument. The technique is similar to the first step of carving a relief image in wood or stone, in which pieces of the matrix are modified and taken away in order to reveal the image intended by the artist. Weeden Island Incised images are separated from the background by an incised outline as well. Some of the images are representations of birds, while othe rs are abstract curvilinear shapes such as curvilinear meanders and simple and compound lobate forms (Willey 1949:413). Within these abstract shapes, an additional incised line often exists, with larger dots at the terminal points of these lines and inte rspersed throughout the line. Willey (Willey 1949: 413) offers the highest praise on the artistic quality of these vessels, stating the craftsmans control over his medium is not excelled anywhere, or at any time, in the pottery of the eastern United State s Usually these decorations are all within the top portion of the vessel. This type is associate with both Weeden Island I and Weeden Island II. Weeden Island Punctated This paste and temper of this type are also very similar to a Weeden Island Plain, m aking it very similar to Weeden Island Incised as well. The curvilinear design patterns are also similar to Weeden Island Incised, except for the point of departure where the design is defined by closely spaced punctation marks instead of the outlined imag e in the previous type. Punctation marks create the background of the central image. Very similar design motifs are used in this type, including the bird images and meandering or lobate designs. Surface designs are also

PAGE 49

49 limited to the top portion of the ve ssel in this type, similar to Weeden Island Incised. This type is mostly affiliated with Weeden Island II. Mercier Red on Buff This type has a similar paste and temper to Weeden Island Plain and a similar paint color to Weeden Island Red, but is distinct from the rest of the assemblage at Bayview because of its buff -colored paste. This type is more commonly found north of the Gulf sites, and is most often associated with the Kolomoki Mounds site (Cordell 2007, personal communication). Weeden Island Red T his type is exactly the same as Weeden Island Plain, but has the addition of red paint on the interior or exterior surface of the vessel. These sherds may also be part of Weeden Island Zoned Red, a type where portions of the vessel are incised and some are as painted red. Keith Incised This type has the same paste and temper as Weeden Island Plain as well. Keith Incised is described as having a band of diagonal cross hatching (Willey 1949:427), sometimes with punctuations at the cross points of the incis ions. The design extends directly from below the rim of the vessel only around the top portion of the vessel. All lines are incised first in one direction, and then the other to create the cross hatch effect. Keith Incised is from both Weeden Island I and Weeden Island II periods. Ruskin Linear Punctated Ruskin Linear Punctated and Weeden Island Plain have similar pastes and tempers. The decoration on Ruskin Linear Punctated is a series of incised and punctuated lines, created by punctuating the surface of the vessel, and when lifting the tool, dragging it across the surface to the next point at which a punctation is made. This results in a similar geometric pattern to Keith Incised, however the technique utilized in creating the decorative motifs is differ ent because it is

PAGE 50

50 punctuated instead of incised This type is mostly related to the Weeden Island II phase, but may be associated with Weeden Island I as well. Cr ooked River Complicated Stamped Crooked River also has an early variety associated with Swift Creek sites, but the paste is tempered with fine sand and mica. The later Weeden Island variety is tempered wi th coarse sand or grit (Willey 1949:435). Decoration of these vessels is similar to Swift Creek Complicated stamped, especially in the technique of using carved wooden paddles, yet the Crooked River variety is much more rectilinear in shape, consisting of a series of zigzag lines or chevrons (Willey 1949:384). Tucker Ridge Pinched This type has a similar paste and temper to Weeden Island Plain, and is identified by a zone of pinched rows of clay that are perpendicular to the vessel opening. This type is affiliated with Weeden Island II phase, and Willey (1949) describes that it is similar to the Coles Creek Ridge Pinched from Louisiana. Indian Pass Incised Indian Pass Incised has a similar paste and temper to Weeden Island Plain, and designs are described by Willey (1949: 426) as a close spaced arrangement of lines giving almost a combed appearance in some cases. The designs on the Indian Pa ss sherds from Bayview look very similar to fingerprints, with their curvilinear structure and concentric ring shapes. The design motifs are most often limited to the top portion of the vessel. Indian Pass is associated with both Weeden Island I and Weeden Island II periods. Dunlap Fabric Impressed Although Willey (19 49: 86, 120) includes fabric impressed sherds in his types at both Fort Walton and the Thomas Site, he does not describe these types in detail in his type descriptions.

PAGE 51

51 The one sherd of this va riety was identified by comparing it to various fabric impressed sherds from the Ripley Bullen collection at the Florida Museum of Natural History. This type is associated mainly with the Deptford period of ceramics, which predates Weeden Island by a signi ficant amount of time. Laboratory Processing and Identification of Sherds Ceramic artifacts were washed and air dried upon returning from the. After washing, artifacts were sorted according to the type of artifact (ceramic, lithic, shell tool). The cera mic artifacts were taken into the Florida Museum of Natural His tory after fumigation, and were analyzed in the Ceramic Technology Lab, with the assistance of Ann Cordell. Upon identifying the sherds accord ing to the established type s, data from all the s hovel tests were put into an Excel spreads heet. The quantitative data were examined from two perspectives, as a separate dataset, and combined with the data in Russo et al. (2006). The results from the 2007 analysis follow Results In all, the 2007 excava tion produced 1683 sherds for analysis, with a total weight of 8050.6 grams (Table 6.1). The results from this excavation are very similar to those from previous work at the site, which is not surprising considering the consistency in methods of excavation and the use of the same typology for identifying sherds based on decoration and paste. I have divided the results into four main categories: Plain sherds and Decorated sherds, with two subdivisions for decorated pottery: Early and Late Weeden Island types Although the actual dates pertaining to when these transitions took place are under scrutiny today, the ordering is consistent throughout much of the literature (see Kohler (1978); Milanich (1994); Percy and Brose (1974); and Willey (1949)) and serve as a way to divide the discussion of the results.

PAGE 52

52 Plain Sherds The most common types of ceramic sherds found in the excavation are plain sherds. These types are found throughout the Weeden Island period, and potentially represent all periods of Weeden Islan d cultural horizons. Plain sherds are divided into burnished, smooth, and residual, meaning sherds which have not been smoothed or burnished. Weeden Island Plain sherds are also noted, but this identification is only possible when a sherd has the diagnosti c folded or incised rim. It is possible that some of the plain sherds cross -mend with decorated sherds. Russo (2008, personal communication) has noted that many Weeden Island ceramic vessels are 20 to 80% plain, so that many of the sherds identified as pla in could in fact be part of a decorated vessel. The 2007 data show that a very large portion of the ceramic sherds are plain (n= 702), and ac count for 51.27% of the weight (4127.5 grams). From this category, burnished plain vessels are most common (n=334) and account for 24.57% of the total weight of the total ceramic assemblage (1977.9 grams). Weeden Island Plain sherds are less common (n=59), and account for 7.15% of the total weight of the assemblage (575.9 grams). Decorated Sherds Several elite type de corated varieties are found throughout the Weeden Island period, and do not have a clearly defined temporal affiliation within the larger sphere of Weeden Island. This is largely to do with their ubiquitous nature in many Weeden Island contexts throughout the region and time periods. At Bayview, the types encountered are Weeden Island Incised (n=21), Weeden Island Punctated (n=16), Weeden Island Red (n=9), Weeden Island Appliqued (n=1) and one decorated sherd with a Weeden Island rim. Together these make up 2.98% of the total weight of the assemblage (239.7 grams).

PAGE 53

53 Early Weeden Island The earliest type of pottery at Bayview is a Dunlap Fabric Impressed sherd (n=1, 4.1 grams), from the Deptford period. Swift Creek sherds are more common (n=53) and are account for 4.70% of the weight of the total assemblage (378.6 grams). The early Weeden Island decorated sherds, including Carrabelle Incised (n=17), Carrabelle Punctated (n=20), and Keith Incised (n=13) make up 3.62% of the weight of the total assemblage (291.7 grams). Late Weeden Island Very few elite type decorated sherds from the assemblage are associated with the later phases of Weeden Island. One Mercier Red on Buff sherd was found at Bayview, a type found frequently in the Kolomoki area, yet very infre quently in the Florida Panhandle. Additionally, Indian Pass Incised (n=1), Ruskin Linear Punctated (n=1) and Tucker Ridge Pinched (n=3) sherds were found, all accounting for 1.0% of the total weight of the assemblage (80.4 grams). Wakulla Check stamped is also commonly associated with later phases of Weeden Island by many researchers (Kohler 1976; Kohler 1978; Milanich 1994; and Percy and Brose 1974). Check stamped pottery in general is considered to mark the decline in incised and punctuated forms of d ecoration, indicating a significant shift to less individualistic design. Wakulla Check Stamped is the most commonly found decorated ceramic type at Bayview (n=276) and represents 26.83% of the total weight of ceramics excavated in 2007 (2160.3 grams). Su mmary The results of the formal stylistic analysis of the ceramic sherds from Bayview indicate that the site may have been occupied from the end of Swift Creek chronology throughout Weeden Island II time periods. This is not uncommon at Weeden Island site s in the area, or regionally, where combinations of many different types such as these are often found. Now that the types of ceramics found at Bayview have been identified, it is possible to combine the results

PAGE 54

54 from this analysis with the spatial informat ion in order to examine the distribution of artifact typ es across the site. In Chapter 6 I will discuss the spatial patterning of the ceramic artifacts at Bayview.

PAGE 55

55 Table 5 1. Total Ceramic Data (2007) Series/Type Count Weight (g ) DEPTFORD Dunlap Fabric Impressed 1 4.1 SWIFT CREEK Swift Creek Complicated Stamped 51 370.1 Crooked River Check Stamped 2 8.5 WEEDEN ISLAND Carrabelle Incised (E) 17 121 Carrabelle Punctated (E) 20 131.3 Keith Incised (E) 13 39.4 Weeden Island Plain (E/L) 59 575.9 Residual Plain (E/L) 144 642.8 Smooth Plain (E/L) 165 930.9 Burnished Plain (E/L) 334 1977.9 Weeden Island Appliqued (E/L) 1 14.7 Weeden Island Decorated (E/L) 1 1.8 Weeden Island Incised (E/L) 21 104.1 Weeden Island Punctated (E/L) 16 73.9 Weeden Island Red (E/L) 9 45.2 Mercier Red on Buff (L) 1 45.2 Indian Pass Incised (L) 1 4.3 Ruskin Linear Punctated (L) 3 25 Tucker Ridge Pinched (L) 1 5.9 Wakulla Check Stamped (L) 276 2160.3 OTHER Cordmarked 2 11.1 UID Appliqued 1 2 UID Complicated Check Stamped 1 6.4 UID Complicated Stamped 4 28.7 UID Decorated 7 58.2 UID Incised 38 146.7 UID Punctated 5 10.8 UID Shell Impressed 1 5.7 UID Stamped 30 186.3 UI D Zoned Stamped 1 5.2 Too Small 402 307.2 TOTAL 1628 8050.6

PAGE 56

56 CHAPTER 6 SPATIAL ANALYSIS OF CERAMIC ARTIFACTS Introduction In this chapter I describe the reasoning, methods and results of s everal tests that were developed to look at the spatial dist ribution of pottery sherds at the site. The first step in looking at the spatial characteristics of Bayview was to determine whether all areas of the site were contemporaneous or multi -component in nature. The second test was to investigate whether there w as nonrandom distribution of ceramic types throughout the site. These finer -scale questions are linked to whether Bayview suggests dual social organization, by comparing the artifact concentrations from opposing areas of the site. A dding the data from Russ o et al. s (2006) report to the data from my excavation in 2007 increased the sample size from 1628 sherds weighing 8050.6 grams to 2589 sherds weighing 13,882.92 grams in total, an adequate sample size for the study at hand. Dividing Bayview into Spatial ly Discrete Clusters Russo et al s (2006) report demonstrated that Bayview was a ring-shaped midden with differential distributions of artifacts across the site. One of their interpretations was that the eastern and western halves of the site showed signs of being very different from each other, in terms of distribution of artifacts, and artifact density. With the combined data from both excavations at Bayview, I divided the site into four quadrants, with the axes of division running north -south and east w est through the middle of the site. In this way, I was able to test for differences between the northern and southern halves of the site, as well as testing for differences between the eastern and western halves of the site. Additionally, I looked at each quadrant separately. From each of these quadrants, I sub -sampled a 60 by60 met er section of the site (Figure 6 1). Only data from the shovel tests that lie in these zones are used in comparing the

PAGE 57

57 quadrants of the site. The field sample numbers (FSN) incl uded in ea ch quadrant are shown in Table 6 2. These quadrants are separated from each other by 30 meters along the north-south axis, and by 20 meters along the east -west axis. Testing for Spatial Contemporeity In order to determine whether a spatial ana lysis is appropriate at Bayview I test ed whether the site is single or multi -component in nature. If the site is multi -component, one possible interpretation of differential clustering could be that the site was occupied in different areas at different ti mes. Once the clusters can be determined to be contemporaneous, inferences on social clustering whether from differential identities of residents or differential activity areas, can be drawn from the spatial data. I approached this question from two diffe rent angles, by extracting samples of shell for radiocarbon dating, as well as testing the proportions of early and late Weeden Island ceramic types from opposite areas of the site Radiocarbon Dating Two samples of oyster shell were taken from Bayview, o ne from each of the two column samples. Each of these samples were drawn from the two lowest complete levels of the columns. The purpose of choosing these levels for analysis was to test whether the site was contemporaneous on both sides of the ring. If s o, the site may have been planned according to prescribed locations of specific groups of people, and possibly according to status or activity areas The two samples were sent to Beta Analysis for radiocarbon dating, initially with the intention to use str aight radiometric dating for each. Due to an unknown laboratory mistake that was noted on the results sheet from the laboratory, AMS dates were obtained from the sample from Column Sampl e 2 AMS is a more precise and reliable measure of radiocarbon dating. The original forms sent back from Beta Analysis are attached at the end of this thesis as Appendix G.

PAGE 58

58 The results from this testing show that both samples from Bayview are essentially from th e same time period. T he conventional radiocarbon dates (2 sigm a) range from 11901290 BP for Column Sample 1 and 1060 1140 for Column Sample 2. The samples become closer in age when the calibrated dates are taken into consideration. Calibration is a more accurate way of presenting radiocarbon dates, because fluctuati on of carbon in the earths atmosphere throughout time causes the radiocarbon dating curve to meander from a simple slope. Beta Analytical determined the calibrated dates for the samples taken from Bayview. In this case, the dates from Column sample 1 (2 s igma calibrated result) are 1280 1050 BP and the dates from Column Sample 2 (2 sigma calibrated result) are 1170 980 BP. Proportions of Early and Late Weeden Island Ceramics The distribution of artifacts at Bayview was shown by Russo et al. (2006) to be hi ghly disproportionate between the east and west halves of the site. It is possible to interpret this difference in a number of different ways. It is possible that the east side of the site was formed first, based on the fact that many Weeden Island sites a re arc -shaped instead of full rings (e.g. Hurlb urt Horseshoe [8OK380] Third Gulf Breez e [8SR8], and Horseshoe Bayou [8WL36] [Russo et al. 2006:69]) As more people joined the site, it may have grown from the initial arc shape to a ring. If this were the case, I would expect to find disproportionate amounts of early and late Weeden Island pottery in the different areas of the site. In Chapter 5, I described the different types of ceramics that were found at Bayview, both in my excavation and in Russo et a l.s (2006) report. Weeden Island is broken into two main temporal categories, Early Weeden Island and Late Weeden Island. There are a number of ceramics found at Bayview that range across both of these time periods, but there only a few that are found sol ely within the Early Weeden Island or Late Weeden Island classifications. I looked

PAGE 59

59 at the east and west halves of the site, using the quadrants described earlier, to compare the proportions of early and late types of pottery found at Bayview. The results of this test suggest that the site was most likely formed initially as a circular midden, due to the fact that the re are similar proportions of early and late ceramics are similar in both halves of the site (Table 7 3). These proportions show that overall the site contains more Late Weeden Island pottery (83.13% by weight in the eastern half and 80.1% by weight in the western half) when compared with Early Weeden Island types (16.87% by weight in the eastern half and 19.9% b y weight in the western half). Ho wever, t hese proportions also demonstrate that eastern side of the site is more associated with Late Weeden Island than Early Weeden Island. A chi square test of association between the eastern and western sides of the site offered an even more significant result, x2 = 16.224 with 1 degree of freedom. There is a less than 0.001 chance that this distribution is random, and that there is a statistically greater proportion of late Weeden Island ceramics in the eastern half of the site. Testing for Dual Social Organization through Spatial Distribution of Ceramic Types At Bayview, there are only a few distinct types of decorated pottery from the expansive Weeden Island typology as described by Willey (1949), and even less of these types dominate the assemblage. These types may represent groups of people who have distinctly different regional origins and/or ethnicities, but who were brought together at Bayview as a singular community. Bartlett and McAnany (2000:118) suggest that potters from the Maya Formative pe riod retained certain regional characteristics in order to express the unique identity of their community in the face of reorganization and homogenization of pottery styles. The site we call Bayview is most easily understood as a local c ommunity connecte d to the extra local larger sphere of Weeden Island as a whole.

PAGE 60

60 While it is interesting to speculate on whether the presence of different types of ceramics indicate different groups of people, the common idea of these types has been that they are part of t he larger sphere of Weeden Island pottery. Burial ce ramics are thought to be similar despite great geographic distances (Milanich 1994; Pluckhahn 2003; Sears 1953, among others) while Milani ch (1994), Luer and Almy (1982), and Kohler (1978, 1980) have arg ued that regional variation exists for both elite and utilitarian wares. Even with this knowledge, it is hard to say whether the individuals at Bayview represented one sedentary group with seasonal visitors from far awa y, or whether they lived together in a singular community with two different backgrounds. If different groups of people were living at Bayview, they would most likely occupy different spaces within the ring. By combining the ceramic data with spatial data, it may be possible to identify patte rns throughout the site that would indicate social differentiation Spatial Comparison of Decorated and Plain Ceramics at Bayview Th is first set of tests examines the extent of differentiation between the east and west halves of the site in terms of deco rated and plain pottery. It has been demonstrated by Russo et al. (2006) that there is a disproportionate amount of artifacts between the two halves of the site based on counts and weights, but not the proportion of decorated and plain ceramics. By map pin g the spatial distributions of different types of ceramics throughout the site it is possible to see whether Bayview reflects a difference betwee n areas in terms of elite and utilitarian wares or possibly identifying ritual areas within the site The res ulting map of decorated sherds indicate that there is a much higher distribution of decorated sherds within the eastern half of the site. There are three possible reasons for this; (1) that the distribution is skewed because there are more sh erds on this s ide of the site; (2) that this side of the site was where more of the affluent individuals lived; or (3) that this side of the site could indicate ritual activity. Similar to Weeden Island burial mounds, the eastern half of Bayview contains more

PAGE 61

61 elite vess els, and a greater diversity of decorated vessels The map of the plain sherds, on the other hand, shows a different distributional pattern, that these sherds are evenly distributed across the entire site, in a distinct ring shape. This is in direct contra st with the decorated sherds. Because of this discrepancy, I tested the distribution of plain and decorated sherds by looking at the proportions of the two types, similar to the earlier test with early and late styles of pottery. In order to test the propo rtions of plain and decorated sherds within the two halves of the site, t he counts and weights of different types of ceramics were grouped together into two classes based on surface treatment, plain and decorated. FSN samples were then grouped together int o the spatial quadrant s using the same areas as the previous test. I combined the northwest and southwest quadrants and the northeast and southeast quadrants into one eastern half and one western half, and calculated the proportions for each. The proporti on of decorated to plain sherds in the eastern half of the site is relatively even (49.69% by weight for decorated sherds and 50.29% by weight for plain sherds), while the proportion in the western half of the site is dominated by plain sherds (27.24% by w eight for decorated sherds and 72.66% by weight for plain sherds) (Table 7 4 ). A chi square test of association between the eastern and western sides of the site substantiated these differences with x2 = 11.799 with 1 degree of freedom. There is a less t han 0.001 chance that this distribution is random, and that there is a statistically greater proportion of decorated pottery on the east side of the site when compared to that of the west side. These resul ts will be disc ussed in Chapter 7 Spatial Comparison of Specific Types of Decorated Ceramics at Bayview By looking at the distributions of the two most common types of ceramic sherds at the site, it may be possible to suggest that these clusters of these types of ceramics indicate two different g roups of people living at Bayview at the same time. This is achieved by looking at the proportion of these types in each quadrant, as well as mapping the distributions. Pluckhahn

PAGE 62

62 (2003) had success with a similar approach at Kolomoki, in looking for discre te social groups. At Bayview, the two most common types of ceramics beyond plain wares are Carrabelle types and Wakulla Check Stamped. The resulting maps show that there are distributional differences between the types of ceramics at the site. The most si gnificant difference between the spatial distribution of ceramics types at the site is between the Wakulla Check Stamped and the Carrabelle sherds, which are located at two opposite sides of the ring. Wakulla Check Stamped, although present throughout the site, is more prevalent along the eastern portion of the site, closest to the coast (Figure 7 4). A little further south of the Wakulla Check Stamped sherds, Carrabelle types are more common, in a distinctly different area of the site (Figure 7 5). In addition to mapping the distributions, I also examined the proportions of these two types in order to see how the proportional distribution was similar or different to the data distribution. I nstead of looking at a proportion of one variable to another, I lo oked at the percentage of each type within the entire quadrant. The co unts and weights of each type were calculated within each quadrant and a percentage was derived based on the count and weight of the combination of the count and weight of all the cerami c artifacts within that quadrant. The proportions for Wakulla Check Stamped in each quadrant shows a similar distribution to the mapped distributions The proportion of Wakulla Check Stamped in the northeast quadrant is 67.72% by weight, while the southeas t quadrant contains 20.96% by weight. Conversely, the northwest quadrant contains 18.7% Wakulla Check Stamped and the southwest quadrant contains 9.11% (Table 7 5 ). Carabelle types account for a lower percentage of the entire assemblage, so the proportions are much lower (Table 7 6 ). The southeast quadrant contains 8.79% Carabelle types, and the

PAGE 63

63 northeast quadrant is 1.93%. The southwest quadrant, however, shows a small increase over the northeast, at 2.42%. The lowest amount of Carrabelle was in the northwest quadrant, where only .64% was present. This shows a distribution centered in th e southeastern portion of the site, slightly into the southwestern and northeastern quadrants. Summary In all the tests performed on the spatial data, the east and west halves of Bayview appear to be different. The eastern half has more material in general, and greater amounts of and diversity of decorated and Late Weeden Island pottery. There is also distinct clustering of different types of ceramics, Carrabelle and Wa kulla Check Stamp. While some possible reasons for this have been alluded to, this will be further discussed in Chapter 8.

PAGE 64

64 Figure 6 1. Field Sample Numbe rs used in comparative analyses = SEAC 2006 FSN = Ellison 2007 FSN = selected area for comparison

PAGE 65

65 Figure 6 2 Distribution of Plain Sherds

PAGE 66

66 Figure 6 3 Distribution of Decorated Sherds

PAGE 67

67 Figure 6 4 Distribution of Carrabelle Sherds

PAGE 68

68 Figure 6 5 Distribution of Wakulla Check Stamped Sherds

PAGE 69

69 Table 6 1. Combined Ceramic Datasets Ceramic Type 2007 Count R usso (2006) Count 2007 Weight Russo (2006) Weight Dunlap Fabric Impressed 1 0 4.1 0 Swift Creek Complicated Stamped 51 28 370.1 143.82 St. Andrews Complicated Stamped 0 3 0 14.95 New River Complicated Stamped 0 1 0 19.76 Basin Bayou Incised 0 1 0 2.79 Santa Rosa Punctated 0 3 0 7.19 Crooked River Check Stamped 2 0 8.5 0 Carrabelle Incised 17 19 121 96.54 Carrabelle Punctated 20 30 131.3 263.59 Keith Incised 13 6 39.4 25.74 Weeden Island Plain 59 93 575.9 1001.52 Residual Plain 144 351 642.8 1581.88 Smooth Plain 499 501 2908.8 2606.14 Weeden Island Appliqued 1 0 14.7 0 Weeden Island Decorated 1 0 1.8 0 Weeden Island Incised (E/L) 21 16 104.1 118.98 Weeden Island Punctated (E/L) 16 15 73.9 67.24 Weeden Island Red (E/L) 9 11 45.2 92.27 Weeden Island Zoned Red 0 1 0 10.07 Mercier Red on Buff (L) 1 0 45.2 0 Weeden Island (Black Paint) (E/L) 0 1 0 7.3 St Petersburg Incised 0 1 0 5.27 Indian Pass Incised (L) 1 0 4.3 0 Ruskin Linear Punctated (L) 3 0 25 0 Tucker Ridge Pinched (L) 1 1 5.9 7.71 Wakulla Check Stamped (L) 276 268 2160.3 1942 Total 1136 1350 7282.3 8014.76 Correlation 0.944582646 0.947131613

PAGE 70

70 Table 6 2. Field Sample Numbers in Each Quadrant Field Sample Numbers Quadr ant Russo et al. (2006) 2007 Northwest 63, 64, 65, 73, 74, 80, 83 11, 62, 63, 64, 66, 67 Northeast 67, 69, 75, 77, 82, 85, 86 12, 40, 42, 44, 76, 78 Southwest 26, 27, 29, 35, 43, 44, 45 1, 2, 3, 7, 8, 10, 15, 17, 18, 21 Southeast 31, 32, 33, 37, 38, 39, 48, 49 9, 28, 31, 32, 33, 36, 37, 38 Table 6 3. Proportions of Early and Late Ceramics between East and West Sides of Site Proportion Early Late Early and Late Early Late Weight Count Weight Count Weight Count Weight Count Weight Count East 232.4 25 1145.3 130 1377.7 155 16.87 19.23 83.13 83.87 West 48.68 14 195.92 14 244.6 28 19.9 50 80.1 50 Table 6 4. Proportions of Plain and Decorated Ceramics between East and West Sides of Site Proportion Decorated Plain Decorated + Plain Dec orated Plain Weight Count Weight Count Weight Count Weight Count Weight Count East 1780.73 277 1802.59 312 3583.32 589 49.69 41.43 50.29 52.97 West 396.51 94 1053.8 178 1450.31 272 27.24 34.56 72.66 65.44 Table 6 5. Proportions of Wakulla Check Stam ped (grams) Russo et. al (2006) 2007 Total Proportion Wakulla Total Wakulla Total Wakulla Total Wakulla East 546.39 1881.12 546 1757.1 1092.39 2905.3 37.6 West 109.92 642.68 81.7 836.8 191.62 1479.48 12.95 Tab le 6 6 Proportions of Carabelle Types (grams) Russo et. al (2006) 2007 Total Proportion Carrabelle Total Carrabelle Total Carrabelle Total Carrabelle Northwest 3.78 363.85 0 228.3 3.78 592.15 0.64 Northeast 6.68 1001.2 13 752.6 19.68 1020.9 1.93 Southwest 16.94 278.83 4.5 608.5 21.44 88 7.33 2.42 Southeast 94.97 879.89 70.6 1005 165.57 1884.4 8.79

PAGE 71

71 CHAPTER 7 FAUNAL ANALYSIS Introduction Faunal analysis is useful in interpreting aspects of social complexity such as ethnic and status differences, by looking at the remains of animal spec ies found within an archaeological assemblage. Many faunal studies are associated with determining higher vs. lower status individuals or groups of people, economic patterns, and technology through faunal remains (Dietler 2001; Emery 2003; Ervynck et al 2003; Hayden 2001; Kirch and ODay 2003; Scott 2001, among others). In this chapter, I compare faunal remains recovered from two column samples taken on opposite sides of the site, the east and west sides. I am concerned with discerning what difference, if a ny, the faunal remains of the two column samples reveal because my ceramic analysis, and Russos (2006) analysis of total artifact densities, both revealed differences in concentrations of ceramic sherds from the eastern and western areas of the site. This chapter is divided into four sections, the theoretical background for using zooarchaeology to interpret social organization, the zooarchaeological recovery and analysis methods, results of the identifications and quantifications, and data analysis of thes e quantifications. The sample size used for this study is extremely small, and therefore minimal interpretation is appropriate. However, the results show some signs of duality in the faunal remains at Bayview that call for further testing by future researc hers. Interpreting Sociality through Faunal Remains Connecting social structure and organization to the use of animals in a given community is familiar to zooarchaeological studies, and many different approaches can be undertaken with the same dataset. Se veral potential indicators of social organization and social activities may be found within the faunal remains at Bayview. This chapter addresses five ways of examining the

PAGE 72

72 data for social meaning. First, the seasonality of the samples will be examined. Th e ceramic assemblage and radiocarbon dates point to the possibility that the site was occupied year -round. However, if the two column samples contain differing seasonal species, it may be that the community size and structure, or possibly the use of differ ent areas of the site, fluctuated throughout the year. Determining the season of capture for particular species is not always possible, however, there are certain times of the year when some species are most likely to be captured for food. Second, the ha bitats for the species found in the columns will be considered. The habitats of the various species can be used to understand whether particular shellfish collection habitats were utilized while others were not. If the two sides of the site contain species from different habitats, it may be that different community members were using different hunting/fishing grounds. Or it might be that there were possible social ties to freshwater vs. saltwater fishing areas. Third, the ritual use of animals will be disc ussed, based on the findings of non-meaty portions of turkey vulture ( Cathartes aura) and white tailed deer ( Odocoileus virginianus ). Non food use of animals at Bayview might offer insight towards understanding the community beyond economic strategies. The differential distribution of ritually important species or body portions as opposed to those with more quotidian purposes may point to the use of different areas of the site for different types of consumption activities, such as day to day sustenance vs. feasting and/or ritual activity. Each of these approaches looks at the assemblage in a different way, in order to arrive at more finely tuned questions pertaining to the use of animal species by the community at Bayview.

PAGE 73

73 Lastly, the relative density of the faunal material will be addressed. In cases of feasting, rapid accumulation of food refuse can result in a high density of animal remains, while daily sustenance may result in lower density of animal remains, mixed in with artifacts and soil matrix. If t he eastern side of the site was more intensively used than the western side, the columns may reveal that the eastern density of faunal remains per level is higher than in the western side. However, if the two columns are similar, it may indicate that fauna l use at the site is less related to social differentiation than what the ceramic analysis has revealed. Laboratory Methods One of the most common problems in any interpretation of zooarchaeological remains is the presumed completeness of the sample from the site. As archaeological methods have developed throughout time, emphasis has shifted towards making the researcher responsible for collecting all possible material when testing. The practice of fine -screening samples to recover more complete zooarchaeo logical assemblages is now common practice (James 1997, Wing and Quitmyer 1985), and also allows the researcher to test for differences between the screen sizes. Screening techniques are of specific interest to coastal sites, where nearly 40% of material is lost through mesh (Russo 2006:57). This bias causes problems for zooarchaeologists attempting to determine the importance of small boned species to the diet of coastal peoples. As described earlier, I excavated the west side column sample in 10 cm levels to sterile soil, which was 60 cm below surface, removed all materials, including soil matrix, from the field, and screened each level in the lab with nested box screens of 1/4, 1/8 and 1/16 inch mesh. These mesh sizes were chosen to conform with cur rently accepted practice, and 1/16 inch mesh is the smallest size commonly used in archaeological sampling (see Wing and Quitmyer 1985). Since time precluded a full analysis, I chose to identify and analyze the faunal assemblages only from levels two (10 2 0 cm below surface) and four (30 40 cm below surface). These two levels were

PAGE 74

74 chosen because the column sample contained 5 levels, but the top 10 cm were disturbed and the excavation reached sterile soil at 55 cmbs. I wanted to compare the shallowest and deepest intact and complete levels, in order to test for changes over time. The east side column sample was excavated by Russo (2006). I identified and analyzed the 1/8 and 1/16 inch mesh assemblage recovered from level four (20 30 cmbs) of Russos excavat ed sample. The inch assemblage from this level was previously analyzed and is published as part of the original report (Russo, et al. 2004). In the interest of time, for the purposes of this thesis, I chose to use the faunal data as an ancillary test of duality between both halves of the site. The comparison of two levels from column sample two may give insight towards change over time at the site, while the two columns may offer information on practices on opposite sides of the site. If this thesis were solely based on zooarchaeological analysis, it would be more prudent to compare each column in its entirety, but as an ancillary dataset, comparing subsamples from each column is adequate. Faunal analysis for each level was conducted following standard pro tocols (e.g. Reitz and Wing 1999). First, all elements were rough sorted to class, when possible. All class -sorted elements were then sorted into identifiable elements and non identifiable and non-diagnostic elements. Specimens were identified by compariso n with the Florida Museum of Natural History Environmental Archaeology collections. All identified elements were weighed and counted, according to the corresponding level of taxonomic identification. Data was entered into an Excel spreadsheet for analysis. Specimens were organized by level number and according to taxonomic level and element. NISP (number of identified specimens present) and MNI (minimum number of individuals) were derived from the raw data. NISP is the initial quantification in faunal analy sis, a straightforward account of specimens

PAGE 75

75 present in each taxonomic category. MNI is a secondary quantification that is based on certain assumptions and is not, therefore, as directly tied to the specimens. MNI is a derived count of the greatest number o f unpaired elements (elements from one side). This value represents the least possible number of individuals present in the material. The NISP and MNI counts, working together, are designed to minimize bias in quantifying and comparing the zooarchaeologica l remains. NISP counts, when used alone, can be heavily biased towards species with more elements, more identifiable elements, and elements that are more easily preserved post depositionally (see Reitz and Wing 2004; Grayson 1973, 1984). MNI can be problem atic as well, in that it is a derived value from NISP counts. The accuracy of MNI is directly related to the accuracy of NISP for any given sample, but corrects for biases of differential element counts or preservability. MNI is more heavily biased by samp le size, and analytical methods (see Reitz and Wing 2004; Grayson 1973, 1984). For example, the column samples at Bayview were excavated in arbitrary, non -culturally distinguished 10 centimeter levels, potentially separating given individuals into more tha n one level, therefore biasing the MNI count for that species throughout the column. As well, MNI is biased in small samples by the relative proportions of rare specimens (i.e. a single specimen of a rare species will be given equal weight as a group of ma ny non -overlapping specimens of a more common species). The samples taken for the purpose of this thesis are not large enough to compensate for this problem, so this bias must be considered in these analyses. In this study, NISP and MNI counts for all species are derived from identifiable elements only. NISP values are given for each level and screen size in the sample, while MNI values are derived for each level by combining screen sizes for the given level by assuming that individuals contributing to each fraction are the same. MNI counts pertaining to bivalve mollusks are based

PAGE 76

76 on the presence of umbos. Elements not containing umbos (i.e. crushed, broken shell) were categorized as unidentified shell, unless identification could be made due to distinct ive shell surface, such as with Chione cancellata. These elements are included in the NISP value for the species, but are not used to derive MNI. Unidentifiable shell and bone fragments were weighed, but not counted, due to the level of breakage and fragil ity of the specimens. It is unlikely that counts for unidentifiable elements would be accurate, and weights are similar whether elements are broken or whole. Preliminary Results of Faunal Identification and Quantification This section describes the initi al quantification of all the data per level and per fraction. In order to be able to combine levels or compare columns, the initial step is to describe each fraction and level as a discrete unit. The faunal material is presented in its entirety within Appe ndices D, E and F. While this section does not analyze the faunal data, it presents a summary of the contents of each level analyzed per fraction size. Column Sample 1, Level 4 Zooarchaeological material from the 1/4 screen of column sample 1 was analyze d and presented as part of the official report of the excavation (Russo et al. 2006). Oyster ( Crassotrea virginic a) and fish heavily dominate the assemblage, with relatively little mammal, reptile and bird remains rounding out the assemblage. I have included the results from this faunal analysis at the end of this thesis in Appendix D. The 1/8 and 1/16 screened samples from this level were minimally analyzed in the previous report, with no identifiable material indicated. In order to obtain a similar data set to compare with the material collected and screened from column sample 2, I analyzed the 1/8 and 1/16 assemblages in their entirety, with the expectation of finding more identifiable material than previously reported. However, without being able to e xamine the previously analyzed material from the fraction, combining MNI counts was not possible.

PAGE 77

77 Therefore, the combined dataset from all fraction sizes is slightly problematic, and may cause bias when considering the MNI from this level. Because of the differences in these two analyses, I will address the and 1/8 fractions separately, but when considering level two and level four from column sample two, I combine the fractions and address their contents as a whole. The most commonly encountered id entifiable genus from the 1/8 sample was Polygyra (NISP/MNI=79, .84 grams). Polygyra a common genus of land snail, is most likely commensal and not used for food. Another genus present in this sample which was not likely used for food is Crepidula (NISP/ MNI=15, .35 grams), or slipper shell, which is mostly known for its parasitic livelihood connected to either mussels or oysters. The only taxa identifiable to species level within the 1/8 and 1/16 inch screened samples from this level is Argopecten irradians (NISP=5, .66 grams). These are not complete specimens, but rather partial valves with intact umbos allowing for identification. In addition to shellfish, two types of fish were encountered, three elements belonging to the order Siluriformes (catfish) an d two to the genus Cynoscion (sea trout). Beyond this, the rest of the 1/8 screened assemblage is comprised of fragmented shell (869.43 grams) and unidentifiable fish bone (NISP=3341, 48.4 grams). The 1/16 screened material from column sample 1 is even less informative. There are no specimens from this assemblage which can be pinpointed to species. The sample is entirely comprised of fragmented bone and shell, and weighs a total of 479.76 grams. Column Sample 2 Column sample 2 contains identifiable mate rial from all screen sizes. I have included the results from this faunal analysis at the end of this thesis, separated into screen sizes in Appendix E. This column sample is significantly different from column sample 1 in terms of overall number of specime ns, species, and horizontal area. The volume from each level is therefore one quarter that of column sample one. The results from each level are described here.

PAGE 78

78 Level 2 As with column sample 1, both vertebrates and invertebrates are present and the datase t for this level can be found at the end of this thesis in Appendix C. The most common fish in level 2 is mullet ( Mugil sp.) (NISP = 34, MNI =2, 3.8 grams). Bony fishes are the dominant vertebrate species in the assemblage and are represented by many diffe rent species: Atlantic croaker ( Micropogonias undulatus ), Spotted seatrout ( Cynoscion nebulosus ), Ladyfish ( Elops saurus ), hardhead catfish ( Arius felis ), Oyster toadfish ( Opsanus tau), mullets (Mugil sp), jacks (Caranx sp), snooks ( Centropomus sp), weakfi sh ( Cynoscion) and flounders ( Paralichthyes ). Although identifiable fish elements represent only 4.43% of the total NISP and .97% of the total weight from this level, fish represent 97.8% of the vertebrate assemblage in terms of NISP and 35.68% in terms of weight. Mammals are the second highest represented class found in the assemblage, and include white -tailed deer ( Odocoileus virginianus ) and deer mouse ( Peromyscus sp.). Neither of these species were likely used for food in this case, considering that the deer element was a shed antler, and the elements from the mouse were found articulated in situ, indicating that it was most likely a commensal species. One bird rib was present and identifiable only to class (Aves). Four turtle elements were identifiable to the genus Kinosternidae, but were not identifiable to species. The invertebrate assemblage contains both bivalve and gastropod species, and the entire sample is heavily dominated by oyster ( Crassotrea virginica ) (NISP = 149, MNI = 82, 1274.26 grams). T his translates to oyster comprising 57.52% of the identifiable NISP from the sample, and to 50.38% of the total weight. Invertebrates in general comprised 93.42% of the total weight of the sample.

PAGE 79

79 Level 4 Level 4 is very similar to level 2 from column sam ple 2. Both vertebrates and invertebrates are present, and again the invertebrates dominate the sample, both in weight and NISP. The results from level 4 are located at the end of this thesis in Appendix F. The dominant vertebrate species in the assemblage are bony fishes, and are represented by several different species: Blue runner ( Caranx crysos ), Spotted weakfish ( Cynoscion nebulosus ), Ladyfish ( Elops saurus ), Flathead mullet ( Mugil cephalus ), mullets ( Mugil sp), jacks ( Caranx sp), and weakfish (Cynosci on sp) Identifiable fish elements from this level represent 1.64% of the total NISP and 1.11% of the total weight from this level, fish represent 99.62% of the vertebrate assemblage in terms of NISP and 88.31% in terms of weight. Birds are the second highest represented vertebrate class found in the assemblage by weight, represented by a single species and element, a turkey vulture ( Cathartes aura) ulna Two turtle elements were identifiable, one to the genus Kinosternidae, and one to Florida softshell tur tle ( Apalone ferox ). Mammals are also present within this level, one Eastern cottontail rabbit ( Sylvilagus floridanus ) and an unidentifiable Rodentia specimen. The invertebrate assemblage from level 4 contains both bivalve and gastropod species, and is again heavily dominated by oyster ( Crassotrea virginica ) (NISP = 188, MNI = 122, 1406.09 grams). This translates to oyster comprising 65.05% of the identifiable NISP from the sample, and to 65.13% of the total weight. Invertebrates in general comprised 94.86% of the total weight of the sample. Data Analysis of Faunal Remains The data analysis takes the initial quantifications a step further, and attempts to address some of the more social questions laid out at the beginning of this chapter. While the sample sizes are very small, it is important nonetheless to evaluate the material at hand, in order to

PAGE 80

80 provide starting points for future research. While the same datasets are used to address each question, the data are manipulated in different ways. Each way th e data are utilized is explained in detail within each subsection. Season of Occupation The first question I posed at the beginning of this chapter was whether species selection at Bayview was based on seasonal rounds with people exploiting different par ts of the coastal environment throughout the year, or whether species at Bayview indicated a year round occupation. From Russo et al.s original (2006) data, it appears as though Bayview was occupied in the summer and fall, due to the presence of both large bay scallops ( Argopecten irradians ) which are in season from July to November (Russo and Quitmyer 1996:217), and a large amount of mullet ( Mugil sp.) which usually run during the late summer and early fall (Purdy 1991:114). While bay scallops are found w ithin the two column samples, they are highly fragmented and therefore not easily quantifiable. Mullet, on the other hand, is present in varying amounts between the two columns, and may indicate that there is a difference in season of occupation or use bet ween the two sides of the site. In order to compare the mullet remains between the two samples, I calculated the relative frequency of this species within each level of the two column samples I analyzed. I compared proportions of mullet within each of t he levels to all species and also to the fish component. The two column samples vary in volume, and relative frequency studies examine the percentage of particular elements within a given population. In level four o f the column excavated by Russo et al. (2 006), mullet comprised a significantly larger portion of the identifiable fish elements in terms of NISP, MNI and weight than in either of the levels of the column sample I extracted (Table 7 3). While relative frequency of mullet from the total sample for each level is similar, the relative frequency of mullet as a proportion of all fish specimens is strikingly different between

PAGE 81

81 the two sides of the site. The sample size is extremely small, but it indicates that while the proportion of mullet is consistent in comparison to all species, mullet is more common in proportion to all fish in the eastern column sample than it is in the western column sample. Between the two levels of the column sample that I excavated, there are also differences in proportions of mullet remains, although they are more challenging to read than between the two columns. The low MNI counts for level 2 (n = 2) and level 4 (n = 1) are not a large enough sample size to make for a solid interpretation of the material, because it would b e misleading to state that level 2 has twice as many mullet as level 4. When comparing MNI of mullet between the two levels, it appears that mullet comprises more of the sample in level four than in level two. However, in terms of weight and NISP, the freq uency of mullet in level two is less than half that of level four (Table 7 3). These results most likely stem from the extremely low sample size, but may be viewed as a way to ask more developed questions in future studies on seasonality at Weeden Island s ites. Habitat Utilization The second question I had concerning the faunal sample considers whether the accumulation of animal species within each column indicates the exploitation of different habitats in the general vicinity of the site. While Bayview is generally considered a coastal site, there are both fresh and saltwater resources, as well as woodland habitats, directly adjacent to the site. Literature on each species found within the two columns was investigated to determine which habitat the specie s most often occupy (Table 7 4 includes both habitat designations and information sources). Four main aquatic habitats were identified within these descriptions: freshwater lakes and streams, shallow sandy or muddy bottom marine habitat, shallow tidal mars hes, and shallow rocky marine habitat. Woodland species are also present, but the particular species within the columns cross into freshwater lakes and streams, or at least the terrestrial

PAGE 82

82 space bordering these water bodies. Because of the versatile nature of these species, they are not appropriate markers of specific habitat use by the community at Bayview. The species that are good indicators of habitat were grouped into the three aquatic habitats, and relative frequencies for the types of species found i n these habitats were calculated. This initial investigation is not meant as a test of relative abundance of those species, but rather an indication of the habitats used at both sides of the site (Table 7 5). In order to address this question, the two le vels of column sample two were combined. This was done primarily because both levels contain species that occupy similar habitats. Although some of the species vary per class, the ecological niches are very similar. For example, in level two there are whit e tailed deer ( Odocoileus virginianus ) present, while in level four there are eastern cottontail rabbits ( Sylvilagus floridanus ). While these animals are vastly different for many reasons, they both represent woodland habitat. Neither are good markers of habitat, either, because of their ubiquitous nature in the Southeast. In contrast, differential shellfish can indicate use of specific habitats, such as oyster beds or sandy shores. However, both levels of the column show species from each of these habitats It is because of these similarities that the two levels were combined, and therefore created a larger sample size to compare with Russos (2006) level four from column one. The percentages of species from marine marsh habitats and sandy bottom marine habitats are very similar in both columns. However, the percentage of species present from rocky bottom marine habitat is much greater in column sample 1 than in column sample 2. In contrast, the percentage of species from freshwater habitats is greater in column sample 2 than in column sample 1.

PAGE 83

83 While relative frequency of particular species within a given sample is important and useful information, it is also useful to see what percentage of the assemblage is from each habitat (Table 7 6). Although reco gnizably biased by the predominant shellfish, the data reveal that the most frequently represented habitat is shallow rocky bottomed marine habitat. This is not surprising considering the amount of oyster ( Crassotrea virginica ) in both columns. The most in teresting pattern that emerges when comparing both columns through habitat utilization is not in the specific comparisons between each habitat, but the overall difference in diversity in habitat utilization. In column 2, almost all of the material is from rocky bottom marine habitat (97.69% by weight, 80.91% by MNI count, 72.57% by NISP). While this is predominantly oyster species, other species are included in this habitat as well. In column 1, on the other hand, rocky bottom marine habitat still dominates the assemblage (87.48% by weight, 69.98% by MNI, and 82.16% by NISP), but sandy bottom marine habitat comes in a closer second (12.01% by weight, 22.15% by MNI, and 15.59% by NISP count) than in column 2 (1.42% by weight, 13.17% by MNI and 11.5% by NISP). Utilization of more diverse habitat may indicate that the eastern inhabitants of the site may have had a broader access to resources than the inhabitants on the western side. Ritual Use of Animals The third question I had concerning the faunal material was the presence of remains from one non -dietary species (a turkey vulture) and a non -dietary portion of another species (a shed deer antler). Interpretational problems can arise in a zooarchaeological interpretation limited to dietary considerations sinc e many species and elements were used for tools, ceremonial ite ms, and the like (Ervynck 2003: 433434). In the west -side column sample, I identified an ulna from a turkey vulture ( Cathartes aura) and a shed antler from a deer ( Odocoileus virginianus ). Thes e specimens probably have less to do with consumption than use as either tools or ritual objects. Weeden Island ceramics often portray bird symbolism, and mortuary vessels have been found

PAGE 84

84 with turkey vultures portrayed as effigies (Percy and Brose 1978). T he turkey vulture, as a carrion animal, carries cross -cultural taboos against eating, and the presence of a non-meaty element such as an ulna indicates the possible use of the birds wing for ritual practice, especially since this species is rarely, if eve r, encountered in food refuse contexts. Deer antlers were also widely used in ritual contexts, as parts of ritual headdresses in ethnographic examples from the southeast (Power 2004:186) but more frequently deer antler was used as a tool for pressure flak ing lithic tools. I lean toward the ritual interpretation because there are very few stone tools found at coastal Weeden Island sites, and only one quartzite point was recovered during my excavation at Bayview. The contextual data for these interpretations are limited to the column sample in which they were discovered, so it is not possible to make a solid interpretation. Future investigations might consider the possibility that the two sides of the site differentially accumulated food vs non-food refuse, a nd that these might be linked to spatially discrete activities at the site. Relative Density Relative density of the columns may also have bearing on the establishment and use of ritual space at Bayview. If the deposit on the eastern side is more dense, this may indicate that the area was used primarily for depositing food refuse, while the western half may indicate habitation or even ritual use, especially when considering the two elements described above. While each of these columns represents a minute fraction of the entire site, they may offer insight towards the differential densities between the two halves of the site. The two column samples were of different sizes, so their quantities cannot be directly compared. However, relative density of remains can be calculated based on the cubic measures of each. The relative density of animal remains from each level reveals that in terms of NISP, level four of column sample one is much more dense per cubic centimeter. However in terms of

PAGE 85

85 weight and MNI, it i s very similar to the other two levels and the other column (Table 7.6). The method of analysis of the fraction from level four of column sample one is directly linked to this seemingly inconsistent result. The NISP counts and weight from column sample one were derived from counting subsamples of the crushed shell and bone and multiplying it by the percentage that class represented in the total assemblage (see Russo 2006:58 for detailed explanation). In my analysis I did not calculate NISP for unidentifi ed shell and bone, and therefore have a much lower count. However, weight was calculated for each of the levels in column sample two. Considering the relative density of weight all three levels are very similar. This indicates that the rate of deposition within each level of the two columns is very similar. This is contrary to the ceramic analysi s, which shows that the eastern side of the site was more intensively used. This is a significant departure from the ceramic results, and may indicate that social differences between individuals occupying the site are more related to the distribution of ceramic artifacts and not as much to faunal detritus. This is also visible in the distribution of shell refuse at the site (Figure 7 1). The distribution of shell f ound in the shovel tests from the 2007 excavation show a relatively even distribution throughout the site, with the exception of one very deep and dense deposit on the eastern side of the site. It may be that the use of animals for food was less socially o rdered than the use of certain types of ceramics. Summary In general, there is not much difference between the two column samples as far as the most commonly exploited species, or the ratio of invertebrate to vertebrate species. In both samples, the bulk of the material is comprised of shellfish. However, it is in the details where there are subtle differences between the two column samples. These differences are quite possibly tied to the small sample size available for this analysis, yet still may be indicators of

PAGE 86

86 differences in economic strategy and in social organization at the site. Seasonality is still a puzzling factor at Bayview. Although it appears as though there are slight differences in the amount of mullet between the two sides of the site, th is species represents a very slight fraction of the total faunal assemblage. While scallops are present, these also represent a very small percentage of the assemblage. The presence of these seasonal markers indicate that the site was most likely occupied during the late summer and early fall, however without reliable seasonal markers for spring and winter, it is not out of the question that the site was occupied during these seasons as well. As far as habitat is concerned, it appears that the rocky bottome d marine habitat is the most frequently used area by both sides of the site, however the east side of the site has more diversity overall when it comes to habitat utilization. Using a greater amount of habitats may be linked to seasonality, because certain habitats are more appropriate during different parts of the year. This might point to a greater degree of sedentary use or more sedentary social groups on the eastern side of the site than the western. Or it might point to the possibility that the people on the eastern side of the site had preferential access to a wider diversity of resources, which correlates well with the data from the ceramic analysis. The recovery of non -meaty portions of animals associated with ritual behavior from the western column sample may indicate that this was a specialized area for other activities not directly related to diet, however it is impossible to distinguish diet from ritual, and diet from tool making when considering animal remains. This is why comparing the two colum ns in terms of any single activity type is entirely inappropriate. There are many potential ways of interpreting the datasets presented within this thesis, and the greatest setback to interpreting the faunal remains is the sample size. However, it is not t he goal of this thesis to answer all the greater questions about Weeden Island culture, but to further the field of discovery and research by offering new ways of thinking about the data.

PAGE 87

87 Figure 7 1. Distribution of Shell Weight from shovel tests (2007 e xcavation)

PAGE 88

88 Table 7 1 Summary of Faunal Material by Class Column Sample 1, Level 4 NISP %NISP MNI %MNI Weight (g) %Weight Mammalia 2 0.01% 1 0.16% 10.6 0.10% Aves 1 0.01% 1 0.16% 0.1 0.00% Reptilia 24 0.14% 2 0.32% 7 0.06% Osteichthyes* 1875 10 .97% 44 7.05% 101.14 0.93% Crustacea 10 0.06% 6 0.96% 2.3 0.02% Mollusca* 15185 88.82% 570 91.35% 10779.85 98.89% Column Sample 2, Level 2 NISP %NISP MNI %MNI Weight (g) %Weight Mammalia 2 0.13% 2 1.05% 0.298 0.01% Aves 1 0.06% 1 0.53% 2.55 0.12 % Reptilia 2 0.13% 2 1.05% 0.336 0.02% Osteichthyes 1321 83.40% 6 3.16% 39.032 1.87% Mollusca 258 16.29% 179 94.21% 2047.089 97.98% Column Sample 2, Level 4 NISP %NISP MNI %MNI Weight (g) %Weight Mammalia 23 1.54% 2 1.49% 19.75 0.78% Aves 1 0.07% 1 0.75% too small too small Reptilia 5 0.34% 2 1.49% 0.09 0.00% Osteichthyes 1280 85.96% 16 11.94% 24.718 0.98% Mollusca 180 12.09% 113 84.33% 2362.766 93.42% Column Sample 1, Level 4 contains estimated counts and weights, explained in Russo (2006) Table 7 2 Relative frequency of Mullet (Mugil sp.) per Column Sample Level NISP Mugil sp. MNI Mugil sp. Wt Mugil sp. % of total Osteichthyes NISP % of total Osteichthyes MNI % of total Osteichthyes weight %NISP Total %MNI Total %Wt Total CS 1 Lev el 4 256 10 29.9 13.65% 22.73% 29.56% 1.50% 1.60% 0.26% CS 2 Level 2 34 2 3.8 2.66% 12.50% 15.37% 2.28% 1.49% 0.15% CS 2 Level 4 12 1 2.53 0.91% 16.67% 6.48% 0.76% 0.53% 0.12%

PAGE 89

89 Table 7 3. Habitat Information on Species from Columns 1 and 2 Species Habi tat Odocoileus virginianus woodland areas Peromyscus sp. woodland areas and open grasslands Sylvilagus floridanus woodland areas and open grasslands Cathartes aura woodland areas and open grasslands Terrepene carolina open woodlands (World Biodiversi ty Database 2008) Apalone ferox freshwater, brackish tidal streams, muddy bottoms (Coultas and Hsieh 1997) Kinosternidae freshwater, brackish tidal streams, muddy bottoms (Coultas and Hsieh 1997) Elops saurus brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Alosa mediocris brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Clupeidae brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Siluriformes freshwater (fishbase. org 2008) Ariidae shallow water, sandy bottom (Abbott and Morris 2001) Arius felis shallow water, sandy bottom (Abbott and Morris 2001) Caranx sp. brackish, marine (fishbase.org 2008) Archosargus probatocephalus bays and estuaries, oyster beds (fishb ase.org 2008) Cynoscion nebulosus brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Menticirrhus littoralis brackish, marine, (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Micropogonis undulatus brackish, marin e (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Pogonias cromis brackish, marine, (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Mugil sp. brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Paralic hthyes sp. brackish, marine (fishbase.org 2008), tidal marshes (Coultas and Hsieh 1997) Micropterus salmoides freshwater, muddy bottoms (fishbase.org 2008) Opsanus sp. marine (fishbase.org 2008), oyster beds (Smithsonian Marine Station at Fort Pierce 20 08) Mennipes mercenaria shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Station at Fort Pierce 2008) Cirripedia any marine habitat (Smithsonian Marine Station at Fort Pierce 2008) Busycon contrarium estuaries, oyster reefs an d tidal creeks (Smithsonian Marine Station at Fort Pierce 2008) Crepidula sp. Rocky bottom shallow marine (Smithsonian Marine Station at Fort Pierce 2008) Melongena corona shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Stat ion at Fort Pierce 2008) Fasciolaria tulipa shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Station at Fort Pierce 2008) Murex sp. shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Station at Fort Pie rce 2008) Neverita duplicata shallow marine sandy bottom (Smithsonian Marine Station at Fort Pierce 2008) Urosalpinx tampaensis shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Station at Fort Pierce 2008)

PAGE 90

90 Table 7 3 Continue d. Species Habitat Urosalpinx perrugata shallow marine, usually seagrass beds and oyster reefs (Smithsonian Marine Station at Fort Pierce 2008) Littorina irrorata brackish water marshes at or above water level on grasses (Animal Diversity Web 2008) M odulus modulus shallow seagrass marshes (Smithsonian Marine Station at Fort Pierce 2008) Euglandina rosea ground dwelling snail found throughout the southeastern US (Animal Diversity Web 2008) Chione cancellata shallow marine, sandy bottom dweller, com monly found in seagrass beds (Abbott and Morris 2001) Macrocallista nimbosa shallow sandy or muddy marine habitat (Abbott and Morris 2001) Ostrea equestris shallow water, rocky bottom (Abbott and Morris 2001) Trachycardium egmontianum shallow marine, sandy bottom dweller (Abbott and Morris 2001) Spisula solidissima shallow marine, sandy bottom dweller (Abbott and Morris 2001) Lopha frons mangrove forests attached to roots (Abbott and Morris 2001) Argopecten irradians shallow water (Abbott and Morri s 2001) Crassotrea virginica shallow water, rocky bottom (Abbott and Morris 2001) Table 7 4. Number of Species Present by Habitat Habitat # of Species in CS 1 # of Species in CS 2 %Species in CS 1 %Species in CS 1 sandy bottom marine 9 6 22.50% 26.09% rocky bottom marine 12 4 30.00% 17.39% marine marsh 12 6 27.50% 26.09% Freshwater 3 3 7.50% 13.04% Table 7 5. Relative Frequency of NISP, MNI and Weight of Species by Habitat Column Sample 1 NISP MNI Wt (g) %NISP %MNI %Wt Sandy bottom mari ne 2407 107 1193.97 15.59% 22.15% 12.01% Rocky bottom marine 12682 338 8694.38 82.16% 69.98% 87.48% Marine marsh 315 32 45.41 2.04% 6.63% 0.46% Freshwater 32 6 4.85 0.21% 1.24% 0.05% Column Sample 2 NISP MNI Wt (g) %NISP %MNI %Wt Sandy bottom marine 65 49 40.11 11.50% 13.17% 1.42% Rocky bottom marine 410 301 2766.678 72.57% 80.91% 97.69% Marine marsh 79 17 24.772 13.98% 4.57% 0.87% Freshwater 11 5 0.566 1.95% 1.34% 0.02%

PAGE 91

91 Table 7.6 Relative Density of Column Sample Levels NISP MNI Wt ( g) NISP/cm 3 MNI/cm 3 Wt(g)/cm 3 CS 1 Level 4 17107 624 11382.05 0.68 0.01 0.46 CS 2 Level 2 1489 134 2529.15 0.24 0.02 0.4 CS 2 Level 4 1584 190 2090.02 0.25 0.03 0.33 Note: for density measures, CS 1 = 50x50x10cm3, CS 2 = 25x25x10cm3

PAGE 92

92 CHAPTER 8 DISCUSSION Introduction By referring back to the data obtained duri ng this investigation, I will address the initial questions laid out in the introduction to this thesis. The central question is whether the distribution of artifacts at Bayview shows infer the social organization of a dual society, similar to what Lvi -Strauss describes ethnographically (1963) and how that might be visible in the archaeological record, using the ideas put forth by Means (2007). More specifically, in terms of the methods of test ing done at Bayview, I will look at whether there are distinct types of ceramics found throughout the site that could infer that there are multiple groups of people living together in one community In order to do this, there needs to be some evidence that people were occupying different areas of the site simultaneously. I will also consider whether the spatial organization of the site reflects a social structure that recognizes distinct social groups, and whether social inequality is visible through the di fferential distribution of faunal remains and possibly artifacts. Inferring Social Structure through Spatial Analysis Spatial patterning and variation within an intra -site context can indicate different possible interpretations pertaining to social structure It could mean the introduction of new ideas into an area, carried by a few different people at the site, or it could represent different family groups with different material culture. It could also represent different occupational periods at the site. Since the material from both types fits within the ring pattern of artifacts found in both the plain sherds and the Weeden Island series, I am inclined to think that this spatial pattern infers that there is a level of social difference between the people living in the northeast portion of the site and those living in the southeast. This could stem from a number of possible factors, including

PAGE 93

93 access to coastal resources by more elite families, or the exact opposite, that the more elites were further away f rom the coast so that those on the opposite side of the village were more involved with food preparation and disposal. Means (2007:94) identifies that spatially discrete clusters of ceramic attributes also could correlate to discrete social groups, includ ing dual social divisions like moieties or corporate social groups along the lines of lineages, clans or Levi -Straussian houses. Alternatively, the argument could be made that those living near the coast hosted new people for a short amount of time that s et up camp on the other side of the village. Considerably less material is found on the western side of the site than the eastern side. This leads to a number of interpretations, each of which are equally potential and could also be complementary. If cert ain individuals hold a greater amount of prestige, it could be that certain Weeden Island cultural beliefs prescribe that the eastern portion of the site holds greater power Weeden Island burial mounds are known for their east -facing caches of highly ritualized ceramic items, and most ritual activities also take place on the east side of the mounds. It is possible that the village is a reflection of the burial mound, or perhaps even a fractal representation of the world view of the Weeden Island people sim ilar to that described by Tambiah (1985) in Southeast Asia. Alternatively, this could be simply a matter of economics, and location next to the shore and maritime resources could be one reason that there is more material there. Higher ranked individuals could be occupying this area because of privileged access to resources. As disparate as these interpretations may appear, it is possible that all are correct, depending on the way they are viewed. Lastly, the most specific tests conducted with the ceramic data concerned looking at individual types within the larger context of the site as a whole. Considering that it is most likely that the site was occupied as a ring from its formation, at least two clusters around the ring

PAGE 94

94 suggest that ceramic artifacts re presenting two distinct groups of people are present. Traditionally, the Wakulla Check Stamped and Carabelle types of ceramics are considered to be at opposite ends of the temporal spectrum for Weeden Island, yet it appears that these two groups occupied Bayview at a similar time. It will take much more research into the spatial analysis of Weeden Island communities to see if this pattern occurs at other sites as well. Contrasting Site Areas through Faunal Analysis The analysis of the two column samples has shown very little difference between the species from the two areas of the site. Both columns show a heavy dominance on shellfish and marine species of fish throughout. A few instances of small rodents, along with very small turtles and land snails are most likely commensal species and may not have been consumed. The faunal evidence contrasts with the ceramic evidence in that there is not as much variety in the types and proportions of food remains tha t people we re disposing between the two sides of the site Is this because food preparation and consumption are not as variable between the groups as their material culture? Are the similarities between the two column samples an indication of lowered status differences between the areas of the site? If so, then are the ceramic data misleading? An alternative to this interpretation would be that the sample taken for faunal analysis was not large enough, or that too few areas of the site were sampled. This is a definite possibility, although the midden in both column samples was intact and undisturbed. While the information available in the column samples is much more subtle than in the ceramic data, there are still differences visible between the two sides of the site. The utilization of habitat is more diver se in the eastern sample, which could indicate that the inhabitants of this side had preferential access to more areas than the western side. However, this could indicate that particular habitats were only used seasonally, and that the western side of the site was only occupied during part of the year. The use of ritual animals is highly probable at Bayview,

PAGE 95

95 considering the identification of non-meaty portions of deer and turkey vulture. Found only on the western side of the site, this could indicate that this was an area that was used for ritual purposes. Finally, the relative density poses a challenging question to the interpretation of the use of space at Bayview. The three levels analyzed show a very similar density in terms of animal detritus. This is v ery different from the ceramic analysis, which indicates that the eastern side of the site was used much more intensively. It is possible that the use of animals in this case is not a good measure of social duality. However, it is more likely that these co nflicting results are due to the extremely small sample size analyzed. At other sites in the region, there are different species that dominate the assemblage. Does this have to do with environmental limitations or cultural choice? Two species I had expec ted to encounter, which were not at Bayview, were the large whelks ( Busycon sp.) and crown conch ( Melongena corona) that are so numerous at other sites. It could be that the environment at Bayview was not well suited for these species, but at Hares Hammoc k (less than 5 miles east of Bayview and on the same water body), one of the more predominant shellfish species was crown conch (Russo 2008, personal communication). These issues are important elements to consider when interpreting the data from this site, and are paramount in setting future research goals.

PAGE 96

96 CHAPTER 9 CONCLUSION AND SUGGE STIONS FOR FUTURE RE SEARCH My first goal in this project was to incorporate different datasets in order to answer one central question, whether Bayview shows signs of dual social organization. My second goal was to illustrate how these lines of evidence differ in meaning when taken into context with each other. For instance, when considering the zooarchaeological data alone, the interpretation would be that the site does not show signs of dual social or ganization, yet between the eastern and western halves of the site there are subtle status differences within the distribution of ceramic artifacts. There are significant differences between the zooarchaeological sample size and the ceramic sample size, which undoubtedly affects the level of interpretation possible. The ceramic data show that there are many types of Weeden Island pottery found at Bayview, and that this site is placed well within the temporal frameworks build around pottery typologies. Yet the combination of these two datasets with spatial information shows that there are complementary social organization factors at work at Bayview, which indicate possible separations of ethnicity and status, and differential a ctivity areas. Although the data from Bayview has been separated into opposing areas of the site, between east and west, and north and south halves, it is possible that the data may easily fit into a concentric model as well. Undoubtedly more than two axes could exist at Bayview as well as other sites, and that there are certain factors in social organization that are not visible archaeologically. It is up to future research to determine a finer scale of understanding in this vein. Taking the lead from pr evious researchers specializing in circular communities such as Russo, Means, Thompson and Pluckhahn, spatial analysis was chosen as a method of research that would offer more in terms of the understanding of the soc ial dynamics present at the site. Howe ver, Bayview is the only coastal Weeden Island site that has been approached in this way

PAGE 97

97 up to this point. In order to test whether this is indicative of Weeden Island culture as a whole, multiple sites need to be analyzed using this same methodology, in order to compare whether ring villages differ from each other in types of materials or if they reflect a similarity in site size and shape. The regional data show that the Woodland Southeast was much more socially integrated and complex than has been previously thought. I would also suggest that more zooarchaeological analysis be undertaken at coastal sites in the region, to see if the initial data found at Bayview are indicative of status differences at other sites. Finally, I would suggest that more of an effort be made to organize the data that is already on hand concerning Weeden Island ceramics, in order that we might be able to discern regional distributions of types as well as associations of these types with others within particular sites. This way, it may be possible to identify certain affiliations between groups of people or regions, and offer more insight into how social interactions played out between Weeden Island communities. This thesis is only a starting point for understanding that there are more complex interactions occurring between groups during this time period. The next step is to identify what these interactions consisted of, and what they meant to the people living in these communities.

PAGE 98

98 APPENDIX A SHOVEL TEST INFORMAT ION STP FSN G rid North Grid East UTM North UTM East Disturbance Depth Midden Top Midden Bottom Max Depth 1 1 830 370 3329441 634345 none 70 2 2 830 360 3329444 634332 30 70 3 3 830 350 3329438 634321 40 70 4 4 832 332 3329438 634305 50 70 5 5 860 410 332 9460 634372 27 55 6 negative 880 430 3329479 634390 none 50 7 6 810 420 3329412 634389 none 47 51 78 8 7 820 390 3329424 637357 none 60 9 8 780 390 3329384 634362 none 50 10 9 820 431 3329423 634402 none 28 52 83 11 10 800 390 3329401 634362 none 48 12 negative 760 390 3329361 634369 none 60 13 11 860 390 3329461 634359 24 32 40 60 14 12 870 430 3329470 634394 44 60 15 13 840 430 3329445 634396 none 60 16 negative 750 370 3329354 634348 none 60 17 14 770 370 3329374 634340 60 60 18 15 790 370 3329392 634346 none 50 19 16 747 351 3329354 634329 64 72 20 negative 770 350 3329376 634324 48 48 21 17 790 350 3329393 634323 35 35 56 65 22 18 810 350 3329415 634319 none 83 23 19 820 410 3329421 634377 none 52 24 20 7 80 410 3329382 634382 none 14 27 51 25 21 810 370 3329415 634336 none 10 30 58 26 22 840 410 3329439 634380 19 70 27 23 800 410 3329400 634381 none 83 28 24 760 410 3329359 634388 none 80 29 25 750 430 3329350 634409 25 65 30 26 760 445 3329360 634420 90 90

PAGE 99

99 31 27 770 430 3329370 634406 80 80 32 28 800 445 3329401 634419 none 64 33 29 810 330 3329416 634303 33 45 60 70 34 30 840 390 3329440 634358 none 52 35 31 801 430 3329406 634405 none 11 31 50 36 32 780 445 3329380 634420 35 35 37 33 820 419 3329422 634389 none 10 23 40 38 34 790 330 3329396 634303 none 68 39 35 760 470 3329360 634443 none 73 40 36 780 470 3329376 634444 none 86 41 37 800 470 3329398 634443 none 21 74 76 42 38 820 470 3329415 634440 none 6 27 71 43 39 840 470 3329435 634435 32 32 50 140 44A 40 860 468 3329450 634429 49 49 44B 41 859 468 3329450 634429 15 93 45 42 880 470 3329476 634434 none 44 74 103 46 43 800 490 3329414 634460 68 104 47 negative 830 290 3329431 634250 55 60 48 nega tive 860 452 3329453 634414 60 60 49 44 880 450 3329476 634413 none 60 50 45 880 490 3329473 634453 none 10 22 60 51 negative 830 270 3329425 634224 55 55 52 negative 810 270 3329408 634232 55 55 53 47 810 290 3329409 634251 30 60 54 negati ve 810 310 3329414 634273 56 72 55 negative 790 310 3329391 634273 50 50 56 48 770 310 3329372 634275 41 80 57 negative 770 330 3329368 634294 60 60 58 49 830 310 3329343 634283 76 78 59 50 850 350 3329456 634314 none 70 60 51 850 330 332 9459 634302 none 68 61 52 820 488 3329415 634459 none 11 19 84 62 53 840 490 3329436 634458 50 62 63 54 860 490 3329453 634457 none 95 64 55 900 490 3329505 634454 none 130 65 negative 760 490 3329354 634467 none 48

PAGE 100

100 66 56 780 490 3329396 634462 none 60 67 57 850 290 3329453 634286 none 70 68 negative 850 270 3329459 634232 46 62 69 negative 870 270 3329476 634231 none 43 70 negative 890 270 3329497 634234 42 55 71 negative 900 290 3329506 634254 53 53 72 negative 890 290 33 29497 634254 50 50 73 negative 870 290 3329481 634253 74 74 74 59 870 310 3329475 634270 15 73 75 60 890 310 3329493 634273 25 55 76 negative 910 310 3329511 634266 18 50 77 61 910 330 3329513 634289 none 80 78 62 890 350 3329493 634316 45 70 79 63 890 370 3329492 634337 50 110 80 64 870 370 3329475 634338 none 60 81 65 860 330 3329468 634293 72 72 82 negative 870 330 3329475 634317 50 50 83 66 880 390 3329476 634352 none 68 84 67 900 390 3329501 634352 22 22 37 96 85 68 917 390 3329517 634352 none 10 40 73 86 69 945 390 3329542 634349 none 70 87 negative 940 390 3329537 634351 none 40 88 70 920 370 3329522 634330 none 64 89 71 960 410 3329552 634372 none 42 90 72 965 410 3329558 634371 14 72 91 73 945 410 3329538 634374 39 39 92 74 930 410 3329531 634371 none 12 19 72 93 75 890 410 3329488 634375 65 65 94 negative 910 430 3329508 634392 23 50 95 76 910 450 3329505 634413 33 78 96 77 930 450 3329527 634416 30 93 97 negative 920 470 3329514 634433 none 50 98 negative 903 471 3329500 634440 34 95

PAGE 101

101 APPENDIX B ARTIFACT DATABASE STP FSN Grid North Grid East UTM North UTM East Sherds (n) Sherds (g) Debitage Stone Tools Shell (kg) 1 1 830 370 3329441 634345 8 70.1 2 2 830 360 3329444 634332 2 5.5 3 3 830 350 3329438 634321 5 47.3 4 4 832 332 3329438 634305 24 55.3 2 1 0.2 5 5 860 410 3329460 634372 15 115.6 6 negative 880 430 3329479 634390 7 6 810 420 3329412 634389 21 120.4 4.2 8 7 820 390 3329424 637357 18 90.3 1 9 8 780 390 3329384 634362 8 38.9 10 9 820 431 3329423 634402 52 404.9 2 22.8 11 10 800 390 3329401 634362 10 103.4 12 negative 760 390 3329361 634369 13 11 860 390 3329461 634359 7 19.9 0.2 14 12 870 430 3329470 634394 3 7.9 15 13 840 430 3329445 634396 4 18.7 0.3 16 negative 750 370 3329354 634348 17 14 770 370 3329374 634340 2 11.5 18 15 790 370 3329392 634346 3 11.3 19 16 747 351 3329354 634329 1 20.5 20 negative 770 350 3329376 634324 21 17 790 350 3329393 634323 5 23.7 0.5 22 18 810 350 3329415 634319 19 94 23 19 820 410 3329421 634377 8 49.3 24 20 780 410 3329382 634382 23 123 1.2 25 21 810 370 3329415 634336 29 133 9.4 26 22 840 410 3329439 634380 10 22.7 27 23 800 410 3329400 634381 24 240.6 1 28 24 760 410 3329359 634388 10 46.7 29 25 750 430 3329350 634409 4 32.5 30 26 760 445 3329360 634420 4 68.6 31 27 770 430 3329370 634406 2 7.8

PAGE 102

102 32 28 800 445 3329401 634419 5 24.8 33 29 810 330 3329416 634303 85 34 2.6 6 34 30 840 390 3329440 634358 3 14.9 35 31 801 430 3329406 634405 25 162.7 7.6 36 32 780 445 3329380 634420 3 13.5 37 33 820 419 3329422 634389 5 33 6.9 38 34 790 330 3329396 634303 31 161.8 5 0.4 39 35 760 470 3329360 634443 1 6 40 36 780 470 3329376 634444 4 41.9 41 37 800 470 3329398 634443 61 255.3 1 5.2 42 38 820 470 3329415 634440 22 79 1 1 2.7 43 39 840 470 3329435 634435 143 905.8 5 2 5.4 44A 40 860 468 3329450 634429 7 21.8 44B 41 859 468 3329450 634429 95 50 2.4 0.5 45 42 880 470 3329476 634434 107 352.1 1 0.1 46 43 800 490 3329414 634460 6 13.2 47 negative 830 290 3329431 634250 48 negative 860 452 3329453 634414 49 44 880 450 3329476 634413 44 24.9 50 45 880 490 3329473 634453 3 2.9 1 9.8 51 negative 830 270 3329425 634224 1 3 52 negative 810 270 3329408 634232 53 47 810 290 3329409 634251 1 0.9 54 negative 810 310 3329414 634273 55 negative 790 310 3329391 634273 56 48 770 310 3329372 634275 3 11.9 57 negative 770 330 3329368 634294 58 49 830 310 3329343 634283 6 7.5 59 50 850 350 3329456 634314 9 39.1 1 60 51 850 330 3329459 634302 14 57 61 52 820 488 3329415 634459 19 70.3 1.9 62 53 840 490 3329436 634458 4 18.9 63 54 860 490 3329453 634457 85 279.4 0.2 64 55 900 490 3329505 634454 152 644.2 1 8.8 65 negative 760 490 3329354 634467 66 56 780 490 3329396 634462 1 7.4

PAGE 103

103 67 57 850 290 3329453 634286 21 79.2 1 68 negative 850 270 3329459 634232 69 negative 870 270 3329476 634231 1 0.8 70 negative 890 270 3329497 634234 71 negative 900 290 3329506 634254 72 negative 890 290 3329497 634254 73 negative 870 290 3329481 634253 74 59 870 310 3329475 634270 3 10 75 60 890 310 3329493 634273 4 9 76 negative 910 310 3329511 634266 77 61 910 330 3329513 634289 78 311.2 1 78 62 890 350 3329493 634316 8 41.1 79 63 890 370 3329492 634337 12 42.2 1 1 80 64 870 370 3329475 634338 15 54.2 81 65 860 330 3329468 634293 57 130. 7 0.2 82 negative 870 330 3329475 634317 83 66 880 390 3329476 634352 20 42.7 1 0.2 84 67 900 390 3329501 634352 14 35.7 1.5 85 68 917 390 3329517 634352 17 46.9 2 7.8 86 69 945 390 3329542 634349 30 138.7 1 0.3 87 negative 940 390 3329537 634351 88 70 920 370 3329522 634330 36 230.8 89 71 960 410 3329552 634372 13 39.1 1 2 90 72 965 410 3329558 634371 35 124.8 0.4 91 73 945 410 3329538 634374 20 90.3 0.3 92 74 930 410 3329531 634371 27 81.1 1.5 93 75 890 410 3329488 6 34375 10 40.9 1 94 negative 910 430 3329508 634392 95 76 910 450 3329505 634413 21 188.2 96 77 930 450 3329527 634416 48 299.2 1 97 negative 920 470 3329514 634433 98 negative 903 471 3329500 634440 37 158.8 2 3.4

PAGE 104

104 APPENDIX C CERAM IC DA TA FSN STP Grid North Grid East Weight Count Rim Count Type 1 1 830 370 40 4 burnished plain 1 1 830 370 28.6 2 residual plain 1 1 830 370 0.7 2 too small 2 2 830 360 4.3 1 burnished plain 2 2 830 360 1 1 weeden island incised 3 3 830 350 22.6 1 residual plain 3 3 830 350 14.7 2 burnished plain 3 3 830 350 9.6 2 smooth plain 4 4 830 332 2.8 6 too small 4 4 830 332 19.7 5 burnished plain 4 4 830 332 5.5 1 smooth plain 4 4 830 332 8.2 2 residual plain 4 4 830 332 6 1 1 weeden is land incised 4 4 830 332 0.6 1 weeden island plain 4 4 830 332 1.6 4 keith incised 4 4 830 332 5.8 3 UID incised 5 15 860 410 27.2 2 1 swift creek complicated stamped 5 15 860 410 47 3 1 swift creek complicated stamped 5 15 860 410 11.4 1 1 swift creek complicated stamped 5 15 860 410 2.2 1 UID incised 5 15 860 410 9.3 1 UID decorated 5 15 860 410 11.1 3 UID check stamped 5 15 860 410 1.8 2 burnished plain 5 15 860 410 4.5 4 too small 6 7 870 430 6.7 1 1 weeden island plain 6 7 870 430 4.3 1 residual plain 6 7 870 430 2.9 5 too small 6 7 870 430 7.8 2 UID incised 6 7 870 430 5.8 1 keith incised 6 7 870 430 14.9 1 swift creek complicated stamped 6 7 870 430 29.3 6 smooth plain 6 7 870 430 48.5 3 burnished plain 7 8 820 390 15 3 3 weeden island plain 7 8 820 390 45.3 1 1 wakulla check stamped 7 8 820 390 10.8 1 UID check stamped 7 8 820 390 2 6 too small 7 8 820 390 1.2 1 UID incised 7 8 820 390 1.2 1 keith incised 7 8 820 390 6.7 1 residual plain 7 8 820 390 7.7 2 burnished plain 9 10 820 430 105.1 14 burnished plain

PAGE 105

105 9 10 820 430 13.3 1 UID stamped 9 10 820 430 6 1 1 wakulla check stamped 9 10 820 430 81.8 4 UID check stamped 9 10 820 430 14 2 curvilinear complicated stamped 9 10 820 430 5.2 1 1 weeden i sland incised 9 10 820 430 38.4 2 UID incised 9 10 820 430 9.3 2 carrabelle punctated 9 10 820 430 69 10 smooth plain 9 10 820 430 28 4 residual plain 9 10 820 430 19.3 2 1 weeden island plain 9 10 820 430 8.2 4 4 burnished plain 9 10 820 430 4. 4 4 too small 10 11 800 390 65.1 5 burnished plain 10 11 800 390 19.1 2 smooth plain 10 11 800 390 9 1 residual plain 10 11 800 390 6.3 1 residual plain 10 9 780 390 29.1 5 burnished plain 10 9 780 390 4.4 1 smooth plain 10 9 780 390 0.4 2 t oo small 10 9 780 390 4.5 1 weeden island incised 11 13 860 390 11.1 2 cordmarked 11 13 860 390 3.4 3 burnished plain 12 14 870 430 1.1 1 1 weeden island plain 12 14 870 430 6.6 2 UID check stamped 13 15 870 430 18.1 2 UID check stamped 13 15 870 430 0.5 1 too small 14 17 770 370 10.4 1 UID check stamped 15 18 790 370 4 2 1 burnished plain 15 18 790 370 7.1 1 residual plain 16 19 750 350 21.1 1 burnished plain 17 21 790 350 4.4 2 UID incised 17 21 790 350 10.9 1 burnished plain 17 21 790 350 8.2 2 residual plain 18 22 810 350 6.5 1 weeden island incised 18 22 810 350 2.8 4 too small 18 22 810 350 3.3 1 carrabelle incised 18 22 810 350 4.9 1 swift creek complicated stamped 18 22 810 350 26.2 3 residual plain 18 22 810 350 33.7 8 burnished plain 18 22 810 350 14.5 2 smooth plain 19 23 820 410 25 2 burnished plain 19 23 820 410 6.8 3 smooth plain 19 23 820 410 8 1 UID check stamped 19 23 820 410 4.5 1 swift creek complicated stamped 19 23 820 410 4.5 1 1 weeden is land plain 20 24 780 410 41.1 10 1 residual plain 20 24 780 410 7.6 2 burnished plain 20 24 780 410 40.7 6 smooth plain

PAGE 106

106 20 24 780 410 27.8 1 1 weeden island plain 20 24 780 410 1.7 3 too small 20 24 780 410 2.6 1 carrabelle punctated 20 24 780 410 0.8 1 UID punctated 21 25 810 370 54.9 5 burnished plain 21 25 810 370 29.2 1 weeden island punctated 21 25 810 370 7.8 3 smooth plain 21 25 810 370 13.3 2 residual plain 21 25 810 370 3.4 1 swift creek complicated stamped 21 25 810 370 1.2 1 carrabelle punctated 21 25 810 370 8.2 11 too small 21 25 810 370 14.7 2 UID incised 22 26 840 410 15.3 5 residual plain 22 26 840 410 2 2 too small 22 26 840 410 0.7 1 UID stamped 22 26 840 410 4.5 2 UID incised 23 27 800 410 157.6 13 bur nished plain 23 27 800 410 5.3 2 residual plain 23 27 800 410 21.1 1 1 UID decorated 23 27 800 410 0.5 1 too small 23 27 800 410 15.1 1 weeden island incised 23 27 800 410 39.8 2 2 swift creek complicated stamped 24 28 760 410 4.1 1 dunlap fabric impressed 24 28 760 410 20 1 UID incised 24 28 760 410 4.1 1 c rooked river variant/swift creek complicated stamped variant 24 28 760 410 5.4 1 weeden island red 24 28 760 410 6.2 3 burnished plain 24 28 760 410 6.4 2 smooth plain 25 29 750 430 18.3 2 burnished plain 25 29 750 430 12.9 1 smooth plain 25 29 750 430 0.8 1 too small 26 30 760 445 2.5 1 carrabelle punctated 26 30 760 445 52.2 2 1 carrabelle incised 26 30 760 445 13.3 1 smooth plain 27 31 770 430 6.3 1 UID check stamped 2 7 31 770 430 1.4 4 UID incised 28 32 800 445 13 1 UID check stamped 28 32 800 445 0.8 1 burnished plain 28 32 800 445 8.3 2 smooth plain 28 32 800 445 2.3 1 residual plain 29 33 810 330 168.8 26 burnished plain 29 33 810 330 93.3 12 smooth pla in 29 33 810 330 10.6 17 too small 29 33 810 330 14.9 4 residual plain 29 33 810 330 8.8 4 carrabelle incised 29 33 810 330 5 3 UID incised 29 33 810 330 1.5 1 UID decorated

PAGE 107

107 29 33 810 330 25.2 1 carrabelle punctated 29 33 810 330 13.2 2 2 weed en island plain 30 34 840 390 11.5 2 1 residual plain 30 34 840 390 3.4 1 smooth plain 31 35 800 430 14.7 1 simple stamped 31 35 800 430 17.7 1 UID stamped 31 35 800 430 4.7 2 UID incised 31 35 800 430 23.7 4 burnished plain 31 35 800 430 17.3 2 smooth plain 31 35 800 430 6.7 3 residual plain 31 35 800 430 34.1 1 1 carrabelle punctated 31 35 800 430 43.5 1 1 weeden island plain 32 36 780 445 8.3 2 residual plain 32 36 780 445 5 1 weeden island incised 33 37 820 445 11.1 2 2 wakulla che ck stamped 33 37 820 445 9.4 2 UID check stamped 33 37 820 445 7.5 1 smooth plain 33 37 820 445 4.7 1 burnished plain 34 38 790 330 34.3 10 residual plain 34 38 790 330 45.5 8 burnished plain 34 38 790 330 6.5 7 too small 34 38 790 330 8.6 3 weeden island incised 34 38 790 330 1 1 weeden island punctated 34 38 790 330 64.4 6 2 weeden island plain 35 39 760 470 5.1 1 UID incised 36 40 780 470 18.3 2 burnished plain 36 40 780 470 6.2 1 residual plain 36 40 780 470 16.4 1 swift creek c omplicated stamped 37 41 800 470 19.1 23 too small 37 41 800 470 5.8 4 residual plain 37 41 800 470 10.2 1 weeden island red 37 41 800 470 16.4 2 carrabelle punctated 37 41 800 470 2.5 1 UID decorated 37 41 800 470 1.8 1 UID stamped 37 41 800 470 24 1 UID check stamped 37 41 800 470 1.1 1 keith incised 37 41 800 470 8.8 1 carrabelle incised 37 41 800 470 2 1 UID appliqued 37 41 800 470 15.6 1 1 weeden island plain 37 41 800 470 9.6 3 2 UID stamped 37 41 800 470 2 2 carrabelle punctat ed 37 41 800 470 5.1 1 UID incised 37 41 800 470 37.6 5 burnished plain 37 41 800 470 33.4 7 burnished plain 37 41 800 470 59.6 6 smooth plain 38A 42 820 470 0.9 2 too small 38A 42 820 470 0.4 1 linear punctated 38A 42 820 470 10.1 4 1 smooth plain 38A 42 820 470 12.1 4 burnished plain

PAGE 108

108 38A 42 820 470 8 1 weeden island residual red 38A 42 820 470 6.5 2 2 weeden island plain 38A 42 820 470 9.7 4 UID check stamped 38B 42 820 470 13.7 2 burnished plain 38B 42 820 470 9.9 1 UID decorated 38B 42 820 470 2.9 1 residual plain 39 43 840 470 97.9 8 8 weeden island plain 39 43 840 470 153 25 UID check stamped 39 43 840 470 45 42 too small 39 43 840 470 139.8 10 10 wakulla check stamped 39 43 840 470 35.4 1 UID check stamped 39 43 840 470 11.5 5 carrabelle incised 39 43 840 470 25 2 1 carrabelle incised 39 43 840 470 4.4 1 1 weeden island red/franklin 39 43 840 470 12.6 1 1 carrabelle punctated 39 43 840 470 11.2 1 1 keith incised var. 39 43 840 470 25 3 1 ruskin linear punctated 39 43 840 470 13.3 2 swift creek complicated stamped 39 43 840 470 11.1 3 UID incised 39 43 840 470 3.2 1 cobmarked (?) 39 43 840 470 69.5 4 UID stamped 39 43 840 470 15.3 3 2 burnished plain 39 43 840 470 98.7 12 burnished plain 39 43 840 470 86.6 8 1 smooth plain 39 43 840 470 49.9 7 residual plain 40 44A 860 468 3.1 1 burnished plain 40 44A 860 468 5 1 UID stamped 40 44A 860 468 3.5 1 weeden island incised 40 44A 860 468 4 1 1 weeden island plain 40 44A 860 468 5.8 3 UID check sta mped 41 44B 860 468 35.1 3 3 wakulla check stamped 41 44B 860 468 230.1 34 UID check stamped 41 44B 860 468 92.5 14 burnished plain 41 44B 860 468 20.9 3 smooth plain 41 44B 860 468 5.8 2 residual plain 41 44B 860 468 9.8 1 rectilinear com plicated stamped 41 44B 860 468 42.4 2 1 swift creek complicated stamped 41 44B 860 468 10.4 5 UID stamped 41 44B 860 468 21.2 2 1 weeden island incised 41 44B 860 468 0.6 1 UID incised 41 44B 860 468 1.8 1 1 weeden island decorated 41 44B 86 0 468 5 2 weeden island punctated 41 44B 860 468 5.7 3 3 weeden island plain 41 44B 860 468 18.6 16 too small 42 45 880 470 46.6 12 1 burnished plain 42 45 880 470 152.5 21 UID check stamped 42 45 880 470 16.3 7 smooth plain 42 45 880 470 15.5 6 residual plain

PAGE 109

109 42 45 880 470 16.8 28 too small 42 45 880 470 9.3 4 weeden island punctated 42 45 880 470 0.6 1 UID incised 42 45 880 470 90.5 5 5 wakulla check stamped 42 45 880 470 4.9 1 carrabelle punctated 42 45 880 470 0.7 1 weeden island red 43 46 800 490 6.6 3 burnished plain 43 46 800 490 6.3 3 smooth plain 44 49 880 450 12.4 1 1 weeden island plain 44 49 880 450 4.9 1 weeden island punctated 44 49 880 450 7.3 2 burnished plain 47 53 810 290 0.7 1 too small 48 56 770 310 11.8 4 simple stamped 49 58 830 310 2.1 1 smooth plain 49 58 830 310 5 2 residual plain 49 58 830 310 0.3 2 too small 50 59 850 350 26.2 1 weeden island plain 50 59 850 350 4 6 too small 50 59 850 350 5.7 1 smooth plain 50 59 850 350 2.8 1 weed en island incised 51 60 850 330 3.8 4 too small 51 60 850 330 46.5 8 burnished plain 51 60 850 330 4.6 1 weeden island punctated 51 60 850 330 1.9 1 weeden island incised 53 62 840 490 5.9 1 1 wakulla check stamped 53 62 840 490 2.1 1 residual p lain 54 63 860 490 18.8 34 1 too small 54 63 860 490 5 1 weeden island punctated 54 63 860 490 3.1 1 weeden island red 54 63 860 490 120.7 19 UID check stamped 54 63 860 490 8.6 4 1 wakulla check stamped 54 63 860 490 47.4 12 burnished plain 54 63 860 490 17.6 4 smooth plain 54 63 860 490 8.6 5 residual plain 54 63 860 490 4.9 1 UID complicated stamped 54 63 860 490 5.7 1 UID shell impressed 54 63 860 490 37.7 3 3 wakulla check stamped 55A 64 900 490 10 3 swift creek complicated stamped 55A 64 900 490 80.2 17 UID check stamped 55A 64 900 490 2.7 1 weeden island punctated 55A 64 900 490 16.4 1 1 weeden island plain 55A 64 900 490 35.5 9 1 burnished plain 55A 64 900 490 7.4 3 smooth plain 55A 64 900 490 11 14 too small 55A 64 900 490 6.5 2 UID incised 55A 64 900 490 2.5 1 weeden island incised 55A 64 900 490 10.4 2 UID stamped 55B 64 900 490 226.1 27 UID check stamped 55B 64 900 490 6.4 1 UID complicated check stamped

PAGE 110

110 55B 64 900 490 2.9 1 simple stamped 55B 64 900 490 3 1 swift creek complicated stamped 55B 64 900 490 45.5 17 1 burnished plain 55B 64 900 490 67.8 14 1 smooth plain 55B 64 900 490 56.1 6 residual plain 55B 64 900 490 14 4 4 weeden island plain 55B 64 900 490 11.6 14 too small 55B 64 900 490 17.5 2 2 wakulla check stamped 55B 64 900 490 4.4 1 crooked river check stamped 55B 64 900 490 6.2 1 keith incised 55B 64 900 490 1.1 1 UID incised 55B 64 900 490 4.9 2 1 weeden island punctated 56 66 780 490 7.3 1 UID check stamped 57 67 850 290 0.8 1 weeden island incised 57 67 850 290 5.9 1 tucker ridge pinched 57 67 850 290 4.9 1 carrabelle incised 57 67 850 290 2.9 5 too small 57 67 850 290 52.2 9 burnished plain 57 67 850 290 12.2 5 smooth plain 59 74 870 310 2.6 1 keith incised 59 74 870 310 3.9 1 UID check stamped 59 74 870 310 3.5 1 burnished plain 60 75 890 310 5.6 2 smooth plain 60 75 890 310 3.3 2 residual plain 61 77 910 330 49 11 1 swift creek complicated stamped 61 77 910 330 23 27 too small 61 77 910 330 93.7 17 burnished plain 61 77 910 330 65.4 6 smooth plain 61 77 910 330 17.4 5 residual plain 61 77 910 330 10.7 1 UID decorated 61 77 910 330 2.9 1 UID incised 61 77 910 330 1.8 1 1 weeden island plain 61 77 910 330 2.7 1 weeden island red 61 77 910 330 17.9 4 1 swift creek complicated stamped 61 77 910 330 7.2 3 3 swift creek complicated stamped 61 77 910 330 16.4 4 swift creek complicated stamped 63 79 890 370 14.7 1 1 weeden island appliqued 63 79 890 370 12.8 5 2 weeden island plain 63 79 890 370 1.5 1 burnished plain 63 79 890 370 12 4 1 residual plain 63 79 890 370 0.8 1 too small 65 81 860 330 12.1 20 too small 65 81 860 330 4.6 1 UID punctated 65 81 860 330 35.1 8 burnished plain 65 81 860 330 30.3 10 smooth plain 65 81 860 330 18.8 5 residual plain

PAGE 111

111 65 81 860 330 7.9 2 2 weeden island plain 65 81 860 330 4.9 1 1 swift creek complicated stamped 65 81 860 330 5.1 2 1 keith incised 65 81 860 330 7.7 3 1 carrabelle punctated 65 81 860 330 3.6 2 UID incised 65 81 860 330 4.6 1 UID punctated 66 83 880 390 3.5 5 too small 66 83 880 390 13.1 3 burnished plain 66 83 880 390 8.9 3 smooth plain 66 83 880 390 4.8 3 residual plain 66 83 880 390 4.8 3 swift creek complicated stamped 66 83 880 390 6 1 1 simple stamped 66 83 880 390 1.1 1 UID stamped 67 84 900 390 2.1 4 too small 67 84 900 390 5.8 2 burnished plain 67 84 900 390 9.3 1 smooth plain 67 84 900 390 1.8 1 residual plain 67 84 900 390 5.5 1 simple stamped 67 84 900 390 4.6 1 keith incised 67 84 900 390 4.3 1 indian pass incised 67 84 900 390 1.5 1 1 weeden island plain 67 84 900 390 0.4 1 UID punctated 68 85 917 390 7.3 1 swift creek complicated stamped 68 85 917 390 23.1 4 burnished plain 68 85 917 390 5.3 2 residual plain 68 85 917 39 0 5.9 8 too small 68 85 917 390 5.1 1 UID stamped 69 86 945 390 6 9 too small 69 86 945 390 31.3 6 residual plain 69 86 945 390 58.5 8 burnished plain 69 86 945 390 10.6 1 1 weeden island plain 69 86 945 390 32 6 UID check stamped 70 88 920 370 8.6 1 1 weeden island red 70 88 920 370 2.1 1 swift creek complicated stamped 70 88 920 370 3.4 1 carrabelle punctated 70 88 920 370 7.3 2 weeden island punctated 70 88 920 370 8.1 1 1 weeden island plain 70 88 920 370 4 6 too small 70 88 920 3 70 104 9 burnished plain 70 88 920 370 60 9 smooth plain 70 88 920 370 33.1 7 residual plain 71 89 960 410 13.3 1 1 weeden island plain 71 89 960 410 1.6 1 carrabelle punctated 71 89 960 410 16.5 4 burnished plain 71 89 960 410 2.2 1 smooth pla in 71 89 960 410 5.4 6 too small 72 90 965 410 6.9 9 too small

PAGE 112

112 72 90 965 410 2.8 1 1 weeden island incised 72 90 965 410 25.5 8 UID check stamped 72 90 965 410 8.9 4 burnished plain 72 90 965 410 14.4 1 1 wakulla check stamped 72 90 965 410 15.6 4 residual plain 72 90 965 410 11 4 smooth plain 72 90 965 410 38.5 3 3 weeden island plain 73 91 945 410 6.2 1 carrabelle punctated 73 91 945 410 13.4 2 weeden island incised 73 91 945 410 31.5 2 UID check stamped 73 91 945 410 17.2 5 smooth plain 73 91 945 410 2.9 3 too small 73 91 945 410 2.2 3 too small 73 91 945 410 16.4 6 burnished plain 74 92 930 410 13.4 5 UID check stamped 74 92 930 410 14.6 1 swift creek complicated stamped 74 92 930 410 4.5 2 burnished plain 74 92 930 410 8.1 8 too small 74 92 930 410 6.7 3 residual plain 74 92 930 410 29.9 4 smooth plain 74 92 930 410 0.8 1 UID stamped 74 92 930 410 2.4 1 1 weeden island plain 75 93 890 410 32.3 5 UID check stamped 75 93 890 410 0.8 3 too small 75 93 890 410 2.1 1 weeden island red 75 93 890 410 5.2 1 burnished plain 76 95 910 480 45.2 1 1 mercier red on buff 76 95 910 480 5.2 1 UID zoned stamped 76 95 910 480 2.4 3 too small 76 95 910 480 31.3 7 UID check stamped 76 95 910 480 75.2 1 1 weeden island plain 76 95 910 480 3.3 1 weeden island incised 76 95 910 480 1.6 1 carrabelle punctated 76 95 910 480 15.1 3 burnished plain 76 95 910 480 8 3 1 residual plain 77 96 930 450 253.9 28 UID check stamped 77 96 930 450 14.5 3 residual plain 77 96 930 450 1.9 1 smooth plain 77 96 930 450 5 7 too small 77 96 930 450 15.1 6 burnished plain 77 96 930 450 7.7 1 swift creek complicated stamped 78 98 900 470 104.3 10 UID check stamped 78 98 900 470 11.3 13 too small 78 98 900 470 15.2 4 burnished plain 78 98 900 470 7.5 2 smooth plain 78 98 900 470 4.1 1 residual plain 78 98 900 470 9 1 1 weeden island plain 78 98 900 470 6.5 1 carrabelle incised

PAGE 113

113 APPENDIX D COLUMN SAMPLE 1 LEVE L 4 1/8 Inch Sample Group Taxa Element NISP MN I Side Portion Wt (g) Osteichthyes UID Osteichthyes N/A 1127 N/A N/A N/A 14.79 Siluriformes pectoral spine 1 1 partial 0.16 Siluriformes maxilla 1 R partial 0.04 Siluriformes maxilla 1 L complete 0.05 Cynoscion sp. atlas 1 1 N/A complete 0.03 Cynoscion sp. otolith 1 R partial 0.07 Micropogonias undulatus otolith 2 1 L complete 0.2 Mollusca UID Mollusca N/A N/A N/A N/A 869.43 Gastropoda Crepidula sp. valve 15 15 N/A 0.35 Polygyridae valve 79 79 N/A 0.84 Bivalvia Argopecten irradians 5 5 partial 0.66 Subtotal 1233 102 886.62 1/16 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Animalia UID Animalia N/A N/A N/A N/A N/A 479.76 Subtotal 479.76 Total 1233 102 1366.38

PAGE 114

114 APPENDIX E COLUMN SAMPLE 2 LEVE L 2 Column Sample 2 Level 2 1/4 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Mammalia UID Mammalia N/A N/A N/A N/A N/A 7.76 Odocoileus virginianus antler 11 1 L partial 11.1 Osteichthyes UID Osteichthy es N/A 122 N/A N/A N/A 2.72 Ariidae articular 1 1 L partial 0.13 Caranx sp. vertebra 3 1 N/A complete 0.27 Cynoscion sp. vertebra 2 1 N/A partial 0.22 Mugil sp. vertebra 30 1 N/A complete 3.49 Mugil sp. hyomandibular 1 L partial 0.14 Mugil sp. hyomandibular 1 R partial 0.13 Paralichthyes sp. vertebra 2 1 complete 0.16 Mollusca UID Mollusca N/A N/A N/A N/A N/A 835.12 Gastropoda Noetia ponderosa valve 1 1 R complete 1.99 Bivalvia Chione cancellata valve 2 2 N/A partial 0.6 Argopecten irr adians valve 5 5 N/A partial 2.92 Crassotrea virginica valve 82 82 L partial 725.88 Crassotrea virginica valve 67 R partial 548.38 Subtotal 330 96 2141.01 1/8 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Mammalia UID Mamm alia longbone shaft 10 N/A N/A N/A 0.83 Peromyscus sp. mandible 1 1 L complete 0.03 Peromyscus sp. femur 1 R distal and shaft 0.03 Reptilia Testudines carapace 1 fragment 0.06 Osteichthyes UID Osteichthyes N/A 1092 N/A N/A N/A 16.04 Elops saurus vertebra 6 1 N/A complete 0.21 Arius felis vertebra 4 1 N/A complete 0.1 Caranx sp. premaxilla 1 1 L partial 0.04 Cynoscion nebulosus otolith 2 2 R complete 0.62 Cynoscion sp. vertebra 2 N/A complete 0.12 Micropogonias undulatus vertebra 1 1 N/ A complete 0.05 Mugil sp quadratus 2 2 R partial 0.04 Centropomus sp. vertebra 5 1 N/A complete 0.2 Serranidae dentary 1 1 R partial 0.03 Mollusca UID Mollusca N/A N/A N/A N/A N/A 247.82 Subtotal 1129 11 266.22

PAGE 115

115 1/16 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Animalia UID Animalia N/A N/A N/A N/A N/A 121.83 Aves Aves rib 1 1 part too small Reptilia Kinosternidae pubis 2 2 R part 0.008 Kinosternidae ischium 1 L part 0.005 Kinosternidae maxilla 1 L part 0.017 Osteichthyes Opsanus tau atlas 1 1 N/A complete 0.007 Opsanus sp. vertebra 1 N/A complete 0.001 Invertebrata Polygyridae valve 23 23 N/A complete 0.056 Subtotal 30 27 121.924 TOTAL 1489 134 2529.154

PAGE 116

116 APPENDIX F COL UMN SAMPLE 2 LEVEL 4 Column Sample 2 Level 4 1/4 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Mammalia Sylvilagus floridanus distal femur epiphysis 1 1 R complete 0.29 Aves Cathartes aura ulna 1 1 R fragments and distal port ion 2.55 Reptilia Apalone ferox carapace 1 1 fragment 0.33 Osteichthyes UID Osteichthyes N/A 216 N/A N/A N/A 4.71 Caranx crysos hyomandibular 1 2 L partial 0.11 Caranx sp. vertebra 2 N/A complete 0.26 Cynoscion sp. quadrate 1 1 R complete 1.12 Mugil cephalus atlas 1 1 N/A complete 0.15 Mugil sp. operculum 1 L partial 0.14 Mugil sp. vertebra 9 N/A complete 1.12 Invertebrata UID Mollusca N/A N/A N/A N/A fragment 231.14 Gastropoda Crepidula sp. valve 24 24 N/A complete 3.04 Fasciolaria tulipa valve 3 3 N/A partial 43.77 Chicoreus pomum valve 3 3 N/A partial 42.35 Bivalvia Argopecten irradians valve 13 15 R partial 17.68 Argopecten irradians valve 15 L fragment 13.65 Argopecten irradians valve partial 145.14 Crassotrea virginic a valve 122 122 R partial 715.86 Crassotrea virginica valve 66 L partial 690.23 Subtotal 480 174 1913.64 1/8 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Osteichthyes UID Osteichthyes N/A 1079 N/A N/A N/A Caranx sp. hyomandibular 2 2 R partial 15.04 Caranx sp. hyomandibular 2 L partial 0.11 Cynoscion nebulosus otolith 2 2 R complete 0.1 Mugil sp. hyomandibular 1 1 R partial 1.12 Invertebrata UID Mollusca 144.91 Subtotal 1086 5 161.28

PAGE 117

117 1/16 Inch Sample Group Taxa Element NISP MNI Side Portion Wt (g) Animalia UID Animalia N/A N/A N/A N/A N/A 83.74 Mammalia Rodentia radius 1 1 R proximal 0.008 Reptilia Kinosternidae pubis 1 1 L partial 0.006 Osteichthyes Elops saurus vertebra 3 1 N/A partial 0.012 Cynoscion sp. vertebra 1 1 N/A complete too small Gastropoda Polygyridae valve 12 12 N/A complete 0.039 Subtotal 18 16 83.805 TOTAL 1584 195 2158.725

PAGE 118

118 APPENDIX G RADIOCARBON RESULTS FROM BETA ANALYTIC

PAGE 119

119

PAGE 120

120 LIST OF REFERE NCES Anderson, David G. and Robert C. Mainfort, Jr. (editors) 2002 The Woodland Southeast University of Alabama Press, Tuscaloosa. Bartlett, Mary Lee and Patricia A. McAnany 2000 Crafting communities: the materialization of Formative M aya identi ties. In The Archaeology of Communities: A New World Perspective edited by Marcello A. Canuto and Jason Yaeger, pp. 102122. Routledge, New York. Bender, Barbara 2002 Time and Landscape. Current Anthropology 43(Supplemental Issue):S103-S112. Bense, Judith A. 1994 Archaeology of the Southeastern United States: Paleoindian to World War I Academic Press, San Diego. 1998 Santa Rosa -Swift Creek in Northwestern Florida. In A World Engr aved: Archaeology of the Swift Creek Culture ed ited by Mark Williams and Daniel T. Elliot, pp. 247 2 73. University of Alabama Press, Tuscaloosa. Bourdieu, Pierre 1977 Outline of A Theory of Practice Cambridge University Press, Cambridge. Brose, David S. 1979 An Interpretation of the Hopewellian Traits in Florida. In Hopewell Archaeology: The Chillicothe Conference edited by David S. Brose and Nomi Greber, pp. 141 150. Kent State University Press, Kent. Brose, David S. and Nomi Greber (editors) 1979 Hopewell Archaeology: The Chillicothe Conference. Kent State University Press Kent. Canuto, Marcello A. and Jason Yaeger (editors) 2000 The Archaeology of Communities: A New World Perspective. Routledge, New York. Claasen, Cheryl 1998 Shells Cambridge University Press, Cambridge Cleland, C. E. ( editor) 1977 For the Director: Research Essays in Honor of James B. Griffin. Anthropological Papers, no. 61. Museum of Anthropology, University of Michigan, Ann Arbor. Cordell, Ann S. 1984 Ceramic Technology at a Weeden Island Period Archaeological Site In North Florida. Ceramic Notes No. 2 Occasional Publications of the Ceramic Technology Laboratory, Florida State Museum, Gainesville.

PAGE 121

121 Dietler, Michael 2001 Theorizing the Feast: Rituals of Consumpt ion, Commensal Politics, and Power in African Contexts. In Feasts: Arc haeological and Ethnographic Perspectives on Food, Politics and Power e dited by Michael Dietler a nd Brian Hayden, pp. 65 114. Smithsonian Institution Press, Washington D.C. Doran, Glenn H. and Bruce J. Piat ek 1985 Archaeological Investigations at Naval Li ve Oaks, Studies in Spatial Patterning and Chrononology in the Gulf Coast of Florida. Submitted to the National Park Service, Sout heast Archaeological Center, Tallahassee, Florida. Emery, Kitty F 2003 The Noble Beast: Status and Differential Access to Animals in the Maya World World Archaeology 34(3):498515. 2007 Assessing the impact of ancient Maya animal use. Journal for Nature Conservation 15(3):184195. Erlandson, Jon M. 2001 The A rchaeology of Aquatic Adaptations: Paradigms for a New Millennium. Journal of Archaeological Research 32:287350. Ervynck, Anton, Wim Van Neer, Heide Huster Plogmann, and Jorg Schibler 2003 Beyond Affluence: The Zooarchaeology of Luxury World Archaeology 34(3):428441. Fewkes, Jesse W. 1924 Preliminary Archaeological Investigations at Weedon Island, Florida. Smithsonian Miscellaneous Collections 76(13):126. Giddens, Anthony 1979 Central Problems in Social Theory: Action, Struggle and Contr adiction in Social Analysis. University of California Press, Berkeley. Hayden, Brian 2001 Fabulous Feasts: A Prolegomenon to the Importance of Feasting. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power edit ed by Michael Dietler and Brian Hayden, pp. 23 64. Smithsonian Institution Press, Washington D.C. Heckenberger, Michael J. 2004 The Symbolic Economy of Power: Plazas as Persons. In The Ecology of Power edited by Michael J. Heckenberge r, pp. 291347. Routledge, NewYork. Hodder, Ian 1991 Interpretive Archaeology and its Role. American Antiquity 56(1):718.

PAGE 122

122 Hodder, Ian (cont.) 1982 Theoretical Archaeology: A Reactionary View. Symbolic and Structural Archaeology edited by Ian Hodder, pp. 1 16. Cambridge University Press, Cambridge. Ingold, Tim 1993 The Temporality of Landscape. World Archaeology 34(2):152174. Jackson, H. Edwin, and Susan L. Scott 2003 Patterns of Elite Faunal Utilization at Moundville, Alabama. American An tiquity 68(3):552572. Jones ODay, Sharyn 2004 Past and present perspectives on secular ritual: food and the fisherwomen of the Lau Islands, Fiji. In Behaviour Behind Bones edited by Sharyn Jones ODay, pp. 1531 61. Proceedings of the 9t h ICAZ Conference, Durham 2002, Vol. 1. Oxbow Books, O xbow. Kirch, Patrick V. and Sharyn Jones O'Day 2003 New Archaeological Insights into Food and Status: A Case Study from Pre -Contact Hawaii World Archaeology 34(3):484497. Knight, Vernon James 2001 Feasting and the Emergence of Platform Mound Ceremonialism in Eastern North America. In Feasts: Archaeological and Ethnographic Perspectives on Food, Politics and Power edited by Michael D ietler and Brian Hayden, pp. 311333. Smi thsonian Institution Press, Washington D.C. Kohler, Timothy 1975 The Garden Patch Site: A Minor Weeden Isl and Ceremonial Center on the North Peninsular Florida Gulf Coast. U npublished Masters thesis, Department of Anthropology, Univers ity of Florida, Gainesville. 1978 The Social and Chronological Dimensions of V illage Occupation at a North Florida Weeden Island Period Site. Unpublished PhD dissertation, Department of Anthropology, University of Florida, Gainesville. Ku s, Susan M. 1983 The Social Representation of Space: Dimensioning the Cosmological and the Quotidian. In Archaeological Hammers and Theories edited by James A. Moore and Arthur S. Keene. Academic Press, Burlington, Massachusetts. Lele, Veere ndra P. 2006 Material Habits, Identity, Semeiotic. Journal of Social Archaeology 6(1):4870. Lvi -Strauss Claude 1963 Structural Anthropology Basic Books, London.

PAGE 123

123 Longacre, William A. 1970 Archaeology as Anthropology Anthropological Papers of the University of Arizona No.17. The University of Arizona Press, Tuscon. Luer, George M. and Marion M. Almy 1982 A Definition of the Manasota Culture. Florida Anthropologist 35:3458. Means, Bernard K. 2007 Circular Villages of the Monongahela Tradition University of Alabama Press, Tuscaloosa. Milanich, Jerald T. 1974 Life in a 9th Century Indian Household: A Weede n Island Fall -Winter Site on the Upper Apalachicola River, Florida. In Florida Bureau of Historic Sites and Properti es Bulletin 4 pp. 1 44. Florida Depa rtment of State, Division of Archives, History, and Records Management, Tallahassee 1994 Archaeology of Precolumbian Florida. University Press of Florida Gainesville. 2002 Weeden Island Cultures. In The Wo odland Southeast edited by David G. Anderson and Robert C. Mainfort, Jr., pp. 352372. University of Alabama Press, Tuscaloosa. Milanich Jerald T., Ann S. Cordell, Vernon J. Knight, Jr., Timothy A. Kohle r, and Brenda A. Sigler Lavelle 1997 Archae ology of Northern Florida, A.D. 200-900: The McKeithen Weeden Island Culture University Press of Florida, Gainesville. Moore, C. B. 1918 The Northwestern Florida Coast Revisited. Journal of the Academy of Natural Sciences of Philadelphia 16:513 581. Nanfro, Claire Elizabeth 2004 An Analysis of Faunal Remains from th e Bird Hammock Site (8Wa30). Unpublished Masters thesis, Department of Anthropology, Florida State University, Tallahassee. Power, Susan C. 2004 Early Art of the Southeas tern Indians: Feathered Serpents & Winged Beings. University of Georgia Press, Athens. Pluckhahn, Thomas J. 2003 Kolomoki: Settlement, Ceremony and Status in the Deep South, A.D. 350 to 750. University of Alabama Press, Tuscaloosa. Purdy, B arbara A. 1991 The Art and Archaeology of Floridas Wetlands CRC Press, Boca Raton.

PAGE 124

124 Reitz, Elizabeth J., Lee A. Newsom, and Sylvia Scudder (editors) 1996 Case Studies in Environmental Archaeology Int erdisciplinary Contributions to Archaeology, Series editor Michael Jochim. Plenum, New York. Reitz, Elizabeth J., and Irvy R. Quitmyer 1988 Faunal Remains from Two Coastal Georgia Sift Creek Sites. Southeastern Archaeology 7(2):95 108. Reitz, Elizabeth & Elizabeth Wing 1999 Zooarchaeology Cambridge University Press, Cambridge. Russo, Michael 2004 Measuring Shell Rings for Social Inequality. Signs of Power: The Rise of Cultural Complexity in the Southeast edited by Jon L. Gibson and Philip J. Carr, pp. 26 70. University of Ala bama Press, Tuscaloosa. Russo, Michael and Gregory Heide 2002 The Joseph Reed Shell Ring. The Florida Anthropologist 55(2):6787. Russo, Michael, Marg o Schwadron, and Emily M. Yates 2006 Archaeological Investigation of the Bayview Site ( 8BY137): A We eden Island Shell Midden. Submitted to Tyndall Air Force Base, Panama City, Florida. Russ o, Michael and Irvy R. Quitmyer 1996 Sedentism in Coastal Populations of South Florida. In Case Studies in Environmental Archaeology edited by Elizab eth J. Reitz, Lee A. Newsom, and Sylvia Scudder, Series editor Michael Joch im, pp. 213232. Plenum, New York. Saunders, Rebecca 2004 The Stratigraphic Sequence at Rollins Shell Ring: Implications for Ring Function. The Florida Anthropologist 57(4):249 270. Savage, Stephen H. 1990 GIS in Archaeological Research. In Interpreting Space: GIS and Archaeology edited by Kathleen Allen, M. S., Stanton W. Green, and Ezra B.W. Zubrow, pp. 330355. Taylor and Francis New York Schmidt, Peter R 1983 An Alternative to a Strictly Materialist Perspective. American Antiquity 48(1):6979. Schmidtt, David N. and Karen D. Lupo 2008 Do faunal remains reflect socioeconomic st atus? An ethnoarchaeological study among Central African farmers in the nor thern Congo Basin Journal of Anthropological Archaeology, in press.

PAGE 125

125 Scott, Elizabeth M. 2001 Food and Social Relations at Nina Plantation. American Anthropologist 103(3):671691. Sears, William H. 1953 Kolomoki Burial Mounds and the Weeden Isl and Mortuary Complex. American Antiquity 18(3):223229. 1954 The Sociopolitical Organization of Pre -Co lumbian Cultures on the Gulf Coastal Plain. American Anthropologist 56(3):339346. 1961 The Study of Social and Religious Systems in North American Archaeology. Current Anthropology 2(3):223246. Slemmons Lawson, Arianna 2005 Vertebrate Fauna from the Refuge Fire Tower Site (8Wa14): A Study of Coastal Subsistence in the Early Woodland Period. Unpublished Masters Thesis, Departm ent of Anthropology, Florida State University, Tallahassee. Snow Frankie and Keith Stephenson 1999 Swift Creek Designs: A Tool for Monitoring Interaction. In A World Engraved: Archaeology of the Swift Creek Culture edited by Mark Williams and Daniel T. Elliot, pp. 99 111. University of Alabama Press, Tuscaloosa. Stoltman James B. and Frankie Snow 1999 Cultural Interaction within Swift Creek Society: People, Pots and Paddles. In A World Engraved: Archaeology of the Swift Creek C ulture edited by Mark Williams an d Daniel T. Elliot, pp. 130 153. University of Alabama Press, T uscaloosa. Steward, Julian 1974 Theory of Culture Change: The Method of Multilinear Evolution. University of Illinois Press, Champaign. Steponait is, Vincas P. 1986 Prehistoric Archaeology in the Southeastern United States, 1970 1985. Annual Review of Anthropology 15:363404. Stine, Roy S. and David P. Lanter 1990 Considerations for Archaeology Database Design. In Interpreting Space: GIS and Archaeology edited by Kathleen Allen, M. S., Stanton W. Green, and Ezra B.W. Zubrow, pp. 80 89. Taylor and Francis New York Tambiah, Stanley Jeyaraja 1985 The Galactic Polity in Southeast Asia. In Culture Thought and Social Action: An Anthropological Perspective edited by S. J. Tambiah, pp. 252 286. Harvard University Press, Cambridge.

PAGE 126

126 Thompson, Victor D. 2007 Articulating Activity Areas and Formation Processes at the Sapelo Island Shell Ring Complex. Southeastern Arc haeology 26(1):91107. Thompson, Victor D., Matthew D. Reynolds, Bryan Haley, Richard Jefferies, Jay K. Johnson, and Laura Humphries 2004 The Sapelo Shell Ring Complex: Shallow Geophysics on a Georgia Sea Island. Southeastern Archaeology 23(2):192201. Trigger, Bruce G. 1989 A History of Archaeological Thought Cambridge University Press, London. Wh eatley, David and Mark Gillings 2002 Spatial Technology and Archaeology: The Archaeological Applications of GIS Taylor and Francis, New Yor k. Williams, Mark and Daniel T. Elliot (editors) 1999 A World Engraved: Archaeology of the Swift Creek Culture University of Alabama Press, Tuscaloosa. Willey, Gordon R. 1949 Archaeology of the Florida Gulf Coast Smithsonian Miscellaneous Colle ctions 113, Washington, D.C. 1958 Burial Mounds on the Gulf Coastal Plain. American Antiquity 23(3):274284. 1945 The Weeden Island Culture: A Preliminary Definition. American Antiquity 10(3): 225254. Wobst, H. M. 1977 Stylistic Behavi or and Information Exchange. In For the Director: Research Essays in Honor of James B. Griffin edited by C. E. Cleland, pp. 317342. Anthropological Papers, no. 61. Museum of Anthropology, University of Michigan, Ann Arbor. Yaeger, Jason 2000 The social construction of communities in t he Classic Maya countryside: strategies of affiliation in western Belize. In The Archaeology of Communities: A New World Perspective edited by Marcello A. Canuto and Jas on Yaeger, pp. 123142. Routledge New York. Yesner, David R. 1980 Maritime Hunter Gatherers: Ecology and Prehistory. Current Anthropology 21(6):727765.

PAGE 127

127 BIOGRAPHICAL SKETCH Tria Marie Ellison was born i n 1974, in Rockford, Illinois. She is the oldest of seven children and the only to have received a college degree thus far. In 2006, s he graduated m agna c um l aude with a BA in a nthropology from UCLA. While a t UCLA she participated in several undergraduate research projects with the Cotsen Institute of Archaeology at UCLA including developing background research for a historic archaeology project in Moyo Province, Uganda, attending archaeological field school in Bocas del Toro, Panama and working with American Indians on many archaeological sites in California. These expe riences developed her passion for archaeology, and led her t o pursue an advanced degree in a nthropology at the University of Florida, where she has been studying on an Alumni Fellowship for the past three years. At the University of Florida, Tria develope d her passion for archaeology of the Southeastern United States. In 2007 she volunteered at the Kingsley Plantation Field School at the Timucuan Ecological and Historic Preserve National Park, in Jacksonville, Florida, under the direction of James Davidson and in 2008 she participated in Thomas Pluckhahn's (USF) field school at Kolomoki Mounds State Park in Early, Georgia. She has presented multiple papers at the Southeastern Archaeological Conference, and the Southern Anthropological Society and has parti cipated in annual round table discussions with other Weeden Island scholars at the Weedon Island Preserve in Tampa. Beyond the world of academia and university studies, Tria enjoys s pending time with her family, traveling around the world and listening to peoples stories She and her husband, Andrew Belcourt are currently expecting their first child, who they are extremely excited to meet. Upon graduation, Tria plans to continue her career in cultural resource management, and has recently accepted a position as a project archaeologist with Cogstone Consulting, Inc. in Santa Ana California.