1 ISOTOPIC INVESTIGATIONS OF ARCHAIC PERIOD SUBSISTENCE AND SETTLEMENT IN THE ST. JOHNS RIVER DRAINAGE, FLORIDA By BRYAN DUANE TUCKER A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009
2 2009 Bryan Duane Tucker
3 To my family and friends
4 ACKNOWLEDGMENTS Id like to thank my committee for the guidance they have provided. John Krigbaum and Ken Sassaman provided invaluable a ssistance in the course of this research. This work would not have been possible without th e technical wizardry of Jason Curtis, and I appreciate his willingness to get data to me on short notice. Paul Thacker and Rebecca Saunders have each read drafts of various parts of this research and their ongoing support th roughout this process is greatly appreciated. I thank John Bloch and Ross Secord, for allowing me the freedom to finish this research while still maintaining financial stability. My wife Megan has been patient and supporting through this process a nd my daughter Eavan has reminded me to stop and have some fun on the way. Last, I would like to acknowledge the late Bob Blakely, for helping to develop the field of bioarchaeology and for introducing me to it.
5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........8 LIST OF FIGURES.........................................................................................................................9 ABSTRACT...................................................................................................................................11 CHAP TER 1 INTRODUCTION..................................................................................................................14 Archaeological Models of Mobility........................................................................................ 19 Dissertation Outline........................................................................................................... .....25 Variation in Burial Treatment......................................................................................... 26 Variation in Diet.............................................................................................................. 27 Variation in Stable Isotope Ratios Recorded in H uman Tooth Enamel.......................... 29 Variation in Diet by Season at Harris Creek and W indover........................................... 30 Conclusion..............................................................................................................................31 2 ISOTOPIC INVESTIGATIONS OF MI DDLE ARCHAIC MORTUARY PRACTICE S AT THE HARRIS CREEK SITE, TICK ISLAND, FLORIDA............................................. 34 Introduction................................................................................................................... ..........34 Mortuary Theory in Archaeology........................................................................................... 35 Status or Rank..................................................................................................................35 Group Mobility, Settlement Patterns and Mortuary Practices ......................................... 36 Beyond Saxe-Binford......................................................................................................38 Harris Creek Burial Mound....................................................................................................38 Previous Investigations at the Site................................................................................... 39 Burial Form.................................................................................................................... .40 Body Positioning.............................................................................................................40 Stable Isotope Analysis........................................................................................................ ...41 Carbon Isotopes...............................................................................................................41 Oxygen Isotopes..............................................................................................................42 Nitrogen Isotopes.............................................................................................................43 Materials and Methods...........................................................................................................43 Results.....................................................................................................................................45 Accelerator Mass Spectrometry Dates............................................................................ 45 Stable Isotopic Values of Fauna l Bone Apatite and Local Water ................................... 46 Stable Isotopic Values of Tooth Enamel......................................................................... 47 Discussion...............................................................................................................................48 Rank and Status...............................................................................................................48 Settlement and Mobility.................................................................................................. 49
6 Function of Mounds........................................................................................................51 Conclusion..............................................................................................................................52 3 DIETARY CONTINUITY AND CHANGE DURING TH E FLORIDA ARCHAI C........... 63 Introduction................................................................................................................... ..........63 Background.............................................................................................................................65 Stable Isotopes and Dietary Reconstruction.................................................................... 65 Apatite to Collagen Spacing and Diet............................................................................. 66 Previous Isotopic Studies of Harris Creek and W indover............................................... 67 Site Descriptions..............................................................................................................68 The Windover site (8BR246)................................................................................... 68 The Harris Creek/Tick Island site (8VO24)............................................................. 69 Materials.................................................................................................................................70 Methods..................................................................................................................................70 Results.....................................................................................................................................71 Discussion...............................................................................................................................72 Conclusion..............................................................................................................................77 4 EVALUATING A SERIAL SAMPL ING METHODOLOGY.............................................. 83 Introduction................................................................................................................... ..........83 Environmental Buffering a nd Isotopic Attenuation ...............................................................84 Tooth and Enamel Development............................................................................................ 86 Maturation and Mineralization Waves............................................................................ 86 Carbonate Substitutions in Enamel................................................................................. 87 Striae of Retzius and Perikymata and Crown Formation Times..................................... 90 Materials and Methods...........................................................................................................91 Sample selection.............................................................................................................. 91 Site Descriptions..............................................................................................................91 External SamplingDental Drill.................................................................................... 92 Internal SamplingMicromill........................................................................................ 92 Pretreatment.....................................................................................................................93 Results.....................................................................................................................................93 Discussion...............................................................................................................................94 Effects of Secondary Mineralization...............................................................................94 Source of Isotopic Variation............................................................................................95 Conclusion..............................................................................................................................97 5 PIERCING THE SEASONAL ROUND: USING STABLE ISOTOPES TO RECONST RUCT HUMAN DIET BY SE ASON AT WINDOVER AND HARRIS CREEK.................................................................................................................................109 Introduction................................................................................................................... ........109 Background...........................................................................................................................114 Site Descriptions............................................................................................................114 Windover (8BR246)...............................................................................................114
7 Harris Creek (8VO24)............................................................................................115 Previous Studies of Paleodiet at W indover and Harris Creek....................................... 116 Season of Site Occupation.............................................................................................116 Serial Sampling............................................................................................................. 119 Methods................................................................................................................................121 Results...................................................................................................................................121 Discussion.............................................................................................................................122 Conclusion............................................................................................................................125 6 CONCLUSION..................................................................................................................... 140 APPENDIX LIST OF REFERENCES.............................................................................................................148 BIOGRAPHICAL SKETCH.......................................................................................................164
8 LIST OF TABLES Table page 2-1 All individuals from the Harris Creek s ite (8VO24) sam pled in this research.................. 53 2-2 Conventional and AMS radiocarbon dates from Harris Creek.......................................... 54 2-3 Harris Creek faunal bone and enamel isotopic (13CPDB, 18OPDB) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW)........................................................................................................................... .........55 2-4 All Harris Creek human burials and dental isotopic 13CPDB, 18OPDB, 15NAIR values..... 56 2-5 Harris Creek local human bur ials and dental isotopic ( 13CPDB, 18OPDB, 15NAIR) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW).................................................................................................. 57 2-6 Harris Creek non-local (southern) hu m an burials and dental isotopic ( 13CPDB, 18OPDB, 15NAIR) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW)................................................................ 58 3-1 Windover and Harris Creek/Tick Island 13C and 15N ratios...........................................79 3-2 Isotopic endpoints for the Is osource linear m ixing model................................................. 79 3-3 Results of Isosource remixing model for Harris C reek and Tick Island............................ 79 4-1 Summary statistics for individua ls sam pled in this research............................................. 99 5-1 Conventional and AMS radiocarbon dates from Harris Creek........................................ 126 5-2 Season of site occupation of Ar chaic sites in the St. Johns Region. ................................ 127 5-3 Summary statistics from Windover (8BR246) and Harris Creek (8VO24) .....................128 5-4 Perikymata counts for selected hum an molars from Harris Creek.................................. 129
9 LIST OF FIGURES Figure page 1-1 Map of Florida with selected Archaic Period sites............................................................33 2-1 Location of the Harris Creek site....................................................................................... 59 2-2 A plot of 18O and 13C ratios of humans and fauna from Harris Creek........................... 60 2-3 Estimated 18O isopatches of river water for the state of Florida...................................... 61 2-4 Distribution of local and non-local burials sam pled in th is research and the location of AMS dated individuals..................................................................................................62 3-1 Adjusted 13C and 15N ratios of bone collagen a nd apatite from Windover and Harris Creek. .....................................................................................................................80 3-2 Difference in mean 13C ratios between apatite and collagen from Windover and Harris Creek.......................................................................................................................81 3-3 Regressions of protein 13C ratios.....................................................................................82 4-1 Serial sampling methods em ployed in this study. ...........................................................102 4-2 Appositional nature of enamel formation........................................................................ 103 4-3 Picture of tooth microstructure showing perikymata, enam el prisms and Striae of Retzius. ...................................................................................................................... ......104 4-4 Difference in 18O ratios recovered from modern and prehistoric human molars. ......... 105 4-5 Difference in 13C ratios recovered from modern and prehistoric human molars........... 106 4-6 Distribution of 18O ratios for Western Europe and the Eastern US............................... 107 4-7 Series of 18O ratios from each sampling method for the modern teeth.......................... 108 5-1 Milanich and Fairbanks (1980) model of m ovement between the river and coast......... 130 5-2 Russo and Ste. Claires (1992) model of m ovement up and down the river and coast by separate ethnic groups................................................................................................. 131 5-3 The location of Windover (8BR246), Ha rris Creek (8VO24) and other Archaic Period sites in the region. ................................................................................................. 132 5-4 The 13C ratios of bone collagen from selected prehistoric sites..................................... 133 5-5 Appositional nature of enamel formation........................................................................ 134
10 5-6 Picture of tooth microstructure showing perikymata, enam el prisms and Striae of Retzius. ...................................................................................................................... ......135 5-7 Sampling tracks on a human molar.................................................................................. 136 5-7 Series of 18O and 13C ratios from Windover. .............................................................. 137 5-8 Series of 18O and 13C ratios from Harris Creekindividuals 4, 8, 35, 36, 38, 46, 91, 100/101. ....................................................................................................................138 5-9 Series of 18O and 13C ratios from Harris Creekindividuals 108, 128, 132, 153, 141, 147/148, 170............................................................................................................139
11 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy ISOTOPIC INVESTIGATIONS OF ARCHAIC PERIOD SUBSISTENCE AND SETTLEMENT IN THE ST JOHNS RIVER DRAINAGE, FLORIDA By Bryan Duane Tucker August 2009 Chair: John Krigbaum Major: Anthropology This research uses stable isotopic analysis and models of human mobility to examine three common themes in archaeological studies of hunter-gatherers; egalitarianism, subsistence, and settlement patterns. Specifically, 13C, 15N and 18O ratios from human bone and teeth are used to evaluate human mobility and identify vari ability in patterns of so cial organization, diet, and settlement during the Archaic Period in the St Johns region of Florid a. Two Archaic Period sites, the Windover site (8BR246) and Harris Cr eek (8VO24), are assayed in this research. First, mortuary variation at Harris Creek is ev aluated as the result of differences in social status or as an outcome of the groups m obility strategy. Two competing hypotheses which explain differences between burials at Harris Creek are evaluated with 13C and 18O ratios from tooth enamel. The results of th e isotopic assay suggest the differe nces in mortuary treatment are not the result of mobility strategies or social status, rather the different burials may show different religious or ethnic preferences exhibited by immigrants to the region. Next, 13C and 15N ratios from human bone are used to assess the contribution of freshwater and marine resources to diet at Windover and Harris Creek. Culture historical models suggest a broad scale subsistence change occurred in the St Johns region between the Early and
12 Middle Archaic Periods when popula tions began to consume large am ounts of riverine resources (Milanich & Fairbanks, 1980). Based on the isotopic data, there is no evidence of a new and intense use of riverine resour ces; both populations relied on moderate amounts of riverine resources as component of their total diet. The results also suggest a marine dietary component was present at both sites which suggests a seasonal o ccupation of the coast. A serial sampling methodology is developed, tested and employed to determine the season of coastal occupation. Serial sampling studies are frequently employed to recover isotopic time series data from tooth enamel. Ho wever, concerns over isot opic attenuation caused by secondary mineralization have remained (Balasse, 2003; Kohn & Cerling, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). Fifteen hum an molars from four modern and eleven prehistoric individuals were a ssayed to develop a sampling methodology and to assess the effects of secondary mineralization. The results suggest that sampling the visible growth structures recovers the most isotopic vari ation and that secondary mineraliz ation is not a major source of isotopic attenuation for researchers sampling enamel carbonate. The sampling methodology suggested developed in the pilot study is used to sample molars from Windover and Harris Creek. Two se ttlement models have been suggested for Archaic Period in the St Johns region. The firs t suggests population occupi ed the St Johns River basin the warm months and move d to the coast or central high lands during the colder months (Milanich & Fairbanks, 1980). An alternative model suggests separate groups occupied the riverine and coastal zones year -round (Russo et al., 1992; Russo & Ste. Claire, 1992). The isotopic data suggest neither m odel is correct. Based on the 13C and 18O ratios and data from floral and faunal studies of the seasonal of site occupation, a ne w scenario is proposed suggesting a summer/fall occupation of the coast.
13 Finally, these results are considered as a w hole and their implications for archaeological models of the region are discussed. The data su ggest a dietary regime which included a summer use of coastal foods was in place as early as ca. 8000 BP and changed little over the next 5000 years. Despite the presence of large freshwat er shell mounds, freshwat er shellfish did not constitute a large portion of the annual diet which suggests th e appearance of shell mounds during the Middle Archaic was not a result of a dietary shift. Furthermore, the presence of extra local individuals in these mounds suggests the mounds were not territorial markers for local groups of isolated hunter-gathe rers; rather the mounds provided a place for contact between different groups in a wider networ k of socially connected groups.
14 CHAPTER 1 INTRODUCTION This study contributes to broad-scale research within an thropology where issues of individual and group mobility are addressed within the framework of hunter-gatherer anthropology. Mobility is a universal property of all humans (Close, 2000) and specific types of mobility may have played a significant role in human evolution (Brantingham, 1998). This research uses stable isotopes to examine hunte r-gatherer mobility at different geographic and temporal scales during the Archaic Period in the St Johns region of Florida. Specifically, stable isotope ratio analyses are used to evaluate hum an mobility by identifying variability in patterns of social organization, subsistence, and settlement patterns. Th e St. Johns River basin and the adjacent Atlantic coast have been the subject of archaeological study for over a century. The area is rich in prehistoric remains with sites ra nging from the Paleoindia n to the Mission Period (Milanich, 1998). Despite long-term study, ma ny aspects of hunter-g ather life during the Archaic Period remain unclear including the degree of mobility required to maintain social networks, harvest natural resources, or constr uct massive freshwater shell mounds found along the river. Advances in stable isotope geoche mistry provide a novel means to interpret mobility in the past and offer fresh pers pectives of old ideas in hunter-g atherer research and the Archaic prehistory of the Southeastern United States. Though hunter-gatherers have been the subjec t of anthropological study since the advent of the discipline, modern hunter-gatherer studies began in the 1960s with projects like the Harvard Kalahari Project (Lee & De Vore, 1968). Such projects were designed to study modern hunter-gatherers to model the behavior of our human ancestors. These projects approached the study of hunter-gatherers from an ahistorical evolutionaryecological pers pective which presumed these groups remained isolated and thus provided a glimpse of the human past (Stiles,
15 1992). Under this model, traits that make up the hunter-gatherer packag e such as residential mobility, seasonal population aggregation, egalita rianism, food sharing, and simple technology were human adaptations to an unpredictable environment. During the late 1980s and throughout the 1990s, the growing post-modern movement introduced a historical perspectiv e into hunter-gatherers studies (Lewin, 1988). This revisionist perspective was accompanied by an explicit recognition that hunting and gathering groups, like all humans, had a long and rich history that shaped the groups current behaviors. The revisionist scholars argued that the San, an African hunter-gathe rer group, were not prehistoric relics shaped by the environment, rather thei r lifeway had been formed by their history of political conflict and European colonialis m (Schrire, 1980; Wilmsen, 1983). Under this historical model, many of the tra its used to define hunter-gathere rs are viewed as a reaction to the dominant culture of aggressive neighbors rather than the result of environmental constraints. Hunter-gatherers of the ethnogr aphic present are best understood as components rather than antecedents of complex so cieties (Sassaman, 2004). The revisionist critique suggests ethnographically documented populations are unsuitable analogs for prehistoric groups. These scholars re cognize that recent hunter-gatherers are a subset of those that existed before the advent of farmi ng and that social organi zations not present today may have been common in the past (Deetz, 1968; Hodder, 1986; Schrire, 1984; Sealy, 2006; Wobst, 1978; Woodburn, 1980). Ar chaeological studies of comp lex hunter-gatherers have demonstrated that many of the defining traits traditionally associated with modern huntergatherers, like small group size, egalitarianism and residential m obility were not present in all hunter-gatherer groups in preh istory (Chapman, 2003; Price & Brown, 1985; Sassaman, 2004). In fact, a subsistence economy ba sed on wild food resources is no l onger directly associated with
16 any particular form of social organization, level of technology, labor relations, intergroup relations, or ideology (Sassaman, 200 4). The dissolution of the hunter-gatherer package of traits has called the category of hunter-gatherers into question with some advocating the abandonment of the category altogether (eg.Burc h, 1998:211). Considering the diversity present in modern and prehistoric groups, hunter-gatherers can be define d simply as people who acquire the majority of their food through hun ting, fishing, and gathering, with no a priori assumptions about other aspects of their soci al system (after Sassaman, 2004). Long histories of interaction with neighbori ng societies coupled w ith reduced diversity among modern hunter-gatherers suggest they are poor analogs for Pleistocene groups. In the absence of modern analogs, the only way to study hunter-gatherers in a world of huntergatherers, before animal husbandry or agricu lture, is through the study of archaeological remains. Though movement is ephemeral and seldom leaves any material trace in the archaeological record, archaeologists have deve loped several techniques to reconstruct human movement in prehistory (Close, 2000). These re constructions are frequent ly based on the source of raw materials used for stone artifacts (e.g. Amick, 1996) ceramics (e.g. Simms et al., 1997), lithic refit studies (Close, 2000), the frequencies of plant remains (Bonzani, 1997), or the types of houses and storage (Smith, 2003). These techni ques are seldom able to clarify the movement of individuals, though a few excepti ons occur under extraordinary ci rcumstances such as Closes (2000) refitting study and the footprints at La etoli (Leakey & Hay, 1979). Most archaeological studies of mobility avoid the individual and st udy broad scale patterns of mobility or they assess mobility in terms of its relationship to the organization of technology (Close, 2000). The archaeological study of mobility is further complicated by the imperfect relationship between movement and material culture (Kelly, 1992:21). Artifacts may be decoupled from the
17 individual rendering the movement of material goods independent from the movement of the individual. As a result, the materials goods may be cached for later use or traded down-the-line. Indicators of mobility in an individuals biological remains are not subject to the problems which complicate archaeological studies based on material culture. Unlike artifacts, biological markers of mobility cannot be decoupled from the individual. Studies of mobility using human remains have focused on two areas of research; biomechanical markers and stable isotope analysis. Bone is a living tissue which re acts to stresses from habi tual activities and thus records evidence of the activ ity in its morphology. Repeat ed activity can affect bones throughout the human skeletal system and the markers left by these forces can provide information about the life history of the indi vidual (Carlson et al., 2007; Larsen et al., 1991; Merbs, 1983; Wanner et al., 2007). For instance, La rsen and Ruff (1994) assert that a general reduction in the size of the femoral midshaft in Contact Period populations from the Georgia Bight is a result of a decrease in individual mobility related to a new focus on farming introduced by the Europeans. Stable isotopes are often employed in the study of prehistoric mobility. Light stable carbon and oxygen isotopes distinguish dissimilar e nvironmental regions a nd, unlike the analysis of artifacts, flora or fauna, a nd settlement patterns, stable is otopes provide direct evidence for mobility at the individual level of analysis (e.g. Montgomery et al., 2005; Price et al., 2000; White et al., 1998). Throughout this research, st able isotope abundances are reported in delta ( ) notation which expresses the ratio of heavy to light isotopes in a sample compared to a standard measure in parts per thousand (per mil or ) (Faure, 1986). Ratios of 13C vary between terrestrial and marine system s and, in the absence of C4 plant consumption, reliably assess the amount of marine foods in the diet. In terms of 18O ratios, environmentally similar regions have
18 similar seasonal 18O averages which fractionate in a pred ictable fashion according to latitude, altitude, and distance from the sea. The 18O values of precipitation are incorporated into the surface water of lakes and rivers and eventually into the biolog ical tissues, including the bones and teeth, of animals and humans (Longinelli, 198 4; Luz et al., 1984). Therefore, ratios of 13C and 18O recovered from these tissues ultimately refl ect the isotopic profile of the environment where the tissue formed and differences in these ra tios inform studies of mobility (Balasse et al., 2002; Bentley and Knipper, 2005; Evans et al. 2006; Hoogewerff et al ., 2001; Prowse et al., 2003; Turner et al., 2005; White et al., 1998, 2001). The independent data produced by the isotopic study of teeth are not subject to the same problems of equifinality as are archaeological st udies of mobility based on material culture or subsistence remains. Stable isotope ratios are locked in tooth enamel and, when combined with paleodemographic profiles, can identify variations in individual and group mobility caused by cultural factors such as post-marital changes in residence. Isotopic data have been used to inform and strengthen archaeological models base d on material culture. White et al. (1998), for instance, demonstrate the presence of ethnic en claves at Teotihuacan using oxygen isotopes. Additionally, stable isotopes have been able to demonstrate the diversity of past hunter-gatherers compared to modern hunter-gatherer groups. Se aly (2006) found prehistori c settlement systems very different from those documented ethnographi cally in the same region. Isotopic data from prehistoric coastal hunter-gatherers in the southe rn Cape of Africa, demonstrate the Cape was divided into territories occupi ed by separate hunter-gatherer gr oups (Sealy, 2006). These groups appear to have practiced an economic or perime ter defense (from Cashda n, 1983) rather than the social boundary defense which applies to most ethnographically documented societies in the region (Sealy, 2006).
19 Archaeological Models of Mobility This research uses stable isotopes to ev aluate human mobility during the Early and Middle Archaic Periods in the St. Johns River region of Florida. The concept of mobility is central to any archaeological or ethnoarchaeologi cal research that studies people who move(d) at a detectable rate (Close, 2000). Mobility is an important topic because changes in mobility are often accompanied by changes in food storage, trad e, territoriality, social and gender inequality, sexual division of labor, subsistence and demography, as well as cultural notions of material wealth, privacy, individuality, cooperation, and competition (Kelly, 1992). Furthermore, mobility is critical for a more nuanced understanding of the variation and patterning found in the archaeological record (Binford, 1980). As Close (2000:50) points out, there are few definitions of mobility. Kelly (1992:44) believes mobility is a property of individuals, who may move in many different ways: alone or in groups, frequently or infrequently, over long or short distances. Close (2000:50) cites Jochim (1976:24) who defines mobility as the potential distance nece ssary to travel per capture. However, mobility is typically associ ated with the act of m oving rather than the potential to move (Kelly, 1992:44) Therefore, Bamforth's (1 988:20) definition may be more appropriate, although referring to bis on, he equates the degree of mobility with the frequency of movements, the distan ce of movements, or the speed of movements. The concept of sedentism is related to and often confused with mobility. Historically, anthropologists considered sedentism and m obility at opposite ends of a continuum with typological categories ranging fr om nomadic to fully sedent ary (e.g. Murdock, 1967). This continuum was often broken into 4 groups:(1) fully migratory or nomadic bands, (2) seminomadic communities whose members wander in bands for at least ha lf of the year but occupy a fixed settlement during some season or seasons, (3) semisedentary communities whose
20 members shift from one fixed settlement to another seasonally or who occupy a single settlement and a substantial portion of the population depart s from seasonally to occupy shifting camps, and (4) relatively permanent settleme nts (Murdock, 1967:159). When sedentism is conceptualized as a point on a continuum of residential mobility, or as a category in opposition to a more mobile state, archaeologists collapse the different dime nsions of mobility (Kelly, 1983). As Kelly (1992:52) points out, sedentism ma y not reduce mobility; in fact logistical mobility is likely to increase as a group becomes more sedentary (Binford, 1980). It may be more productive to consider sedentism a threshold phenomenon de fined by all segments of a population using facilities within a settlem ent during all seasons of a year (Pl og, 1990) or when a portion of the population remains in a residential settle ment throughout the year (Rafferty, 1985). The juxtaposition of mobility and sedentism obscures the fact that mobility is universal, variable, and multi-dimensional (Kelly, 1992; Smith, 2003). The concept of mobility includes several dimensions: (1) individual mobility, or the movement of individuals independent of one another, (2) group mobility, or the coordinated m ovements of individuals, and (3) the frequency and magnitude of movement (Kelly, 1983; 1992). Movement is not equal for all individuals. Some individuals may move more than ot hers (e.g. men vs. woman, young vs. old), and movement also occurs on daily, seasonal, a nd annual scales (Kelly, 1992). Differences in mobility provide different options for dealing with specific foraging dilemmas; some foraging tasks are better accomplished by individuals, others may require group action (Brantingham, 1998). Of the multiple dimensions of mobility, re sidential and logistical are the most often discussed (Binford, 1980; Kelly, 1983; 1995). Binford's (1980) foraging and collecti ng model remains highly influential in archaeological discussions of mobility. Most stud ies of mobility are based on or in reaction to
21 Binford 's mobility typesfora ging/residential and co llecting/logistical. Binford (1980:10) describes two types of mobility strategies; a forager strategy where a group maps onto resources through residential moves and adjustment s in group size, and l ogistically organized collectors who acquire specific resources with specifically organi zed task groups. His foraging model shows seasonal residential moves among a series of resource patches. In addition to areas with patches, this model is applicable to areas wi th relatively even resource distributions such as tropical rainforests or other equa torial settings. (Binford, 1980). In a foraging system, foragers range out daily gathering food on an encounter basi s and return to their residential bases each afternoon or evening (Binford, 1980:5). Binford (1980) allows for considerable variability in the size of the mobile group and in the number of reside ntial moves that are made by the group in the course of an annual cycle. In terms of archaeological residues, a fora ger system is expected to produce two basic spatial contexts: a residential ba se and a location. A residentia l base is the hub of subsistence activities where foraging pa rties originate and where processing, manufacturing, and maintenance occur (Binford, 1980). A location is a place where resources are procured or where extractive tasks are performed (Binford, 1980:9). In settings with limit ed resources, patterns of residential mobility may be limited to a series of very restricted locations such as water holes, increasing the year to year redunda ncy in the use of particular residential camps (Binford, 1980). In general, foragers have high residential mobility and daily food procurement strategies. Residential mobility is divided into a nu mber of dimensions including demography, scheduling, frequency, stability, range and length /duration (Ames, 1991; Chatters, 1987; Kelly, 1983). The demography of a group is highly va riable and may change in response to the patchiness of resources. When resources become scarce, members of a group may not all move
22 at the same time or to the same new lo cation (Yellen, 1977). Sc heduling involves the organization of mobility patterns to deal with s carceness by adjusting to seasonal distributions of resources (Binford, 1983). Frequency refers to the number of reside ntial moves in a year and the duration of each occupation. Stability is the perm anence of a mobility pattern from year to year and results in the redundant use of the landsca pe over several years (Binford, 1983). Range involves the total distance covered in a year and the average distance traveled for each residential move (Binford, 1983). According to Kelly (1992; 1995), the length of occupation of a residential site is determined by weighing the cost and benefit of staying compared to the cost and benefit of moving. The durat ion of residential site occupation varies, ranging from shortterm occupation to year-round use of a location. The duration of site oc cupation also varies by season, often long-term occupations and limite d group movement occur during one season (e.g. winter/summer or rainy/dry) with high group mob ility during other times of the year. Huntergatherers are expected to move more frequen tly in marginal environments when resource productivity is low, unless sufficien t stores are available (Smith, 2003). In the collector or logistical mobility model, collectors organize task groups to leave the residential base for overnight fo rays to procure a specific type of resource. Unlike a foraging based system, collectors maintain stores of food for a portion of each year (Binford, 1980). In addition to residential bases and locations, collecto rs generate three additi onal types of sites: field camps, stations and caches. A field camp is a temporary base for a task group where the group sleeps, eats, and maintains itself while away from the residential base (Binford, 1980). A station is a site where task groups gather info rmation, such as observing the movement of game or other humans (Binford, 1980). Caches are ne cessary for collectors because procurement of resources by small groups for relatively la rge groups requires storage (Binford, 1980).
23 Binfords (1980) categories of residential (f orager) and logistical mobility (collector) have proven to be useful heuristic tools, but they should not be em ployed as dichotomous categories (Binford, 1980). Binford (1980) views th e categories as a graded series from simple to complex and suggests that logistical and resi dential variability be viewed as organizational alternatives which are employed in various comb inations. Most groups practice a complex blend of settlement strategies rather than occupying the poles of a continuum or a single category between sedentary and mobile (B inford, 1980; Lightfoot & Jewett, 1986; Pauketat, 1989; Varien, 1999; Warrick, 1988). Under most models, mobility is related to the quality, quantity and distribution of resources (Binford, 2001; Donald & Mitchell, 1994; Hitchcock, 1982; Kelly, 1983). Mobility allows hunter-gatherers to reduce risk by averag ing over variability a nd simply moving away from scarcity (Cashdan, 1992:237; Halstead & O'Shea, 1989; Kelly, 1983; 1992). However, factors other than the distributi on and collection of resources can affect mobility. For instance, the depletion of water and fuel and the accumulati on of trash, debris, and vermin can influence when and how often a group moves (Sm ith & McNees, 1999; Wandsnider, 1992). In addition to physical and economic demands, social dimensions also affect mobility. Mobility can be influenced by a need to seek sp ouses, allies, shamans, or move in response to sorcery, death, and political concerns (Griffin, 1989; Vickers, 1989). Kelly (1990) points out that residential mobility itself may be culturally valued. Formerly mobile hunter-gatherers often regret their sedentary nature and express a desire to move around in order to visit friends, to see what is happening elsewhere, or to relieve boredom (Kelly, 1992:48). Though social complexity and mobility are no longer directly correlated, changes in behavior and social organizationa l associated with agriculture and political complexity often
24 correlate with relatively sedent ary societies. Evaluating sede ntism at different social and temporal scales allows models of the adoption of agriculture and/ or the emergence of political complexity to be assessed (Gallivan, 2002) Traditionally, increasing sedentism and social/political complexity have been associated with the emergence of agriculture (Christenson, 1980:52-53; Cohen, 1977; Rindos, 1983). However, st udies of complex hunte r-gatherers suggest areas with a rich supply of f ood resources can support relatively sedentary societies with large populations and complex political and soci al systems (Erlandson, 2001; Moseley, 1992). Many complex hunter-gatherer societies subsist on aquatic resources (Erlandson, 2001; Moseley, 1975; 1992; Rick et al., 2001; Russ o, 1996). Abundant marine and freshwater resources are capable of supporting large sedentary populations (Binford, 1968:332-333; Moseley, 1975; Sealy, 2006) and their richne ss and availability may obviate many of the shortages that necessitate re sidential mobility (Bailey & Mi lner, 2002; Bailey & Parkington, 1988; Binford, 1968; Cohen, 1977). Binford (1990:137) reports that terrestrial hunters moved more frequently over longer distances than di d aquatic hunter-gatherers Aquatic foods are excellent sources of protein and can be harv ested by members of the population who are often excluded from subsistence procurement (ch ildren and seniors) (Russo et al., 1992). Coastal and riverine environments, like those of the St. Johns region, can support a variety of societies with levels of mobility a nd sedentism ranging from seasonally mobile to fully sedentary settlements with complex social a nd political organizations (Perlman, 1983:293; Widmer, 1988). Two models of settlement have been suggested for the St. Johns region during the Middle to Late Period. Milanich and Fa irbanks (1980) suggest th at people spent their summers along the St. Johns River and dispersed to the coast and/or the inte rior during the colder months. Russo and colleagues (1992) critique this model demonstrating that it is a synthetic
25 treatment of flawed data originati ng with researchers in the late 19th century. Russo and Ste. Claire (1992) propose an alternat ive scenario in which separate populations occupied the coast and river valley year-round. These groups may have been sedentary or may have practiced some form of residential mobility in their respective riverine or coastal zone. Delineating the types of mobility and the relati ve level of sedentism practiced by groups in the St. Johns region through time permits th e discursive relationship between settlement pattern and social/political complexity to be examined. A better understanding of this relationship could indicate how and why social complexity arose and what social and/or environmental conditions lead to the high levels of social complexity common today. Dissertation Outline This research reconstructs subsistence and settlement patterns as proxies to evaluate human mobility and identify variability in patte rns of social organization, subsistence, and settlement during the Archaic Period in the St. Johns region of Florida. Two well known Archaic Period sites in the St. Johns region of Florida are assa yed in this research: Windover (8BR246) and Harris Creek (8VO24). The Windover site is an Early Archaic cemetery located near Titusville, Florida where at least 168 human skeletons were recovered from the shallow but persistent pond ( Figure 1-1 ) (Doran 2002). Radiocarbon date s on human skeletal material indicate the interments o ccurred between 7,100 and 7,330 rcybp (Doran 2002). The Harris Creek-Tick Island site is a Mt. Taylor Period mortuary mound on Tick Island in the St. Johns River. Recent AMS dates on f our of the 175 burials recovere d from the site indicate the mortuary layers date to between cal. BP 6793 to 7179 [0.96] (X-9110) and 6482 to 6758 [0.98] (X-9111A), a range of cal BP 697 to 35 radiocarbon years (two sigma). First, isotopic and mortuary va riation in burials at Harris Creek is evaluated as the result of differences in social status or as an out come of the groups mobility strategy. Settlement
26 patterns are reconstructed at th e micro and macro level. The micro level analysis examines movement up and down the St. Johns River and betw een the river and the Atlantic coast. Macro level patterns are evaluated by quantifying the numb er of extra-local immigrants present in the populations. Then stable isotopes are used to identify and quantify the amount of freshwater and marine resources consumed during the Archaic Peri od, both in terms of annual diet at the level of the individual and on a temporally broader scal e by determining when human populations in the region began to make use of marine resources. After the presence of a marine component in the diet is established, a serial sampling methodology is used to evaluate the season of consumption. Finally, the results of these studies are combined to assess culture historical models of the Archaic Period in the St. Johns region. Variation in Burial Treatment Differences in burial treatment are often interpreted to reflect identity or social status of the deceased (e.g. Binford, 1971; Saxe, 1970 ) or tem poral/geo-spatial considerations of corpse disposal (e.g. Buikstra & Char les, 1999; Hofman, 1985). Both of these interpretations are particularly relevant for the popul ation interred at the Harris Creek site. Excavations recovered more than 175 burials which were divided into two groups; tightly flexed secondary burials and loosely flexed primary burials (Jahn & Bulle n, 1978). The difference between primary and secondary burials at Harris Creek has been in terpreted as the earlies t evidence for social differentiation among hunter-gatherers in No rth America (Aten, 1999:17 9). Alternatively, secondary burials are often interpreted as individuals who died while away from a preferred burial site and were processed and stored un til an appropriate time and place for burial (e.g. Buikstra & Charles, 1999; Hofman, 1985). Stable isotopes can be used to test if the differences in burials at Harris Creek reflect social differentiation or temporal/g eospatial aspects of corpse dis posal. Individuals of different
27 social status often have different diets which ma y be recorded in differe nt isotope ratios between burial types (Ambrose et al., 2003; White et al., 2001). Individua ls with higher rank or status, evidenced by secondary burials, may have experien ced greater access to protein sources, such as meat or seafood, than more loosely flexed burials possibly resulting in enriched 15N or 13C values. Differences in 18O values, which are derived primarily from drinking water are not expected if the differences betw een the burial types are rank or st atus-related. A lternatively, if the differences in secondary and primary burials are the result of a need for curation and transport, then no differences woul d be expected in the isotope ra tios of the two burial types. These hypotheses are tested with data derived fr om human molars from the Harris Creek site. The results of the isotopic a ssay indicate the majority of sa mpled burials at Harris Creek belong to two groups of people, one local to the site and one (or more) non-local group that grew up further south on the Florida peninsula. Th e entire local group received post-mortem processing as indicated by their secondary (tight ly flexed) burial. In comparison, only a portion of the non-local southern group received similar bur ial treatment. The presence of primary and secondary burials at Harris Creek does not appear to be related to curation or transportation of the dead. Instead the differences probably result from different cu stoms practiced by the different groups who came together at Harris Creek. Variation in Diet Stable isotopes are also used to reconstruc t the diet of the people interred at Windover and Harris Creek. The use of 13C and 15N ratios from human bone to reconstruct diet is now a common method of bioarchaeologica l investigation (Ambrose et al., 2003; Buikstra & Milner, 1991; Hutchinson et al., 2000; Katz enberg & Saunders, 2000; Kat zenberg et al., 1995; Newsome et al., 2004; Prowse et al., 2004; Richards et al., 2005; Schoeninger, 1989; Schoeninger & Schurr, 1998; Schwarcz et al., 2005; Turner et al., 2005; White & Schwarcz, 1989). Differences
28 in these values can indicate a reliance on mari ne foods or maize and differentiate between the relative trophic positions of the animals consumed. Though both Windover and Harris Creek are mortuary sites not habitation sites, neither is located on the coast and no evidence of marine food has been found at inland sites in the St. Johns region. Therefore, any evidence of marine food in the diets of the people interred at either site would represent so me level and type of mobility between the river valley and the coast. In addition to issues of diet and mobility, stable isotopes can provide evidence to evaluate the presence of monumental ar chitecture in the St. Johns re gion. Many researchers suggest Middle Archaic shell mounds middens resulting from a new and intensive use of riverine resources (e.g. Milanich 1994). Other resear ches assert these mounds are examples of monumental architecture which may or may not be composed of food remains (Randall & Sassaman 2005b). Using the human dietary data, th e assertion that mound c onstruction is related to a shift in subsistence economy is addressed. Isotopic ratios of 13C and 15N from Windover and Harris Cr eek/Tick Island suggest a minor shift in diet between the Early Archaic and Middle Archaic Periods. The results from Harris Creek suggest Mt. Taylor populations were not consuming large numbers of freshwater snails. Based on their 13C and 15N ratios and the estimated isotopic end member values of potential food resources in the region, peopl e at Harris Creek and Windover consumed large amounts of marine/estuarine fish. Since no major shift in subsistence accompanies the introduction of shell mounds in th e region, the appearance of she ll mounds in the region does not seem to be related to a shift in diet and ot her explanations for mound construction should be sought.
29 Variation in Stable Isotope Ratios Recorded in Human Tooth Enamel Most isotopic studies of dentition use a bulk sampling methodology to assess an average isotopic value for the period of enamel formati on. A serial sampling or isotopic zoning (Kohn & Cerling, 2002) approach samples discrete units of enamel represen ting different periods of time. Multiple samples in a sequence collected from a single individual provide time series data which allow isotopic variability to be studied. Secondary mineralization is a major concern in serial sampling st udies (Balasse, 2003; Kohn & Cerling, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). Enamel is only partially mineralized at deposition and secondary mineralizat ion occurs during the maturation phase as the mineral content of the tooth increases until the enamel is fully mineralized. This process extends the period of time over which minerals and isotop es are incorporated into a unit of enamel. Despite this concern, serial sampling methodologi es have been successfully employed in many isotopic studies of faunal dentition (Balasse et al., 2005; Fricke & O'Neil, 1996 ; Gadbury et al., 2000; MacFadden et al., 2004; Wiedema nn et al., 1999; Zazzo et al., 2005). However, this serial sampling methodology has not been frequently app lied to human remains (but see Fuller et al., 2003; Sponheimer et al., 2006). Before this technique can be used to recons truct the seasonal round fo r the inhabitants of Windover and Harris Creek, the effects of secondary mineralization on the serial sampling of human dentition must be determined. I must also establish that the isot opic variation recovered from human dentition reflects environmental inputs and is not the result of internal biological rhythms. To this end, a sample of 15 human mola rs from various contexts is assayed. Four of the fifteen molars are modern, two from Gaines ville, Florida and two from New Castle-on-Tyne, United Kingdom. The remaining eleven teeth are fr om four different prehistoric contexts, Harris Creek (Florida), the West Mouth of Niah Cave (Sarawak, Malaysia), Lobang Jeragan, (Sarawak,
30 Malaysia), and Gua Cha, (Kelan tan, Malaysia). Each molar wa s sampled using two methods; a micomill was used to sample the inner enamel along the project path of mineralization and a dental drill was used to remove samples from the surface of the tooth parallel to the growth structures (perikymata). The re sults of each method are compared to evaluate the effects of secondary mineralization and to evaluate wh ich sampling method recovers more isotopic variation. The results of this study suggest se rially sampling dentition from Harris Creek and Windover parallel to the growth st ructures will provide reliable dietary data with sub-annular resolution. Sampling the visible growth structures recove rs more isotopic variation than sampling parallel to the enamel prisms. These result s do not suggest secondary mineralization is a significant source of isotopic atte nuation in stable isotopic studies of tooth enamel carbonate. Based on the comparison of the modern and prehis toric teeth, no variation is caused by an internal biological rhythm; rather the prehistori c teeth reflect an attenuated record of seasonal variation apparent in 18O values. Variation in Diet by Season at Harris Creek and Windover Serial sampling techniques developed in the previous section are applied to human molars from Windover and Harris Creek to assess the relative amount and timing of marine food consumption. These data allow for the reconstr uction of the seasonal ro und and provide a means to evaluate mobility at the individual level of analysis. Isotopic samples from Windover and Harris Cr eek provide a diachronic view of dietary change in the St. Johns region during the Early and Middle Archaic. This diachronic view of dietary change and seasonal scheduling can provi de information about mobility patterns through time in the region. Changes in mobility and reso urce use may be correlated with changes in settlement patterns and the constr uction of shell mounds and ridges.
31 Data recovered from Windover and Harris Cr eek indicate these populations consumed more marine food during warm months than duri ng the cold months. These data support the bone isotope studies which indica ted a marine component in the diet of people from Windover and Harris Creek. Based on these results and the dist ance of these sites from the sea, it is likely these hunter-gatherer groups were residentially mobile and occupi ed the coast or an estuarine environment for at least a portion of each summer Conclusion This research approaches three themes co mmon in the archaeological study of huntergatherers to evaluate hunter-gatherer mobility: egalitarianism, subsistence, and settlement models. Though each chapter functions as an inde pendent module, they are united by an isotopic methodology and the use of mobility as an explana tion for the variability observed. In addition to the three main themes, other issues are ad dressed including ethnic ity, monumentality, and technical issues of tooth enamel mineralization and development. Taken as a whole, this research provides a significant new source of data to evaluate models of mobility, social diffe rentiation, subsistence and settlem ent in the St. Johns region of Florida. An isotopic approach is well suited to evaluate mobility at the level of the individual and avoid many issues of equifinality present in archaeological studies. Additionally, the serial sampling methods developed here demonstrate th at serial sampling human molars provides reliable data which show dietary and environmenta l change over a period of months rather than years. This increased resolution allows the seasonal round of hunter-gatherers to be assessed isotopically. This technique could be applied to questions in contexts far removed from the Southeastern United States. For instance, this te chnique can be used to clarify the long standing debate over the amount and timing of mari ne foods in human diets during the Mesolithic/Neolithic transition of Europe (Hedge s, 2004; Lidn et al., 2004; Milner et al., 2004;
32 Richards & Hedges, 1999; Rich ards et al., 2003). In a ddition to marine resources, 13C ratios from serially sampled human teeth could reveal the seasons in which significant amounts of maize were consumed. Detailed information on the seasonality of maize consumption correlated with evidence for long term storage facili ties could provide ne w insights into the Woodland/Mississippian transiti on in the Eastern Woodlands.
33 Figure 1-1. Map of Florida with selected Archaic Period sites.
34 CHAPTER 2 1ISOTOPIC INVESTIGATIONS OF MIDD LE ARCHAIC MORTUARY PRACTICES AT THE HARRIS CREEK SITE, TICK ISLAND, FLORIDA1 Introduction The rituals and customs surrounding death and burial are some of the oldest areas of study in anthropology and archaeology (e.g. Gluckm an, 1937; Kroeber, 1927) and remain of interest today (Garazhain & Yazdi, 2008 e.g. ; Horsley, 2008 ; Komar, 2008 ; Semple & Williams, 2007 ). Information about the particip ants, both living and dead, are encoded in the mortuary event, which has been interpreted to reflect a fashionlike phenomena (e.g. Cannon, 2005 ; Kroeber, 1927), the identity of the deceased (e.g. Binford, 1971; Saxe, 1970 ), the identity of the living participants (e.g. Parker-Pearson, 1999), or temporal /geo-spatial considerations of corpse disposal (e.g. Buikstra & Charles, 1999 ; Hofman, 1985). Human remains also contain information about the life history of the individua l, which can be recovered through a variety of macroscopic, microscopic, genetic and chemi cal analyses (Larsen, 1997). When integrated, aspects of the mortuary event and the life histor y of the deceased can combine for more complete and holistic analyses (see Go ldstein, 2006 for a discussion). In this study, we examine mortuary varia tion at an Archaic period mortuary site in present-day Florida. Stable is otopes provide an ecol ogical and geographical life history of the deceased, which may be integrated with mortuary data to produce a more nuanced understanding of burial practice. Specifically, we examine a su ite of isotopic data to evaluate previous interpretations of Archaic mortuary variation at the Harris Creek site on Tick Island. Harris Creek (8VO24) is a Middle Archaic freshwater shell mound in the St. Johns River basin of 1 This chapter is an early version of a manuscript in pr eparation for publication as: Tucker, B., Krigbaum, J. and Quinn, R. 2009. Isotopic Investigations of Mortuary Practice at the Harris Creek Site, Harris Creek, Florida. Submitted to American Antiquity.
35 northeastern Florida (F igure 2-1). Excavations during th e 1960s by Bullen and co-workers recovered more than 175 burials before shell mining destroyed the site (Jahn & Bullen, 1978). Differences among the burials have been interprete d as some of the earliest evidence for social differentiation among hunter-gatherers in North America (Aten, 1999). Stable isotopes bound in human tooth enamel provide an individual life history, including information about childhood residence and diet. Isotopic data can be framed into broader environmental context allowing for multi-scalar analyses not possible with conventional archaeological data. Isotopic ratios of 13C and 18O from human tooth enamel and 15N ratios from human dentin are used to investigate di fferences in diet, stat us/rank, residence and settlement pattern at Harris Creek. Finally, th e functional correlates of mound use are examined for the Archaic people of present-day north-central Florida. Mortuary Theory in Archaeology Mortuary theory allows inferences about the living to be drawn from the burial of the dead. These inferences include evaluations of social status (heterarchi cal vs. hierarchical) as well as mobility and settlement patterns. Additionall y, mortuary theory allows a more nuanced understanding of the role of the mortua ry events for the living and the dead. Status or Rank The Saxe-Binford approach remains an importa nt interpretive framewor k used to identify evidence of social status or so cial rank in mortuary data (Rakita & Buikstra, 2005). This approach assumes a relatively direct relationship between the social status of the deceased and the energy or wealth expended in the burial of the individual (Binfor d, 1971; Tainter, 1978). Elements of mortuary ritual such as comple xity of body treatment, construction/placement of interment facility, extent/durati on of mortuary ritual, material contributions to ritual, and evidence for human sacrifice are frequently asso ciated with social ra nk (Tainter, 1975). The
36 amount of corporate involvement and the degree of disruption the community experiences during the death of one of its members may correlate with the amount of labor or energy used in subsequent funerary rites (Tainter, 1973 ; Ta inter, 1978). Binford (1971) suggests that the number of social identities ritu alized at death varies according to the social position enjoyed in life and that funerary treatment varies with el ements of the social persona. For instance, biological sex could be distingu ished by grave orientation or a ssociated grave goods. Biological age is often expressed by variati ons in method of disposal, grave characteristics, patterning of the body, and/or placement of the grave. Group Mobility, Settlement Patterns and Mortuary Practices Differences between individuals within a buria l population may represent more than facets of an individuals social persona or identity. For instance, in seasonally mobile groups, death might occur a considerable distance from a preferred burial site; in such cases, the deceased could be buried where she/he died (i.e. away from the preferred site) or curated and transported back to a preferred site at a time of group aggr egation. Indeed, Hofman (1985) and Buikstra and Charles (1999) contend that differences betw een primary and secondary burial reflect the circumstances of the group at the time of an i ndividuals death, specifically the groups distance from its preferred burial site. In their respec tive studies, evidence from two Archaic sites, the Ervin site in Tennessee and th e Bullseye site in the Illinois River Va lley, suggests that individuals who died afar were bun dled and transported back to th e site as a part of each groups seasonal round. If timing and logistics were impor tant factors influencing burial, then variation in burial treatment and frequency of secondary burial could be informative about group organization, group mobility, and territory size (B uikstra and Charles, 1999; Hofman, 1985). In addition to issues of distance and timing, the dece aseds social persona could affect the groups decision to curate and transport the remains only a subset of individuals mi ght deserve or require
37 curation. In this model, cemeteries and mortuary mounds may act as seas onal aggregation sites, which serve economic, political, social, and ideo logical functions for seasonally mobile huntergatherers (Hofman, 1985). Burial rites may have played an important role in constituting and/or unifying many of these functions (Hofman, 1985). Mortuary practices also reflect aspects of circumscription an d territoriality. Saxe (1970) suggested that formal disposal areas for burial of the dead are maintained by corporate groups who use descent to legitimize their rights over crucial but restrict ed resources. Goldstein (1976) evaluated Saxes hypothesis using data culled from thirty ethnographic studies. She concurs that the maintenance of a permanent, specialized and bounded disposal area is a means by which a corporate group might seek to legitimize its right to control and claim resources. However, in what has become referred to as the Saxe-Goldste in hypothesis, Goldstein a dds that groups might symbolize this relationship in different waysthe absence of a formal cemetery does not preclude the existence of a lineal descent group (Goldstein, 1976 ). The Saxe-Goldstein hypothesis was re-evalu ated by Schroeder (2001), who addressed the issue of secondary disposal/bur ials using data from the Standard Cross-Cultural Sample of Murdock and White (1969). Schroeders sample consisted of 186 well-documented societies which were linguistically and cultu rally distinct. With regard to the Saxe-Goldstein hypothesis, Schroeder found that there is a relationship between crucial (probabl y fixed) resources and lineal descent groups. Schroeder (2001:90) notes that when this relationship is legitimated by lineal ties to ancestors, it may be reinforced by the ma intenance of formal disposal areas exclusively used for dead. However, the exact relations hip between crucial resources, lineal descent groups, and secondary disposal has yet to be elucidated (Schroeder, 2001).
38 Beyond Saxe-Binford That physical and circumstantial factors affect mortuary practices seems clear; however many researchers argue that these are secondary t o, and independent from, ideological and social factors (e.g. Carr, 1995; Hodder, 1984; Kroeber, 1927; Parker-Pearson, 1999). Rather than simply reflecting the identity of the deceased, thes e (and other) researchers believe that mortuary behavior is fluid and manipulated by the living to meet their own needs (Parker-Pearson, 1999). Some have reconfigured aspects of the Saxe-Bin ford approach, or have defined new approaches altogether (Barrett, 1990; Brown, 1995 ; Ch apman, 1995 ; Hrke, 2002; Lull, 2000 ; Morris, 1991 ; Parker-Pearson, 1993 ). New approaches include: the study of personhood (Gillespie, 2001), gender (Arnold & Wicker, 2001 ; Crown & Fish, 1996; Howell, 1995; Sullivan, 2001 ), and practice/agency theory (Barrett, 1990). Nevertheless, despite these critiques and reassessments, most researchers agree that the Sa xe-Binford approach provides a starting point for analyses and does not exclude other sym bolic or ideological aspects of mortuary investigation. Harris Creek Burial Mound Mortuary practices at Harris Creek, which include mound burial with differences in complexity of body treatment and extent/duratio n of mortuary ritual, have typically been associated with status or rank (Tainter, 1975). However, it is al so possible that the freshwater shell mound constituted a formal burial area used by a descent group to legitimate their rights to specific resources, as proposed by Goldstein (1976) and Saxe (1970) In this project, we use stable isotope ratios to ascertain the relative importance of social ranki ng and/or circumscription of resources by Middle Archaic groups at Harris Creek.
39 Previous Investigations at the Site Prior to its destruction, Harris Creek (8VO24) was composed of a series of shell middens and ridges, along with a substantial, bean-sha ped platform mound constr ucted of freshwater shell, estimated to have been 5.5 to 10.7 m above ground level (Aten, 1999). The site has a long history of investigation, starting with Moore in 1891, fo llowed by Bullen in 1961. Bullen conducted salvage excavations of the northeaster n slope of the mortuary mound. Although more than 175 burial features were recovered, firsth and accounts from people involved in the shell mining operations indicate that hundr eds of burials were destroye d by mining activities prior to Bullens mitigation (Aten, 1999). This suggests the original number of in terments at the site greatly exceeded those r ecovered by Bullen (Aten, 1999). Aten (1999) identified ten st ratigraphic layers within th e mound; the lowest layer was designated Layer 1. The majority of the burials were in layers 3 and 4. Layer 3, the white sand zone, was the earliest mortuary layer and c ontained two types of mo rtuary-related deposits: areas of white sand with burials and lenses of white sand with evidence of fires built atop them. Layer 3 also contained the black zone, a zone of dark, organi c, charcoal-impregnated soil with associated postholes. Following Jahn and Bullen (1978), Aten (1999) sugges ts that this zone represents the remains of a charnel house. Th e black zone in layer 3 is overlain by a more homogeneous Layer 4, constructed of freshwater snail shell (primarily Viviparus sp.) and a dark brown sandy matrix. Jahn and Bullen (1978) reported four conventio nal radiocarbon dates from samples taken from his Harris Creek excavations. Two of these bracket the mortua ry levels, one at the base of level 3 and one well above the mortuary layers in layer 8 (Aten, 1999). Charcoal at the base of layer 3 produced a radiocar bon date of 5450 rcybp (M-1264) and charcoal from layer 8 produced a date of 5320 rcybp (M-1265) (Aten, 1999). Based on these dates, the use range
40 of the two mortuary components could have been as great as 630 radiocarb on years or as short as 30 radiocarbon years (based on 1 deviations). Once cal ibrated (CALIB 5.0.1, 2 ) (Stuiver & Reimer, 1993), Bullens dates range between 6912 and 5112 years BP suggesting mound use could have exceeded 1800 years. Burial Form Deceased individuals at Harris Creek were primar ily interred in layers 3 and 4 of the shell mound. Burial features typically contained one to three indivi duals, although one mass grave included at least 11 individuals (Aten, 1999). Aten (1999) identif ied three variations of burial treatment. Some burials were placed on a small fi re in the bottom of the pit before the grave was filled with white sand; the aforementioned burial of 11+ individuals was in this category. Postholes in and around this grav e suggest it was surrounded by a fence or structure of some kind. Other burials involved sand with no evidence of fire, the interred body being covered with a layer of white sand and shell midden. Finally, some burials were simply placed in a pit and covered with shell midden. Body Positioning The majority of the burials (98%) recovered from Harris Creek were flexed and consisted of two typesloosely flexed, in which the indi vidual was lying on his or her side, and tightly flexed, in which the individual was interred in a vertical or seated posit ion (Aten, 1999). Aten proposed that the loosely flexed burials were interred shortly af ter death (primary burials) and that the tightly flexed vertical burials were in terred after an extended period of exposure above ground (secondary burials). The tig htness of the vertically flexed burials suggests they were partially defleshed prior to burial. Additionally, some bones in tightly flexed burials were associated with the remains of small fires a nd some bones show eviden ce of charring, indicating that these bodies were desiccated and partia lly decomposed prior to burial (Aten, 1999).
41 Stable Isotope Analysis Stable isotopes are commonly used to inves tigate the diet and move ment of animals and humans (e.g. Ambrose et al., 2003; Balasse et al., 2002; Buikstra & Milner, 1991; DeNiro & Epstein, 1978; Dupras & Schwarcz, 2001; Hutc hinson et al., 2000; Katz enberg & Saunders, 2000; Lee-Thorp, 1989; Montgomery et al., 2005; Price et al., 2002; Price et al., 2000; Richards et al., 2001; Schoeninger & Moore, 1992; van de r Merwe, 1982; White et al., 1998). Following convention, the abundance of stable isotopes are reported in delta ( ) notation, which expresses the ratio of heavy to light isotopes in a sample compared to that of a known standard, measured in parts per thousand (per mil or ) (Faure, 1986) using the formula: = [(Rsample/Rstandard) 1] x 1000 where R = 13C/12C, 18O/16O, 15N/14N The standard employed for carbon ( 13C) and oxygen ( 18O) is Pee Dee Belemnite (PDB) or its modern equivalent (NIST 19), and for nitrogen ( 15N) the standard is AIR. Carbon Isotopes In terrestrial systems, stable carbon isotopes vary by photosynthetic pathway, C3, C4, and CAM. C3 plants have 13C ratios from to (avg = .5) and C4 plants have 13C values that range from to (avg = .5) (Smith, 1972; Smith & Epstein, 1971). Edible CAM plants have a range of 13C values that partially overlap those of C3 and C4 plants. CAM plants are not a substantial concern in the diet of prehisto ric inhabitants in eastern North America, although some prickly pear has been re covered from Archaic s ites in Florida (Doran, 2002; Newsom, 1994; Tuross et al., 1994). Stable carbon isotopes also discriminate between terrestrial C3 biota ( 13C values between to 20) and marine systems ( 13C values between to ) (Schoeninger & DeNiro, 1984; Smit h, 1972). In the absence of significant C4 or CAM resources, i,e. in the Arch aic southeastern United States, 13C values can be used to differentiate terrestrial and ma rine diets and the values have been employed in numerous
42 paleodiet studies (Ambrose & DeNiro, 1986; Druc ker & Bocherens, 2004; Dupras et al., 2001; Hutchinson, 2002; Katzenberg et al., 1995; Keenleyside et al., 2006; Richards et al., 2002; Schoeninger & Schurr, 1998; Schwarcz et al., 2005; Tuross et al., 1994; Vogel & Van de Merwe, 1977; White et al., 1998; Wright & Schwarcz, 1998). Oxygen Isotopes The 18O of surface water are determ ined in large part by temp erature, precipitation, and evaporation. The 18O of precipitation is determined by th e ambient temperature and the amount of precipitation. Generally, warmer weather results in enriched 18O values and cooler weather in decreased 18O values (Bryant & Froelic h, 1996). In temperate regions, which have relatively constant rainfall paired with vari ed temperatures throughout the year, 18O values are enriched during the summer months (Gonfian tini, 1985). In tropical regions which have varied rainfall and more constant annual temperatures, 18O values track the amount of rainfall (Njitchoua et al., 1999). Environmentally similar regions have similar seasonal 18O averages which fractionate in a predictable fashion according to latitude, al titude, and distance from the sea. The 18O values of precipitation are incorporated into the surface wate r of lakes and rivers and eventually into the biological tissues of the animals and humans (Longinelli, 1984; Luz et al., 1984). When significant differences in 18O ratios exist between regions, th ese differences can be used to assess migration and immigration in archaeological contexts (B alasse et al., 2002; Bentley & Knipper, 2005; Evans et al., 2006; Hoogewerff et al., 2001; Prowse et al ., 2003; Turner et al., 2005; White et al., 2001; White et al., 1998) ; for example, White et al. (1998) use 18O values to identify and delineate immigrant et hnic enclaves at Teotihuacan.
43 Nitrogen Isotopes Protein consumption and rela tive trophic level of the sp ecies consumed is commonly measured in archaeological assemblages with 15N values of collagen extracted from bone and dentin. The lighter isotope is more easily in corporated into metabolic processes such as ammonia excretion, re sulting in loss of 14N and enrichment of organism tissue in 15N. At the base of the trophic pyramid, nitrogen-fixing flor a such as legumes are typically lower in 15N values and may approach 0 (Shearer & K ohl, 1986). These and other primary producers supply herbivores with low 15N values. As enrichment ensues with excretion, and primary and secondary carnivores consume herbi vores, they take on the enriched 15N resulting in a stepwise function up trophic level (Koch et al., 1994). Additional trophic spaces in marine ecosystems result in higher 15N values relative to terrestrial and freshwater ecosystems (Schoeninger & DeNiro, 1984; Schoeninger et al., 1983). Marine plants 15N ratios are approximately 4 enriched relative to terrestrial plants. Materials and Methods We utilized a subset of Harris Creek burials previously analyzed for 13C, 15N and 18O from tooth enamel and dentine (n = 39; Quinn et al., 2008). We choose burials with the most complete mortuary information and provenience data available (n = 21; Quinn, 1999). We did not include four individuals pr eviously interpreted as immigr ants from coastal and northern regions because these extreme outliers obscured variation present in the remaining sample (Quinn et al., 2008); rather, we focused on indivi duals believed to reside within the St Johns River valley. We enhanced th e dataset with additi onal carbon and oxygen is otopic analysis of tooth enamel from 9 burials, yielding a total of 48 individuals. We denot ed vertically flexed burials as Secondary and loosely flexed burials as Primary, reflecting their delayed or relatively rapid interment after death, respectivel y. Table 2-1 lists the bur ials with associated
44 mortuary information. The combined sample consists of 36 (75%) secondary burials and 12 (25%) primary burials. These ratios approximate the ratios of burial types recovered from the site: ca. 60% tightly flexed (s econdary) and ca. 38% loosely flexed (primary) burials (Aten, 1999). To establish a local isotopic baseline, bone ap atite from 6 deer and enamel from 1 raccoon from the Harris Creek site were assayed for 13C and 18O. Additionally, in an attempt to gain better chronological control over the mortuary layers at Harris Creek, four human bone collagen samples were AMS dated. Two individuals were sampled from layer 3 and two from layer 4. For this study, enamel and dentin were preferred over bone for several reasons. First, tooth enamel is less likely to suffer diagenetic altera tion than bone apatite (Koc h et al., 1997; NielsonMarsh & Hedges, 2000). In addition, enamel a nd primary dentin are not remodeled during an individuals lifetime. Finally, sampling dentin and enamel from the same tooth provides information on different aspects of the diet at the same point in tim e in an individuals life. The 13C ratios in enamel record whole diet, while 13C ratios in dentin preferentially reflect the protein component of the diet (Ambrose & Norr, 1993). Each tooth records the diet consumed during the period of enamel formation. Mineralization and crown formation in human mo lars has been estimated by Smith (1991): first molar (0.1 to 2.5 years after birth), the second mo lar (3.8 to 6.8 years afte r birth), and the third molar (9.5 years to 12.4 years after birth). Bulk samples taken from the surface of the tooth do not encompass the entire period of crown forma tion. Due to the incremental nature of tooth growth, a portion of the enamel growth layers are enveloped in the appositional zone under the cusps and do not reach the surface (Dean & Benyon, 1991). Therefore, though crown formation
45 spans a period of ca. 2.5-3 years, each bulk sample contains only an average of around one to two years of growth, depending on the amount of tooth wear and which molar is sampled. For stable isotopic analysis, the nine additional tooth sample s were ultrasonically cleaned in distilled water and air-dried. Tooth surfaces were inspected under magnification and lightly abraded to remove remaining debris using a lo w speed dental drill (B rasseler UG12) outfitted with a carbide bit (Brass eler #170). Under magnification, a single sample (ca. 1.0 mg) per tooth was taken along the axis of growth. Sample powde rs were not chemically pretreated. Enamel was analyzed using a Micromass PRISM mass sp ectrometer interfaced with a multiprep device. Analytical precision is estimated at better than 0.05 for 13C and 0.1 for 18O. Repeat analyses of three samples from Quinn et al. ( 2008) showed insignificant differences in both isotopes (e.g. Passey et al., 2002). We report 13C values of enamel relative to PDB. Th ese values are enriched compared to the diet 13C values by approximately 9-14 per mil (Lee-Thorp, 1989). The 15N values of collagen from dentin are reported relative to air a nd are enriched by 1-5 per mil relative to diet 15N values (Ambrose & DeNiro, 1986; DeNiro, 1985). We included only 15N values from collagen that yielded C:N values between 2.9-3.6 (Ambrose, 1990). We converted 18OPDB values to Standard Mean Ocean Water (SMOW) after Coplen et al. (1983) and then converted 18OSMOW values to 18O of ingested water ( 18OIW) (after equations derived by Cormie et al., 1994; Iacumin et al., 1996; Luz & Kolodny, 1985; Luz et al., 1984). The 18OIW values from the faunal sample were used to define a range of local values for the vicinity of Harris Creek. Results Accelerator Mass Spectrometry Dates New accelerator mass spectrometry (AMS) data for four human burials from Harris Creek are reported in Table 2-2. Tw o individuals from layer 3 produ ced calibrated (CALIB 5.01) 2
46 sigma dates of cal BP 6793 to 7179 [0.96] (X-9110) and cal BP 6742 to 7030 [0.90] (X9112RA). Two individuals from layer 4 are slightly younger in age at cal BP 6598 to 6891 [0.97] (X-9109A) and cal BP 6482 to 6758 [0.98] (X-9111A). These new AMS dates are uniformly earlier than dates reported by Bulle n (Crane & Griffin, 1965). Based on these new dates, Harris Creek mortuary layers 3 and 4 span a time range of cal BP 697 to 35 radiocarbon years (2 ). Stable Isotopic Values of Faunal Bone Apatite and Local Water Bone from seven deer ( Odocoileus virginianus ) and tooth enamel from one raccoon ( Procyon lotor ) was sampled to establish the range of 18O values for local water. Bone apatite returned 18OPDB values ranging from -3.1 to -0 .3, with an average of -2.0 0.8. Once converted to ingested water (IW) values, the 18OIW values range from -4.5 to -0.8 with an average of -3.0 1.1 (Table 2-3). One deer outli er is considerably enriched (18OIW = -0.8) compared to the remaining samples and is likely non-local in origin. With the outlier removed, the average 18OIW value for deer bone apatite is -3.0% 0.7. Deer are browsers and their 18OIW values are often enriched re lative to local water sources because deer derive a large amount of their water from consumed leaves, which are enriched relative to local water so urces due to evapotransporation (Luz et al., 1990 ). However, this effect is most pronounced in areas of low relati ve humidity where isotopic enrichment of 18O can be as high as 9.7. Florida has high relative humid ity and thus low evapotransporation minimizing 18O enrichment in leaves (Luz et al., 1990 ). The omnivorous raccoon is an obligate drinker and typically consumes surface water while drinking and eating. Its 18OIW value therefore should constitute a robust measure of local water values. The raccoon has a 18OIW value of -3.6, which is 0.2 depleted compared to the average deer 18OIW values of -3.0, supporting
47 minimal enrichment in deer. Based on the fa unal values, and the likelihood of a slight enrichment in the deers 18OIW values, we estimate the range of 18O values for local water sources in the vicinity of Ha rris Creek to be -2.5 to -4.0. Quinn et al. (2008) modeled 18O values for Florida rivers adjusting modern 18O values by -1 to account for a 5C higher average summe r temperatures in the mid-Holocene (Jones et al., 2005). However, despite the evidence for increased summer temperature and increased seasonality, the mean faunal 18OIW from Harris Creek suggest av erage yearly temperature stabilized in the region as ear ly as 6,000 rcybp. Accordingly, in this work we use modern 18O isopatches rather than the -1 adjust ment reported in Qu inn et al. (2008). Stable Isotopic Values of Tooth Enamel A total of 48 human enamel samples returned results for 13C and 18OIW analyses (Table 2-4). 13C values ranged from -8.7 to -12.8 with an average of -11.1 1.0. The 18OIW ranged from -0.3 to -4.2 with an average of -2.2 0.9. Twenty-three dentine samples were assayed for 15N with values ranging from 15.3 to 10.8% and an average of 13.4 1.1 Based on the range of local water 18O values estimated from the faunal samples reported above, we define two groups, one local and one non-local (Figure 2-2). The local group encompasses 18 individuals and the non-local gr oup included the remaining 30 individuals. The local group (n=18) has a 18OIW value range from -2.5 to -4.1 with an average of -3.1 0.5 (Table 2-5). The 18OIW values of the non-local group (n=30) range from -0.3 to -2.5 with an average of -1.7 0.7 (Table 2-6). The 18O values of the designated groups are significantly different (ANOVA, p=.0001). The enriched 18OIW values of the non-local group, based on Florida geography and modeled 18O distribution of Florida wa ters (Dutton et al., 2005 )
48 suggests a more southern origin, perhaps near th e head waters of the St. Johns River (Figure 23). The 13C values for the local group range from .1 to .5 and average .7 0.6 (Table 2-5). The 13C values of the non-local group are more enriched, ranging from -8.7 to 12.8 with an average of -10.7 1.0 (Table 2-6). The two groups are significantly different in their 13C values (ANOVA, p=.0005). In terms of tooth dentine, the local group (n=8) has 15N values that range from 15.2 to 10.8 and an average of 13.6 1.4 (Table 2-5). The non-local group (n=15) has 15N values that range from 15.3 to 12.0 and an average of 13.2 0.9% (Table 2-6). The two groups are not significantly different in 15N values (ANOVA, p=.5774). We found no statistical differences in any of the three isotopic syst ems when separated by burial layer (Figure 2-4). Discussion Rank and Status Aten (1999) suggested differen ces in burials at Harris Creek may reflect difference in social rank. In many societies, elite individuals enjoy access to privileged foods; therefore we hypothesized that differences in diet between high er and lower status i ndividuals in the Harris Creek mound might be manifested in varying isotope ratios (Ambrose et al., 2003; White et al., 2001). Individuals with higher rank or status, evidenced by ex tended mortuary resulting in tightly flexed burials, may have experienced diffe rential access to protein sources, such as meat and marrow, than the more loosely flexed burials. This would result in higher 15N values. In addition, high status access to trad ed food resources from non-local environments such as coastal regions might yield less negative 13C values. Because 18O values are derived primarily from
49 drinking water, rank or status-related differences are not manifested in 18OIW values of tooth enamel. Significant differences in 13C values (ANOVA, p = 0.04) and 18OIW values (ANOVA, p = 0.01) between primary and secondary burials at Harris Creek suggest these individuals consumed food and water from diff erent sources. Differences in 18O of drinking water are not consistent with rank related dietary differences. Instead, significant differences in 18OIW ratios from human molars suggest differe nt areas of childhood residence. All primary (loosely flexed) burials fall within the non-local range of 18OIW values, suggesting these individuals spent their childhood south of the Harri s Creek site (s ee Figure 2-2). The local group is composed entirely of secondary burials. There are no st atistically significant differences in 13C, 15N, or 18OIW values between primary and secondary burials within the non-local group, suggesting no signif icant dietary differences be tween the burial types when place of origin is controlled for. Additionally, though there are significant differences in 13C values between the local and non-local groups, there are no si gnificant differences in 15N values, suggesting that there is no significant difference in the trophic level of the dietary protein between groups. We propose that the isotopic evidence does not support status or rank related differences in diet between the primary and secondary burials, rather the differences are related to region of childhood residence. A more detailed study of differences in burial treatm ent at Harris Creek within the non-local southern group will be possible af ter ongoing detailed demographic work on the assemblage is completed. Settlement and Mobility Isotopic evidence reported above suggests the population inte rred at Harris Creek includes people who were born into at least two differe nt groups, one local and one non-local, perhaps
50 from southern portions of the Florida peninsula. The local group seems to have been fairly sedentary, since their 18O(IW) values correlate with 18O(IW) values of local fauna, and they have a more limited range of 13C values indicating a more restrict ed range of food resources. The non-local southern group, in cont rast, shows a broad range of 18O(IW) values, which based on modeled 18O(IW) ratios of river water are inferred to ra nge south of Harris Creek to the Lake Okeechobee area (Figure 2-3). The southe rn group exhibits a wider range of 13C values compared to the local group, which may reflect va riations in foraging acr oss a large area that could have included the upper (s outhern) St. Johns River, th e Lake Okeechobee area, and the Atlantic coast. The isotopic and burial data presented for Harris Creek do not suppor t the contention that different burial types at a mortua ry site reflect differential as pects of group mobility, as argued for other Archaic groups by Hofman (1985) and Buik stra and Charles (1999). According to such a scenario, groups who spend more time away from the mound would have more secondary burials than groups who spent more time near the mound. Instead, at Harris Creek, the difference in primary and secondary burials suggest s differences in social and ideological factors rather than energetic/physical f actors of curation and transporta tion. We suggest the local group spent their childhood in the vicinity of Harris Creek and were inte rred as adults and we assume they lived in the local area throughout their live s. If primary and secondary burials were a function of the seasonal round and distance from the mound at time of death, more primary burials would be represented in this group. Instead, all cases of primary burial occur within individuals in the non-local sou thern group. For primary burial to have occurred, these people must have been near the mound when they died as adults. Therefore, it is more likely these people joined the local group as adults. Howeve r, not all non-local indi viduals were primary
51 burials; many received the same extended mort uary processing as the local people. One interpretation is that local individuals enjoy a higher status due to their proximity to the mound, and only some southern individuals had or were able to gain such status prior to burial. An alternative explanation is that a portion of the southern group retained alternative religious beliefs and did not require the same type of mortuary treatment as the local population or religiously/ethnically converted non-locals. Spatial analysis of the Harris Creek burials does not reveal any pa tterning in terms of burial treatment, burial layer, 18OIW, 13C, or 15N values (Figure 2-4). Lack of apparent patterning supports an interpretation of a singl e burial population includi ng people from different childhood origins who joined the lo cal group as adults but retained different ethnic/religious beliefs. Function of Mounds Mounds are frequently cited as gathering pl aces for seasonally m obile groups, where ceremony and ritual reinforce alliances and a sh ared sense of identity or common ancestry (e.g. Gibson, 2006; Hofman, 1985; Russo, 1994; Saunders, 1994). The act of mound construction and maintenance constitutes shared r itual by a community that could pl ay a key role in maintaining the identity of these groups. Reinforcing or cr eating a shared sense of identity may be important when large numbers of immigrants are present. In addition to the non-local southern group, there are other immigrants from coastal and mo re northerly regions identified at Harris Creek (these were not included in the present study) (Quinn et al., 2008)1. Based on 18OIW values and 87Sr/86Sr ratios, at least 8% of the burials tested at Harris Creek containe d individuals who were not from the local or non-local southern group (Quinn et al., 2008). This percentage represents the minimum number of immigrants in the population, since some immigrants may not be isotopically distinct. If larg e numbers of immigrants are pr esent in the pop ulation, then
52 communal acts that reinforce group identity such as mound building, feasting, or burial rights, could be an important means of maintaining group unity and in tegrating immigrants into the local population. Conclusion The majority of the burials at Harris Creek belo ng to two groups of people, one local to the site and one (or more) we suggest grew up furthe r south on the Florida peninsula, perhaps as far south as the Lake Okeechobee area. Interestingl y, all people who spen t their childhood in the vicinity of Harris Creek were subjected to post-mortem processing as indicated by their secondary (tightly flexed) buria l. In comparison, only a portion of the non-local southern group received similar secondary treatment. The differences in burial treatment among the nonlocal southern group may be evidence of rank or status among the buria ls, but this is not supported by isotopic differences in diet. In contrast to the models proposed by Hofman (1985) and Buikstra and Charles (1999), the presence of primary and secondary burials at Harris Creek does not appear to be related to curation or tran sportation of the dead. Instead the differences may be evidence of multiple ethnicities or re ligions among those people who join the local population during adulthood. Finally, the large number of non-local people interred at the site suggests an important function of mortuary mounds may be to create and reinforce a shared sense of identity or common ancestry though communal actions and efforts.
53 Table 2-1. All individuals from the Harris Creek site (8VO24) sampled in this research. Burial Modes: F/V= Flexed Vertical, F/R or F/ L = Flexed on Right or Left side, E/B = Extended on Back. Burial No. Tooth Burial Layer Body Position Primary vs. Secondary 4 LM3 3 F/V Secondary 6 LM3 3 F/V Secondary 8 LM2 3 F/R Primary 11, 4 LM3 3 F/R? Primary 16, 23 LM3 4 F/R Primary 19 RM3 3 E/B Primary 25 RM3 5 F/V Secondary 29 LM3 5 F/V Secondary 35 RM3 4 F/V Secondary 36 LM3 4? F/R Primary 38 RM3 4 F/R Primary 46 RM3 4 F/V Secondary 49 LM3 ? ?/S Secondary 65,66,68 LM3 3/4 F/V Secondary 67 RM3 4? F/V Secondary 70 RM3 4 F/V? Secondary 74,75,76 LM3 4 F/V Secondary 83 RM3 3 F/R Primary 88 LM3 3 F/V? Secondary 89 LM3 3 F/V? Secondary 90 LM3 4 F/V? Secondary 91 RM3 4 F/V Secondary 92 RM2 4 F/V Secondary 93 LM3 5 F/L Primary 95 LM2 3 F/R Primary 102,103 LM3 3 F/V? Secondary 105,106,107 RM3 3 F/V Secondary 108 LM3 3 F/L Primary 111 LM3 ? F/V Secondary 113 LM3 4? F/V? Secondary 114,117a RM3 4? F/V Secondary 114,117b LM3 4? F/V Secondary 115 LM3 4? F/V Secondary 123 RM3 3 F/R Primary 132 RM3 3 F/V Secondary 134 RM3 ? F/V Secondary 135 LM1 3? F/V Secondary 137 LM3 ? F/V Secondary 139,140b LM3 3 F/V Secondary 141 RM2 4 F/V Secondary 143 RM3 4 F/V Secondary 147,148 LM3 4/5? F/V Secondary 154,155,156 LM3 3 F/RorL Primary 158 LM3 4 F/V Secondary 159 RM3 4 F/V Secondary 165,166,167 RM3 3/4? F/V Secondary 170a RM3 3? F/V Secondary 170b RM3 3? F/V Secondary
54 Table 2-2. Conventional and AMS radiocarbon dates from Ha rris Creek. New AMS dates are calibrated with CALIB 5.0.1 (Stuiver & Reimer, 1993). Lab No. Layer Sample Uncorrected 14C years B.P. Corrected 14C years B.P. Calibrated Dates, 2 sigma BP Reference M1264 3 charcoal 5450300 [5590 BP:6912 BP] 0.999 [6921 BP:6929 BP] 0.001 Aten 1999 M1265 8 charcoal 53200 [5645 BP:6494 BP] 1.00 Aten 1999 M1268 3 charcoal 54500 [5767 BP:5806 BP] 0.01 [5890 BP:6652 BP] 0.99 Aten 1999 M1270 7 or 8 marine shell 503020 5430 [5081 BP:5095 BP] 0.02 [5112 BP:5587 BP] 0.98 Aten 1999 X9110 3 human bone (Burial 7) 6125 [6793 BP:7179 BP] 0.96 [7195 BP:7244 BP] 0.04 New date, this paper X9112RA 3 human bone (Burial 31) 6053 [6742 BP:7030 BP] 0.90 [7043 BP:7069 BP] 0.02 [7077 BP:7086 BP] 0.01 [7096 BP:7156 BP] 0.07 New date, this paper X9109A 4 human bone (Burial 3) 5904 [6563 BP:6592 BP] 0.03 [6598 BP:6891 BP] 0.97 New date, this paper X9111A 4 human bone (Burial 9) 5825 [6482 BP:6758 BP] 0.98 [6761 BP:6782 BP] 0.02 New date, this paper
55 Table 2-3. Harris Creek faunal bone and enamel isotopic ( 13CPDB, 18OPDB) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW). Species Sample material 13CPDB 18OPDB 18OSMOW 18OIW White-tailed deer ( Odocoileus virginianus ) bone apatite -9.7 -3.1 27.7 -4.5 White-tailed deer ( Odocoileus virginianus ) bone apatite -11.1 -2.6 28.2 -3.9 White-tailed deer ( Odocoileus virginianus ) bone apatite -11.0 -2.1 28.8 -3.2 White-tailed deer ( Odocoileus virginianus ) bone apatite -13.4 -1.8 29.1 -2.8 White-tailed deer ( Odocoileus virginianus ) bone apatite -10.7 -1.8 29.1 -2.8 White-tailed deer ( Odocoileus virginianus ) bone apatite -10.6 -1.7 29.1 -2.7 White-tailed deer ( Odocoileus virginianus ) bone apatite -9.6 -0.3 30.7 -0.8 Raccoon ( Procyon lotor ) enamel -9.9 -2.4 28.4 -3.6 Mean -10.7 -2.0 28.9 -3.0 Std. dev. 1.2 0.8 0.9 1.1
56 Table 2-4. All Harris Creek human burials and dental isotopic 13CPDB, 18OPDB, 15NAIR values Burial No. Primary vs. Secondary 13CPDB 18OPDB 15NAIR 4 Secondary -10.5 -1.2 12.8 6 Secondary -10.1 -1.5 14.9 8 Primary -10.2 -0.9 13.1 11, 4 Primary -11.3 -0.7 13.0 16, 23 Primary -11.7 -1.5 12.6 19 Primary -10.5 -1.6 12.8 25 Secondary -12.4 -2.2 29 Secondary -10.9 -1.2 35 Secondary -11.0 -1.7 36 Primary -10.4 -0.5 38 Primary -8.9 -0.4 46 Secondary -11.8 -2.1 49 Secondary -12.1 -1.8 13.8 65,66,68 Secondary -12.2 -2.3 67 Secondary -11.7 -1.6 15.2 70 Secondary -11.9 -1.8 12.3 74,75,76 Secondary -12.2 -2.5 83 Primary -9.9 -1.2 13.6 88 Secondary -11.9 -1.8 89 Secondary -11.5 -2.3 90 Secondary -12.5 -2.0 91 Secondary -10.6 -1.6 14.2 92 Secondary -10.5 0.1 93 Primary -10.4 -1.2 12.8 95 Primary -11.7 -0.8 13.0 102,103 Secondary -12.8 -1.2 15.3 105,106,107 Secondary -10.2 -1.0 12.0 108 Primary -10.6 -0.7 111 Secondary -12.5 -1.4 12.3 113 Secondary -10.1 -1.6 13.7 114,117a Secondary -11.5 -1.4 114,117b Secondary -12.4 -2.8 115 Secondary -11.5 -1.7 14.3 123 Primary -9.6 -0.5 132 Secondary -11.8 -1.9 13.8 134 Secondary -11.6 -1.5 13.9 135 Secondary -11.5 -0.8 137 Secondary -10.5 -0.9 139,140b Secondary -10.7 -1.5 141 Secondary -10.7 -0.1 143 Secondary -9.6 -1.5 147,148 Secondary -8.7 -0.1 154,155,156 Primary -11.9 -1.4 158 Secondary -11.7 -2.5 10.8 159 Secondary -10.2 -1.5 14.4 165,166,167 Secondary -11.5 -1.9 14.7 170a Secondary -12.2 -1.4 13.3 170b Secondary -10.1 -0.1 -11.1 -1.4 13.4 Mean Std. Deviation 1.0 0.7 1.1
57 Table 2-5. Harris Creek local human burials and dental isotopic ( 13CPDB, 18OPDB, 15NAIR) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW). Burial Modes: F/V= Flexed Vertical, ?/S = recorded as Secondary burial. Burial No. Tooth Burial Layer Burial Mode 13CPDB 18OPDB 18OSMOW 18OIW 15NAIR 25 RM3 5 F/V -12.4 -2.2 28.6 -3.4 35 RM3 4 F/V -11.0 -1.7 29.2 -2.6 46 RM3 4 F/V -11.8 -2.1 28.8 -3.2 49 LM3 ? ?/S -12.1 -1.8 29.1 -2.8 13.8 65,66,68 LM3 3/4 F/V -12.2 -2.3 28.6 -3.4 67 RM3 4? F/V -11.7 -1.6 29.3 -2.5 15.2 70 RM3 4 F/V? -11.9 -1.8 29.1 -2.8 12.3 74,75,76 LM3 4 F/V -12.2 -2.5 28.3 -3.8 88 LM3 3 F/V? -11.9 -1.8 29.1 -2.7 89 LM3 3 F/V? -11.5 -2.3 28.6 -3.4 90 LM3 4 F/V? -12.5 -2.0 28.8 -3.1 91 RM3 4 F/V -10.6 -1.6 29.2 -2.6 14.2 113 LM3 4? F/V? -10.1 -1.6 29.3 -2.5 13.7 114,117b LM3 4? F/V -12.4 -2.8 28.0 -4.1 115 LM3 4? F/V -11.5 -1.7 29.2 -2.7 14.3 132 RM3 3 F/V -11.8 -1.9 29.0 -2.9 13.8 158 LM3 4 F/V -11.7 -2.5 28.3 -3.7 10.8 165,166,167 LM2 3/4? F/V -11.5 -1.9 28.9 -3.0 14.7 -11.7 -2.0 28.9 -3.1 13.6 Mean Standard deviation 0.6 0.4 0.4 0.5 1.3
58 Table 2-6. Harris Creek non-local (southern ) human burials and dental isotopic ( 13CPDB, 18OPDB, 15NAIR) values (per mil, ) and converted 18O-Standard Mean Ocean Water (SMOW) and 18O-Ingested Water (IW). Burial Modes: F/V= Flexed Vertical, F/R or F/L = Flexed on Right or Le ft side, E/B = Extended on Back. Burial No. Tooth Burial Layer Bu rial Mode 13CPDB 18OPDB 18OSMOW 18OIW 15NAIR 4 LM3 3 F/V -10.5 -1.2 29.7 -2.0 12.8 6 LM3 3 F/V -10.1 -1.5 29.4 -2.4 14.9 8 LM2 3 F/R -10.2 -0.9 30.0 -1.7 13.1 11, 4 LM3 3 F/R? -11.3 -0.7 30.3 -1.3 13.0 16, 23 LM3 4 F/R -11.7 -1.5 29.4 -2.4 12.6 19 RM3 3 E/B -10.5 -1.6 29.3 -2.5 12.8 29 LM3 5 F/V -10.9 -1.2 29.7 -2.0 36 LM3 4? F/R -10.4 -0.5 30.4 -1.1 38 RM3 4 F/R -8.9 -0.4 30.5 -0.9 83 RM3 3 F/R -9.9 -1.2 29.7 -2.0 13.6 92 RM2 4 F/V -10.5 0.1 31.0 -0.3 93 LM3 5 F/L -10.4 -1.2 29.7 -2.0 12.8 95 LM2 3 F/R -11.7 -0.8 30.1 -1.5 13.0 108 LM3 3 F/L -10.6 -0.7 30.2 -1.4 111 LM3 ? F/V -12.5 -1.4 29.5 -2.3 12.3 123 RM3 3 F/R -9.6 -0.5 30.5 -1.0 134 RM3 ? F/V -11.6 -1.5 29.4 -2.4 13.9 135 LM1 3? F/V -11.5 -0.8 30.1 -1.4 137 LM3 ? F/V -10.5 -1.0 29.9 -1.7 141 RM2 4 F/V -10.7 -0.1 30.8 -0.6 143 RM3 4 F/V -9.6 -1.5 29.4 -2.4 159 RM3 4 F/V -10.2 -1.5 29.4 -2.4 14.4 102,103 LM3 3 F/V? -12.8 -1.2 29.7 -2.0 15.3 105,106,107 RM3 3 F/V -10.2 -1.0 29.9 -1.8 12.0 114,117a RM3 4? F/V -11.5 -1.4 29.5 -2.3 139,140b LM3 3 F/V -10.7 -1.5 29.4 -2.4 147,148 LM3 4/5? F/V -8.7 -0 .1 30.9 -0.5 154,155,156 LM3 3 F/RorL -11.9 -1.4 29.5 -2.4 170a RM3 3? F/V -12.2 -1.4 29.5 -2.2 13.3 170b RM3 3? F/V -10.1 -0.1 30.9 -0.5 Mean -10.7 -1.0 29.9 -1.7 13.3 Standard deviation 1.0 0.5 0.5 0.7 0.9
59 Figure 2-1. Location of the Harris Creek site.
60 Figure 2-2. A plot of 18O and 13C ratios of humans and fauna from Harris Creek. The shaded area is the range of local 18O ratios.
61 Figure 2-3. Estimated 18O isopatches of river water for the state of Florida. 0 per mil -4 per mil
62 Figure 2-4. Distribution of local and non-local burials sampled in this research and the location of AMS dated individuals. Image courtesy of Asa Randall.
63 CHAPTER 3 DIETARY CONTINUITY AND CHANGE DURING THE FLORIDA ARCHAIC Introduction Traditional cultural evol utionary models for the Archaic Pe riod in the St Johns region of Florida assert the start of the Middle Archaic Period is marked by a new and intense use of riverine resources (Milanich, 1994; Milanich & Fairbanks, 1980) Large mounds composed of freshwater shell along the St Johns River are often cited as evidence of this shift in subsistence (Milanich, 1994; Milanich & Fairbanks, 1980; Ru sso et al., 1992; Wheeler & McGee, 1994). Stable isotope ratios recovere d from human bone from Windover, an Early Archaic site, and Harris Creek, a Middle Archaic site are used to evaluate the dietar y shift. These data are also used to assess models of human mobility for the St Johns region of Fl orida during the Archaic Period. Evidence from Archaic sites in and around th e St Johns River basin suggests the inhabitants of riverine sites in teracted with the coast through seasonal occupation or trade. Sharks teeth and marginella beads have been recovered from Windover (Doran, 2002) and marine shells ( Busycon carica, Busycon contrarium, Strombus gigas ) and shark's teeth have been recovered from most, if not all, Mt. Taylor sites incl uding Harris Creek (Aten, 1999; Wheeler et al., 2000). Middle Arch aic Period sites have been doc umented in the coastal zone and Florida Archaic stemmed point s and baked clay objects, both associated with the Mt. Taylor culture, have been found on many of these sites (Piatek, 1994; Russo & Ste. Claire, 1992; Ste. Claire, 1990). However, the nature of the cont act between riverine Mt Taylor populations and the coast remains unclear. Tw o settlement models have been suggested to explain the relationship of Mt Taylor sites on the St. Johns River to Middle Archaic sites on coast. Milanich and Fairbanks (1980) suggest people occupied the St. Johns River drai nage during the warm
64 months, but split into smaller gr oups during the winter and disper sed to the coast or to the interior highlands. Russo and Ste. Claire (1992 ) offer an alternative model. Based on faunal data from Late Archaic sites they assert that se parate cultures occupied the river basin and coast throughout the year (Russo & Ste. Claire, 1992). Groups may have been mobile, but movement was along the coast or river rather than between them (Russo & Ste. Claire, 1992). By the beginning of the Mt. Taylor peri od, populations inhabited the St. Johns River drainage for at least a portion of each year and ap pear to be heavily reliant on aquatic resources. (McGee, 1995; Milanich, 1994; Milanich & Fa irbanks, 1980; Russo et al., 1992; Wheeler & McGee, 1994). Based on food remains rec overed from Groves Orange Midden (8VO2601), freshwater snails contributed be tween 33 and 87% of the dietary m eat weight at Mt. Taylor sites during the early Mt Taylor Peri od (Wheeler & McGee, 1994). The snails were supplemented by large contributions from freshwat er fish and other aquatic vertebrates (Wheeler & McGee, 1994). During the late Mt Taylor and early Orange Period a shift may have occurred to a reliance on freshwater snails which may have composed as much as 98% of the meat weight in the diet (Russo et al., 1992). In this study, 13C and 15N ratios are used to test the assertion that a significant difference exists between the diets of Middl e Archaic populations a nd the preceding Early Archaic populations. An intense use of riverine resources should be ap parent in the stable isotope ratios of individuals inte rred in the region. If significa nt dietary change did not occur and the inhabitants of the region were not eating a high percentage of freshwater shellfish, then the formation of shell mounds in the region is not explained by a change in subsistence which allows for evaluation of alternative explanations.
65 Background Stable Isotopes and Dietary Reconstruction The use of stable isotopes to reconstruct diet and movement in bioarchaeology is now (Ambrose et al., 2003; Buikstra & Milner, 1991; Dupras & Schwarcz, 2001; Montgomery et al., 2005; Price et al., 2000; Richar ds et al., 2001; Sealy, 2006; White et al., 1998). Many researchers have used stable isotopes to reconstr uct the diet of ancient Floridians (DeLeon, 1998; Hutchinson, 2004; Hutchinson et al ., 2000; Quinn, 1999; Turner et al ., 2005; Tuross et al., 1994). Of the light stable isotopes, carbon and nitrogen are most comm only used to study paleodiet and are the isotopes employed in this study. Carbon isotopes in terrestrial systems vary by photosynthetic pathwa y. There are three photosynthetic pathways C3, C4, and CAM. C3 plants have 13C ratios from to with an average of .5 (Smith, 1972; Smith & Epstein, 1971). C4 plants have 13C values that range from to with a mean of .5 (S mith, 1972; Smith & Epstein, 1971). Edible CAM plants, largely limited to succulents in arid areas with low summer rainfall, have 13C values that overlap with C3 and C4 plants. CAM plants are not a common foodstuff in the Eastern Woodlands, though prickly pear does grow along the eastern coast and has been found at Archaic sites including Wi ndover (Doran, 2002; Newsom, 1994; Tuross et al., 1994). Carbon isotopes also differ between terrestrial and marine systems. Marine vertebrates typically have 13C values of to compared to the to range of C3 plants (Schoeninger & DeNiro, 1984; Smith, 1972). Therefore, in systems where C4 or CAM input is negligible, carbon values differentiate between terrestrial and marine diets. While carbon isotopes clearly distinguish terrestrial C4 from terrestrial C3 resources, the addition of nitrogen isotopes greatly expands th e explanatory power of isotopic reconstruction. Though 13C values can be used to discriminate between trophic levels, 13C are only slightly
66 enriched per level, around 1 (S choeninger & Moore, 1992). This small amount of enrichment limits the use of carbon values fo r the study of trophic level in all but the most controlled systems (Wada, 1980). Nitrogen isotopes exhibit a mu ch larger enrichment factor allows for an improved understanding of terrestrial food chains and distinguishes between terrestrial and marine diets (Norr, 1995). The 15N ratios of marine foodwebs are enriched compared to terrestrial systems because plankton and marine plants at the base of the foodweb are enriched in 15N compared to terrestrial plants (Schoeninger et al., 1983; Schwarcz & Schoeninge r, 1991). In terrestrial and marine systems each successive step in the f ood chain is marked by a 3 to 4 increase in 15N values. Therefore, primary consumers are 3 to 4 enriched relative to primary producers, secondary consumers are 3 to 4 enriched rela tive to primary consumers and so on. This increase is referred to as the trophic level e ffect, and has been well documented (Ambrose & DeNiro, 1986; DeNiro, 1985; De Niro & Epstein, 1981; Katzenbe rg, 1989; Minagawa & Wada, 1984; Schoeninger, 1985; 1989; Sc hoeninger & DeNiro, 1984). Apatite to Collagen Spacing and Diet Bone is divided into two components in prep aration for isotopic sampling: the organic collagen fraction and the inorgani c apatite fraction. The organi c portion comprises 35% of the dry weight of bone and the inorganic component comprises 65% of the dry weight of bone (Fawcett, 1986). Differences in the way carbon isotopes are routed to the two fractions of bone make the offset between the 13C values of apatite and collagen informative. Results of controlled feeding expe riments indicate the 13C value of bone collagen is greatly influenced by the isotopic composition of dietary protein rather than that of th e whole diet (Ambrose & Norr, 1993; Tieszen & Fagre, 1993). The bone apatite valu es differed from the collagen values in that they more closely reflected the composition of the entire diet.
67 Ambrose and Norr (1993) developed a model to distinguish the isotopic composition of each component of the diet based on the difference in 13C between bone collagen and apatite. Since collagen values disproportionately reflect the isotopic composition of dietary protein and apatite values more closely mirror the composition of the whole diet, the difference in the 13C values between the fractions can evaluate the isotopic value of carbohydrates in the diet. When the dietary protein and energy sources have simila r isotopic values, the spacing between them is intermediate, around 5.7% 0.4%. When the 13C value of the dietary energy source is more negative than that of the di etary protein, for example a C3 plant and marine fish diet, the spacing between the collagen and apatite values is quite sm all, from 1.2 % 0.1% to 2.1% 0.2%. If the 13C value of the dietary protein is more negative than the 13C value dietary energy source, for instance, in a terrestrial diet of C3 protein with maize or other C4 plants for dietary energy, the difference between the apatite and collagen valu es is large, 7.2% 0.3% to 11.3% 0.4%. Ambrose and Norr (1993) demonstrated that even a small amount of a C4 energy source (21%) added to a C3 diet would greatly affect the apatite values and thus the apatite-collagen spacing. Previous Isotopic Studies of Harris Creek and Windover Preliminary isotopic studies of diet have been undertaken at Windover and Harris Creek (Quinn, 1999; Tuross et al., 1994). Tuross et al. (1994) conducted an isotopic analysis of the Windover (8Br246) in 1994 and found an average 13C ratio of human bone collagen ( 13Ccol) of .6 and an average 15N from bone collagen (15Ncol) of 11.8. Tuross et al. (1994) suggest the 13C and 15N ratios reflect a riverine based diet with input from terrestrial fauna that consumed CAM plants and C4 grasses. Overall, Tuross et al. (1994) suggest the people at Windover were seasonally opportunistic and made us e of terrestrial, riveri ne, and perhaps small amounts of marine resources.
68 Rhonda Quinn (1999) conducted an isotopic stud y of dentition from the Harris Creek site for her M.A. thesis. Quinn (1999) determined the population at Harris Creek had an average 13C of .4 from human tooth enamel and an average 15N of 13.2 from human dentine. Quinn (1999) interprets these values as evid ence of a reliance on es tuarine and riverine resources, although coastal re sources are not ruled out. Overall, the existing isotopic data from both sites are remark ably similar. The isotopes from both Windover and Harris Creek suggest the popul ations were composed of broad spectrum foragers and hunters who made use of food resour ces from terrestrial, riverine, and estuarine environments. This study builds on the results of the previous studies by sampling additional individuals, sampling both components of bone, and analyzing the resulting data with additional interpretive models to better refine the isotopic life histories of i ndividuals recovered from Windover and Harris Creek. Site Descriptions Though evidence of Archaic Period occupati on is found throughout Florida (Milanich & Fairbanks, 1980), this research is primarily concerned with nort heast Florida and the St. Johns River drainage. Two sites, Windover and Harris Cr eek are sampled in this research. Both sites are considered excellent exampl es of their respective time periods: Windover for the Early Archaic and Harris Creek for the Middle Archaic. A brief description of each site follows. The Windover site (8BR246) The Windover site is an Early Archaic cemeter y located near Titusv ille, Florida (Doran, 2002). The site was discovered in 1982 when a small pond was being filled to make way for a subdivision (Doran, 2002). The site was excavated by Glen Do ran and colleagues during three field seasons from 1984 to 1986 (Doran, 2002). The degree of preservati on at the site was remarkable which resulted in an extensive artifac t assemblage of textiles, floral remains, and
69 tools made of bone, antler and marine shell (D oran, 2002). At least 168 human skeletons were recovered from the shallow but persistent pond. Radiocarbon dates on the human skeletal material indicate the intermen ts occurred between 7,100 and 7,330 rcybp (Doran, 2002). Burials were flexed on their side and he ld down with matting or fabric and wooden stakes in shallow depressions underwater. The Harris Creek/Tick Island site (8VO24) The Harris Creek site is a Mt. Taylor Peri od mortuary mound on Tick Island in the St. Johns River. The mound, constructed in 10 distinct layers, is primarily co mposed of freshwater snail (Viviparus ) shells (Aten, 1999). Bullen recovered four radiocarbon dates from the site ranging from 5450 +/300 to 5030 +\200 rcybp (A ten, 1999). Recently, four new AMS dates from burials at Harris Creek range from (2 ) cal BP 6793-7179 [0.96] (X-9110) to cal BP 64826758 [0.98] (X-9111A), ca. 500 years early on average than the c onventional dates recovered by Bullen. Prior to shell mining the site consisted of a large mound with multiple shell ridges and middens (Aten, 1999). The Harris Creek site has a long history of inves tigation and was first excavated by Clarence B. Moore between 1891 and 1894. After Moores work, the site was mapped and surface collected by Francis Bushnell in 1959. Finally, in 1961, Ripley P. Bullen conducted salvage excavations prior to its dest ruction from shell mining operations. Bullen recovered over 175 human burials which consiste d primarily of two types: loosely flexed primary burials and tightly flexed secondary bur ials (Aten, 1999). Aten (1999) interpreted the different burial modes as evidence of rank or status among the burials. The secondary burials have under gone extensive post-mortem processi ng indicating a higher status than the less processed burials (Aten, 1999). However, recent work by Tucker and Krigbaum (2005) suggests
70 the burials likely represent two or more residential groups and that the differences in burial mode are related to different mortuary custom s rather than differences in status. Materials Additional isotopic data are co llected from Windover and Ha rris Creek. A total of 62 human bone samples were collected in the course of this research. Twenty proximal carpal phalanges were selected from Windover represen ting 10 males and 10 females. Only adults were chosen for analysis and they range in age from ca. 21 to 82. These samples were chosen to complement those sampled by Tuross et al. (1994). Forty-two samples from Harris Creek were chosen to complement those collected by Quinn (1999). Various skeletal elements were sel ected from individuals buried in layers 3 and 4 at Harris Creek, including both tightly flexed and loosely flexed individual s. Due to the highly fragmentary nature of the collection age and sex could not be reliably assessed for the majority of the individuals sampled. Methods Collagen and apatite samples were prepared based on methods detailed elsewhere (LeeThorp, 1989). Bone for collagen samples was clean ed, crushed, and then demineralized in ~50.0 ml of 0.2 m HCL. The HCL was refreshed as needed until demineraliz ation was complete. After decanting the HCL, ~50 ml of 0.125 m NaOH wa s added to remove humic acids. After 12 hours the samples were rinsed to neutral, and ~50 ml of 10-3 m HCL was added. The solution was decanted and condensed to ~5ml or less in a 65C oven. The resulting sample was freeze dried. The apatite samples from both sites were prep ared with similar methods. Bone for the apatite samples was mechanically cleaned and crus hed into a fine powder. Samples were treated with ~12 ml of a solution of 50% Clorox and 50% distilled water, which was replaced as needed
71 until all organics were removed from the sample Next, ~12 ml of 0.2m acetic acid was added and the sample was allowed to sit 24 to 36 hours. Finally, the samples were rinsed and freeze dried. Prior to the Clorox solution, the Windover was treated to remove Rhoplex, a preservative, which was applied to the samples. Previous isotopic research on remains from Windover demonstrated that the preservative applied to the remains, Rhoplex, is difficult to remove and would be expected to skew 13C values derived from bone collagen (Tuross et al., 1994). Fortunately, Rhoplex is an organic molecule and should not interfere with 13C values derived from th e inorganic bone apatite fraction of bone. Nevertheless, Toluene was used to remove the Rhoplex from each bone sample, in anticipation of analyzing the organic fraction. However, due to the complexities of removing all the Rhoplex from the organic fractio n of bone from Windover, that research is not being pursued at this time. The Rhoplex and Toluene treatment should not complicate the collection of data from bone apatite. Results Harris Creek: The average 13C of human bone apatite ( 13Cap) from Harris Creek (n=42) is .9. The average 13Ccol is -16.5 and the average 15Ncol is 12.5 (see Table 3-1). The average collagen to apatite spacing for Harri s Creek is 6.3. This value was calculated by subtracting the 13Ccol from the 13Cap for each individual and averaging the results. Windover: The average 13Cap (n=20) for Windover is .2. The average 13Ccol value from Tuross et al. (1994), -15.6 is paired with the apatite value in Table 3-1. The average collagen to apatite spacing for Windover was 5.4. This spacing was calculated differently from the average for Harris Creek. Unfortunately, it was not possible to sample the same individuals as Tuross et al. (1994). Theref ore, collagen to apatite spacings could not be calculated for individuals. Instead the average 13Cap for the site was subtracted from the average
72 13Ccol for the site and the average taken. As a c ontrol, the average spacing for Harris Creek was recalculated in this fashion and re sulted in a difference of only 0.3. Discussion Overall, the results of the is otopic assays for Windover and Harris Creek are similar, the 13C and 15N ratios of collagen and apatite, as well as the collagen to ap atite spacing, from the two sites are all within 1.0 (Table 3-1). The average 13Cap is -10.2 from Windover and 9.9 from Harris Creek, a difference of only 0.3. The apatite ratios from both sites reflect a bulk diet that borders between a marine and a terrestrial C3 diet. Marine vertebrates typically have 13C values of -19 to -9 compared to the -35 to -20 range of C3 plants and the animals that feed upon them (Schoeninger & DeNiro, 1984; Smith, 1972). However, due to isotopic fractionation 13Cap, 13Ccol and 15Ncol do not directly represent th e average values of the food consumed. To calculate the 13C and 15N values of the diet, the 13Cap must be adjusted by 9.5, 13Ccol by between -0.8 to -1.0 (1.0 is used here), and 15Ncol by between -2.5 and 3.4 (-3.0 is used here) (Bocherens & Druc ker, 2003; DeNiro & Epstein, 1981; Kelly, 2000; Minagawa & Wada, 1984; Vanderklift & Ponsard, 2003). The adjusted 13Cap values are -19.7 for Windover and -19.4 for Harris Creek. These values cluster at near the inte rsection of marine, fres hwater, and terrestrial resources (Figure 31). Therefore, based on an analys is of bulk diet, these ratios sugge st all three types of foods were consumed in moderate amounts. The 15N ratios from Windover and Harris Creek differ by only 0.7. Like the 13Cap, the 15Ncol values of bone collagen fall be tween a marine and terrestrial C3 diet. The 15Ncol ratios from both sites are higher than would be expected from a diet consisting of 100% terrestrial C3 resources and fall into the low range of what could be expected fr om a diet consisting of marine and estuarine foods. Tuross et al. (1994) considered the possibility that the 15Ncol ratios could
73 be explained by the use of marine shellfish in the diet, but this inte rpretation was dismissed because no accumulations of marine shells were found during excavations. Windover is an inland mortuary site without associated habitation and would not necessarily show the consumption of marine foods even if they were present in the diet. As such the 15Ncol ratios of both populations support the assertion that marine, freshwater, and terrestrial C3 foods were consumed. The average 13Ccol values from Windover and Harri s Creek differ by only 0.9. The average 13Ccol from both sites is lower than ex pected for a diet composed of C3 resources. The 13Ccol values at Windover and Harris Creek, like the 13Cap values, suggest a mixed diet of marine, freshwater, and terrestrial food resources. The influence of marine protein is apparent in both sites. Finally, the average collagen to apatite sp acings at Windover and Harris Creek were 5.4 and 6.3 respectively, a difference of 0.9. Th e ap-col spacings indicate that the 13C ratios of the dietary protein and car bohydrates consumed by the peopl e buried at Windover and Harris Creek were similar (Figure 3-2). This monoiso topic diet could result from a diet of all C3 resources, all C4 resources, or a diet that included a mix of C3 and C4 carbohydrates and proteins (Norr, 1995). The small differences in isotopic ratios between the two sites are not what would be expected if one population was consuming 30 to 80% of their dietary meat weight from freshwater fish or shellfish and the other was not. Kellner and Schoeninger (2007) recently develo ped a protein regression model to interpret 13C values recovered from hu man bone. This model controls for protein in the diet with three regression lines: one for C3 protein, one for marine protein, and one for C4 protein. The model also provides information about carbohydrates in the di et. The upper ends of the regression lines
74 indicate more C4 carbohydrates in the diet while the lowe r ends of the regr ession line indicate more carbohydrates from C3 sources. When the average (2 std. devs.) 13Ccol and 13Cap values from Windover and Harris Creek are plotted on the regressions, they are intermediate to the C3 and marine protein lines, sugges ting these populations consumed bot h protein types (Figure 3-3). These results support the interp retation of a substantial mari ne component in the diet. Surprisingly, the earlier population from Windover plots closer to the marine line and may have consumed more marine protein th an the population from Harris Creek. To refine the diets of the people at Windover and Harris Creek, IsoSource is used to model the dietary inputs at Windover and Harris Creek. IsoSource is a visual Basic computer program designed to incorporate multiple isotopic sources into a single mixing model (Phillips & Gregg, 2003). The user inputs the isotopic signatures of the sources (foods), the mixtures (population averages), the source increment and mass balan ce tolerances. Five broad categories of food resources with different isotopic ra nges were defined for the region: freshwater fish, marine fish, terrestrial game, marine mollusks, and freshwater snails/mollusks. The average isotopic values for each of these categories were determined from published samples collected from Florida (Table 3-2) (Hutchinson, 2004; Tuross et al., 1994). The average isotopic values of the populations at Windover and Harri s Creek were entered into the program and different combinations of the five source categories were calculated in an atte mpt to match the human values. To produce robust estimates, a source increment of 1% and a tolerance of .5 were used. In the absence of a single solution for th e mixtures and resources the program returns a range of feasible combinations, which are show n in Table 3-3. Based on the stable isotope values the people at Windover a nd Harris Creek were consuming the majority of their diet as marine fish with smaller contribution of freshwat er fish, terrestrial game and fresh and saltwater
75 shellfish. There is a slight increase in the amount of freshwater fish and terrestrial game in the diets of the population at Harris Creek. A significant marine component is suggested by this model, however, caution must be used when interpreting these values. Linear models, such as IsoSource, make several assumptions that may not be accurate. These models assume that an equal proportion of carbon and nitrogen are incorporated into the consumers tissue from a given food resource. This may not always be true if the food resources in the model have strongly divergent C:N ratios or different digestibility factors. Though models, Isoerror and Isoconc, have been developed to help deal with these problems, they are unable to handle mo re than 3 potential sources. Fortunately these factors should not greatly impact these models because they are based on the 13C and 15N of bone collagen of consumers and potential food s ources. As such, the contribution of plant resources is excluded from these models. This exclusion does not impl y that plant resources were not an important dietary re source during the Florida Archaic period, rather in diets with adequate protein intake, the 13C of collagen has been shown to primarily reflect the protein portion of the diet (Ambrose & No rr, 1993; Tieszen & Fagre, 1993). In terms of N, plants have significantly lower amounts of digestible N and only contribute signifi cant amounts of N to the collagen of the consumer when very little anim al protein is available (Hedges, 2004; Newsome et al., 2004; Richards & Hedges, 1999) Terrestrial meat, marine fis h, and shellfish have all been shown to have roughly similar C:N ratios and dige stibility factors (Newsome et al. 2004) and the C and N in meat and fish are 100% di gestible (Pritchard & Robbins, 1990). Finally, an additional potential source of error is the sample sizes of the fauna used to calculate the end members for each source category. If the sample sizes are small they may not accurately define the range of isotopic value for e ach category. Therefore, the ranges presented
76 in Table 3-3 should only be used as a measure of relative import ance, not as a percentage of actual diet. Despite the potenti al sources of error, all the data presented here suggest a previously understated marine contribution to human diet at Windover and Harris Creek. Taken as a whole, these results broadly suppor t the assertion that Mt. Taylor groups were adapted to aquatic subsistence systems, but sugg est that these groups reli ed more on marine or estuarine resources than previously noted. In this respect th ese groups may have been similar to the population recovered from Bay West, a Middle Archaic pond burial in Collier County, Florida. Isotopic studies by De Len (1998) suggest the people bur ied at Bay West consumed a diet rich in marine resources. Based on the results from Windover, this adaptation may have had roots in the Early Archaic, rather than being a Middle Archaic innovation. Tools and ornaments made from marine shell or sharks teeth are f ound on most Mt. Taylor sites and marginella beads and sharks teeth tools were recovered from Windover (Doran, 2002; Wheel er et al., 2000). In light of the isotopic data, these marine resources may be better inte rpreted as evidence of coastal occupations, rather than evidence of trade with coastal groups or the result of short term logistical foraging trips to the co ast. Rising sea levels may have destroyed most evidence of the coastal occupations of these groups, however, the Tomoka Mound Complex and other coastal pre-pottery sites may be evid ence of these occupations (Pia tek, 1994; Russo, 1992; Russo, 1996; Russo & Ste. Claire, 1992; Ste. Claire, 1990). Marine resources, including marine shellfish, were definitely exploited by the Late Archaic as evidenced by season of capture studies at Tomoka Stone Park (8VO2571), Useppa Isla nd-Calusa Ridge (8LL51), and Horrs Island (8CR209) (Quitmyer et al., 1997; Russo, 1998 ). The isotopic data collected from Harris Creek indicate the Mt Taylor people building the mound were not consuming large amounts of freshwater snail. This sugge sts layers of mounded
77 snail shell were not foods remains and were intentionally used as a building material rather than merely deposited as midden. The use of freshwat er shellfish as a building material has been argued by Claassen (1991; 1996) fo r Archaic Period mounds in the mid-South. Sassaman (2004) suggests the use of freshwater snail as building material in the St J ohns region and Randall and Sassaman (2005b) cite layers of clean shell and slabs of re deposited concreted midden at Hontoon Island, as evidence of intentional cons truction. The mound at Harris Creek exhibited similar layers alternating layers of midden a nd clean shell, though redeposited midden was not reported (Aten 1999). If the shell was not a major food resource, the use of shell as a building material indicates that Mt Tayl or Period shell mounds were more than trash dumps that gradually grew in size and were made m eaningful as part of the landscape. These mounds were likely intentional monumental structur es created by communal efforts of seasonally mobile groups. Conclusion Analyses of 13C and 15N ratios in bone collagen and 13C ratios in bone apatite from Windover and Harris Creek indicate a minor shift in diet between the Early Archaic and Middle Archaic Periods. Contrary to widely held beliefs (Wheeler and McGee, 1994; Wheeler et al., 2000; Russo, 1992) if Harris Creek is representati ve, Mount Taylor populations in the St. Johns region were not consuming large numbers of freshwater snails. Surprisingly, based on their 13C and 15N ratios and the estimated isotopic end member values of potential food resource from the region, people at Harris Creek and Windover consumed large amounts of marine/estuarine fish. Based on the isotopic data gath ered from these two sites, no major shift in resource use occurred between the Early and Middle Archaic in northeastern Florida. The data suggest that during the Mt Taylor Period, an existing coastal aquatic adaptati on may have been extended into the river basin. As such the impetus for the cr eation of shell mounds along the river valley must be attributed to something other than a heavy reliance on riverine resources. Obviously
78 something changed around 7500 BP which resulted in the creation of fres hwater shell mounds. As we move away from explanatory schemes base d on evolutionary thought, we are also able to move away from explanatory schemes driven by environmental change. New avenues of research based on archaeologies of landscape an d practice may prove more fruitful (Randall & Sassaman, 2005a). In the end, the beginning of the Shell Mound Archaic may have had more to do with what was in peoples heads than in their bellies.
79 Table 3-1. Windover and Harris Creek/Tick Island 13C and 15N ratios Site 13Ccol 13Ccol adj. 13Cap 13Cap adj. Col -Ap 15Ncol 15Ncol adj. References Windover .6 .6 .2 .7 5.4 11.8 8.8 Tuross et al. 1994; This paper Harris Creek/Tick Island .5 .5 .9 .4 6.3 12.5 9.5 This paper Difference 0.9 0.9 0.3 0.3 0.9 0.7 0.7 Table 3-2. Isotopic endpoints for th e Isosource linear mixing model. Food type 13C 15N freshwater fish -23.5 9.6 marine fish -15.2 9.5 terrestrial game -21.3 7.6 marine shellfish -15.7 7.8 freshwater snail -25.7 4 Table 3-3. Results of Isosource remixing model for Harris Creek and Tick Island. IsoSource Freshwater Fish Marine Fish Terrestrial Marine Shellfish Freshwater Snail Windover 0-22% 12-90% 0-30% 0-71% 0-17% Harris Creek/Tick Island 2-33% 38-80% 0-27% 0-31% 0-6%
80 13C and 15N ratios of bone collagen and apatite from Windover and Harris Creek 0 2 4 6 8 10 12 14 -30 -25 -20 -15 -10 -5 13C15N Local Terrestrial Mammals Local Marine/Estuarine Local Freshwater Winover Collagen Harris Creek Collagen Windover Apatite Harris Creek/Tick Island Apatite 13C and 15N ratios of bone collagen and apatite from Windover and Harris Creek 0 2 4 6 8 10 12 14 -30 -25 -20 -15 -10 -5 13C15N Local Terrestrial Mammals Local Marine/Estuarine Local Freshwater Winover Collagen Harris Creek Collagen Windover Apatite Harris Creek/Tick Island Apatite Figure 3-1. Adjusted 13C and 15N ratios of bone collagen a nd apatite from Windover and Harris Creek
81 Figure 3-2. Difference in mean 13C ratios between apatite and collagen from Windover and Harris Creek. 0123456789101112 Harris Creek Windover Monoisotopic Marine Protein & C3 Carbs Terrestrial Protein & C4 Carbs
82 Figure 3-3. Regre ssions of protein 13C ratios. Regressions from Kellner and Schoeninger 2007. Sites averages are shown with 2 standard deviations. Diet by Protein Source -18 -16 -14 -12 -10 -8 -6 -4 -2 0 -25-20-15-10 -5 0 13C col C3 Protein Marine Protein C4 Protein Harris Creek Windove r 13 C ap
83 CHAPTER 4 EVALUATING A SERIAL SAMPLING METHODOLOGY Introduction Serial sampling or isotopic zo ning studies (Kohn & Cerling, 2002) assess changes in diet or environment recorded as a series of stable is otopic ratios in biological tissues (Balasse et al., 2002; Balasse et al., 2005; Fricke & O'Neil, 1996 ; Gadbury et al., 2000; Kohn et al., 1998; MacFadden et al., 2004; Passey et al., 2002; Straight et al., 20 04; Wiedemann et al., 1999). Tooth enamel is the tissue of choice because the appositional nature of enamel development produces growth layers encapsulating isotope ratios from different periods of enamel formation. Sampling discrete increments of enamel produces isotopic data which allow changes in ecology and diet that occurred during en amel formation to be observed. Several factors complicate serial samp ling studies including environmental and physiological processes which buffer or attenuate the isotopic signal recorded in tooth enamel (Balasse, 2003). One of greatest concerns in serial sampling huma n and faunal teeth is secondary mineralization. During amelogenesis, enamel is only partially mineralized as it is deposited as along appositional growth fronts. The remaining mineral content of the enamel is introduced during the maturation phase by a proc ess known as secondary mineralization (Avery, 2000; Schroeder, 1991). This process significantly prolongs the pe riod of enamel maturation and introduces isotopes into the enam el from food and water ingested after the formation of the appositional growth structur e (Balasse, 2003; Kohn & Cerlin g, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). Serial sampling studies are infrequently applied to humans, in part because human teeth are small and bunodont compared to the larger, more hypsodont teeth of fauna. However, serial sampling studies of human and hominin remains have been successful. For example, human
84 molars from Wharram Percy were sectioned (as opposed to sampled) to infer the timing of weaning in a Medieval British population (Fuller et al. 2003). The dent al remains of a ~2mya old robust australopithecine, Parathropus robustus, were effectively serially sampled by laser ablation to provide dietary and enviro nmental data (Sponheimer et al. 2006). A recent study of mammalian enamel demonstrates that isotopic sampling the growth layers between the inner and outer enamel recovers isotopic varia tion which remains intact after the maturation phase (Humphrey et al., 2007). Hu mphrey et al. (2007) identified changes in strontium/calcium ratios in infant enamel before and after birth and these changes were retained after mineralization was complete. Richards et al (2008) also used la ser ablation to recover strontium isotopes from Neanderthal enamel wh ich showed the Neanderthal moved over 20km during its lifetime. The success of these studie s, coupled with the widespread success using faunal teeth, suggests the effects of secondary mi neralization do not signif icantly impact isotopic studies. This research evaluates the effects of sec ondary mineralization in human tooth enamel by sampling teeth along two different axes: (1) cross-cutti ng the appositional front s, roughly parallel to the enamel prisms and (2) parallel to the pe rikymata on the tooth surface (Figure 4-1). We also compare the amount of variation recovere d from modern and prehistoric samples to determine if the variation recovered results from environmental signals or an internal biological rhythm. Environmental Buffering and Isotopic Attenuation Initially the layers of enamel were thought to record the 18O values of ingested water in a relatively direct fashion (Frick e & O'Neil, 1996 ; Wiedemann et al., 1999). Despite the early successes, concerns over the le ngth of time enamel requires to fully mineralize emerged (Balasse, 2003; Balasse et al., 2002; Passey & Cerli ng, 2002). Enamel is essentially a
85 hydroxyapatite mineral which is precipitated disc ontinuously at different times and rates over a period of weeks or months (Sakae & Hirai, 1982; Suga, 1982; 1989) which may result in an attenuated (time averaged) isot opic signal (Balasse, 2003; Bala sse et al., 2002; Hoppe et al., 2004; Zazzo et al., 2005). Waves of mineralizati on do not follow the growth structures and as such the isotopic signature associated with the form ation of the structures may be lost or greatly attenuated. The influence of later minera lization phases may prev ent the recovery of chronological signals from the gr owth structures (Balasse, 2 003; Kohn & Cerling, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). Factors unrelated to mineralization can also complicate the interpretation of serially sampled data. Surface water produces a reservoir e ffect as precipitation is mixed and stored in lakes and rivers. Such water sources ultimately reflect seasonal variat ion isotopically, though the isotopic values are muted (Balasse, 2003; K ohn & Cerling, 2002). Body water constitutes another reservoir in which oxygen is otopes are stored and mixed in the organism before they are locked into tooth enamel (Bala sse 2003; Kohn and Cerling 2002). Despite these sources of poten tial buffering, serial sampling is a proven means to recover isotopic time series data from teeth (Balasse et al., 2002; Balasse et al ., 2006; Balasse et al., 2005; Fricke & O'Neil, 1996 ; Gadbury et al., 2000; Kohn et al., 1998; MacFadden et al., 2004; Nelson, 2005; Sharp & Cerling, 1998; Sponheimer et al., 2006; Straight et al., 2004; Zazzo et al., 2006). For example, Gadbury et al. (2000) interpreted serially sampled 13C and 18O ratios as evidence of a drought which affected herds of bis on prior to the catastroph ic event that formed the Hudson-Meng bone bed in Nebraska. Addi tionally, Balasse and colleagues (2005; 2006) have investigated the seasonal cons umption of seaweed by sheep with 18O and 13C values recovered by serial sampling.
86 Variation in 18O is interpreted as a proxy to seasonal 18O variation in precipitation and surface water, an assertion suppor ted by studies of African gazelle and other fauna in which the primary source of 18O variation was seasonal fluctuation of 18O values in precipitation (Balasse, 2003; Fricke & O'Neil, 1996 ; Gadbury et al., 2000; Kohn et al., 1998; MacFadden et al., 2004; Wiedemann et al., 1999). However, variation in 18O values may also result from changes in developmental physiology including ch anges in diet, metabolism, or water turnover rates (Kohn et al., 1998). Tooth and Enamel Development Teeth begin to form as enamel is secreted in layers which stack on the previous layer until the crown is fully formed (Figure 4-2.). Th ere are three phases of enamel development; differentiation, secretion, and maturation. (Ave ry, 2000; Moss-Salentijn et al., 1997). The maturation stage is of particular interest to researchers involved in serial isotopic sampling of mammalian tooth enamel. Maturation and Mineralization Waves New enamel matrix is transformed into matu re crystalline enamel during the process of enamel maturation (Schroeder, 1991). These ch anges include growth of enamel crystals, concentration and hardening of the mineralizing structure, selective changes in the composition of the enamel matrix, a decrease in the volume of organic matri x, and a loss of water (Schroeder, 1991). In general, these changes occu r through the process of mineralization. As soon as amelogenesis is complete the matr ix begins a progressive and discontinuous process of mineralization (Weinmann et al 1942; Suga et al. 1970, 1987; Suga 1979, 1982; Sakae and Hirai 1982; Engel and Hilding 1984; Mo ss-Salentijn et al. 1997). Mineralization proceeds in a series of four phases or waves (Suga et al., 1987). These successive waves of mineralization proceed as successive fronts in multiple directions across the enamel layer
87 between the enamel surface and the dentine-enam el junction (DEJ) (Balasse, 2003; Suga et al., 1987). During mineralization, small crystal seeds of ap atite are deposited in the newly secreted enamel (Avery, 2000; Schroeder, 1991). The ini tial phase of enamel mi neralization occurs as weakly (ca. 20%) mineralized protein matrix is secreted. An exception occurs in the first layer of enamel secreted adjacent to the DEJ, this la yer (ca. 8um thick) is more heavily mineralized (ca. 60%) than later enamel and is less affected by the successive waves of mineralization (Sakae & Hirai, 1982; Suga, 1982). The second phase of mineralization increases mineral content as the wave progresses from the surface toward the DEJ (S uga et al., 1987). A tertiary phase occurs as a wave of increased mineralization moves from th e DEJ to the surface of the tooth (Suga et al., 1987). Lastly, during the quatern ary phase, the narrow outer laye r of enamel (ca. 15um) which lags behind in mineralization, undergoes a rapid mi neralization to become the most mineralized tissue in the human body ca. 95% (Suga, 1989; S uga et al., 1987). The high mineral content results in the loss of almost all wate r and organic material during maturation. The increasing mineral content du ring mineralization is a result of the growth of the apatite crystals. The growth of these crystals ha s been documented during the third wave of mineralization (Allan 1967). In itially, the young crystals are 1.5nm thick and 30nm wide. By the end of the third phase th e crystals increase to 20-60 nm thick and 30-90 nm wide. Carbonate Substitutions in Enamel Attenuation caused by mineralization waves is a major concern in serial sampling studies (Balasse, 2003; Balasse et al., 2002; Passey & Ce rling, 2002; Zazzo et al., 2005). However, the manner in which carbonate substitutions occur in enamel suggests these waves do not result in significant buffering when sampling the struct ural carbonate component of tooth enamel.
88 The principal component of enamel is hydroxyapatite Ca10(PO4)6(OH)2 but carbonates and trace minerals are frequently s ubstituted into the hydroxyapatite matrix (Weatherell & Robinson, 1973). Carbonate in enamel mineral is the third most common constituent following calcium and phosphate. The chemical and structural features of enamel carbonate have been studied extensively because carbonate is thought to be related to cariogenesi s (Boyde, 1979). However, Moreno and Aoba (1990) suggest that hydroxya patites containing carbonate are more thermodynamically stable than hydroxyapatites without carbonate, indicati ng carbonate may not be the weak link which precipitates the formation of carious lesions. Sydney-Zax et al. (1991) examined the carbonate content of forming, maturing, and mature enamel in 22 human teeth and 46 bovine teeth. They found that carbona te concentration was highest in newly secreted enamel and began to decrease at the beginning of the maturation stage, finally reaching the lowest concentration in mature enamel. They suggest the decreasing carbonate concentration with the increasing mineralization of en amel results from the dilution of carbonate rich enamel with carbonate poor enamel. Robinson et al. (199 5) support these results demonstrating that as mineralization advances the carbonate concentration, as well as trace minerals such as magnesium and fluoride, are reduced. In addition to humans and bovines, the early secretory enamel of porcine enamel is enriched in carbonate as well (A oba & Moreno, 1990). The formati on of a carbonate rich apatite during early enamel deposition may be a generalize d process in several sp ecies and may reflect a mechanism of mineral precipita tion (Aoba & Moreno, 1990; Hiller et al., 1975; Sydney-Zax et al., 1991). For example, Hiller et al. (1975) reporte d similar enrichment in the development of a rat incisor.
89 Carbonate is incorporated into the hydroxyapat ite molecule at two di fferent sites: (1) in place of the PO4 or phosphate group, a Type B substitution, and (2) in place of the OH or hydroxyl ion, a Type A substituti on (Elliott et al., 1985). Using X-ray diffraction on newly formed secretory enamel, Aoba and Moreno (1990) detected carbonate only in the position normally occupied by PO4 (a Type B substitution). However, incorporation of carbonate into the Type A or hydroxyl position becomes evident as amelogensis proceeds (Aoba & Moreno, 1990). X-ray diffraction showed a significant increase in the parameter between the secretory stage and mature enamel which Aoba and Moreno (19 90) attribute substitution of the hydroxyl ion with carbonate (a Type A substitution). Most carbonate in mature enamel, is at the PO4 position (Type B); only ca. 10-15% of the carbonate ions in human mature enamel are in the OH position (Type A) (Elliott et al., 1985). Aoba (1996) suggests the selective occupation of hydroxyl sites by carbonate in the later stages of enamel formation is due to a competition fo r two sites, one of which (A-type) is favored energetically but limited in number while the B-type site is less favored but present in greater numbers (Aoba, 1996; Sydney-Zax et al., 1991). When there are lower concentrations of carbonate the energetic cons iderations prevail. However, when carbonate is available in greater concentration, the probability of occupation at either site is largely determined by the absolute number of sites. If this model is correct, there may be differences in the mineralizing environment during different mineralizing pha ses (Aoba, 1996). The preferential B-type substitution in early development supports the pr esence of carbonate rich fluid during initial enamel secretion and mineralization. Supporting evidence is found in porcine enamel; the liquid phase surrounding the forming enamel mine ral contains significant amounts of HCO3 ions (Aoba & Moreno, 1987).
90 The difference in the timing of the two substi tutions is important for researchers sampling enamel carbonates. Of the total carbonate in mature enamel 10-15% are in the OH (Type A) position and the remaining 85-90% occupy the PO4 (Type B) position (Elliott et al., 1985). The majority of the carbonate present in newly formed enamel is in the PO4 position, and there is a preference for the OH sites to be filled as mineralization proceeds (Aoba & Moreno, 1987). Therefore, the majority of the car bonate in each sample of mature enamel (ca. 85-90%) is present at the point of enamel secretion and chronologically asso ciated with the visibl e growth structure. Only ca. 10-15% of the total carbonate in a sa mple of mature enamel is introduced during subsequent waves of mineralization. Striae of Retzius and Perikyma ta and Crown Formation Times Serial sampling requires the relative am ount of time encompassed by each sample be known. Striae of Retzius or Retz ius lines are growth lines fre quently discussed in serial sampling literature. More specia lly, such studies often involve th e surface manifestation of these lines, called perikymata (Figure 4-3) As tooth growth proceeds, th e sequential layers of enamel produce appositional lines that can be seen after tooth formation is complete. These features represent the successive positions of the enamel forming front as tooth growth proceeds. The growth rate of teeth is not constant and the spacing of lines of Retzius and perikymata vary throughout the tooth. Shellis (1998) found the rate of enamel apposit ion in the lateral enamel of human permanent dentition increase s from 2-3um/day in the initial stages to 5-6um in the final stages as enamel formation process. Based on a count of cross-striations, lines of Retzius and perikymata have an average periodicity of ca. 9 days in humans and great apes (Benyon et al., 1998; Dean et al., 2001; FitzGera ld, 1998). All estimates for Homo are between 6 and 11 days (summarized in FitzGerald, 1998).
91 Histological examination of periodic growth features in dental enamel has been shown to provide better estimates of crown formation times than more tradition ra diological examinations (Reid & Dean, 2006). Radiologi cal studies often have probl ems detecting the onset of mineralization and do not have the resolution to identify the near microscopic changes which occur during dental development (Benyon et al ., 1998; Reid & Dean, 2006). Reid and Dean (2006) collected data on crown formation time fr om 326 molars from southern Africa, northern Europe and North America. Based on the periodic ity of Striae of Retziu s and cross-striations, the crown formation time for various teeth was calcul ated. Molars exhibited very little variation in crown formation time compared to other t ypes of teeth; all molars completed enamel formation over a period of 3.0 to 3.4 years (R eid & Dean, 2006). Based on Reid and Deans (2006: Figure 4) estimates of ch ronological age for each decile of crown height, roughly 2 years of growth is present along the external tooth su rface, the remaining year of growth is folded inside later growth layers and does not reach the tooth surface. Materials and Methods Sample selection Fifteen human molars were selected to asse ss the effects of secondary mineralization through serially sampling. Modern teeth were obta ined from extant collections in Gainesville, Florida and New Castle-on-Tyne England. Two teeth were selected for sampling from each locality. Unfortunately, the indi vidual life histories of the individuals are not known. The remaining 11 teeth are from 3 diffe rent archaeological contexts: Ni ah Cave West Mouth, Borneo, Niah Cave Lobang Jeragan, Borneo, and Gua Cha, Malaysia. Site Descriptions Individuals from Niah Cave were recovere d from the West Mouth portion of the cave system. Niah Cave is located in Sarawak, Bo rneo and contains one of the most extensive
92 stratified records of human occupa tion in Southeast Asia (Barker, 2002). The site spans the late Pleistocene and Holocene and ove r 170 human burials have been recovered (Krigbaum, 2005). Lobang Jeragan is separate site within the Niah Cave complex. Over 60 Neolthic burials were recovered from the site. Gua Cha is a rock shelte r in northeast Malaysia. The site was occupied during the Mid Holocene by broad spectrum hunter-gatherers and late r used as a burial site by Neolithic populations (Bellwood, 1997). External SamplingDental Drill Tooth surfaces were mechanica lly cleaned with distilled H2O and inspected under magnification. The external surface of each molar was lightly abraded with a Brasseler dental drill and carbide bit (B rasseler #170) to remove remaining de bris. A series of six to eight samples, between .4 and 1.0 mg were removed un der magnification, parallel to the perikymata starting just above the CEJ and proceeding to ward the occlusal su rface (Figure 4-1B). Internal SamplingMicromill A New Wave Research-Merchantek MicroMill sy stem fitted with a carbide bit (Brasseler #170) was used to remove samples from the inte rior of each tooth. The Micromill is a microsampling system specifically designed to collect small samples for isotopic or chemical analysis. The user defines the sample areas with the graphi cal interface and the drilling is fully automated. After the drilling is complete the user collects the sample and prepares it for isotopic assay. Prior to sampling, the molars were mechanically clean ed in distilled water. Once dry, each tooth specimen was sectioned along the mesio-distal mid-section and half of each specimen was mounted on a glass slide. The fi rst sample was collected between the DEJ and the tooth surface under the tip of the cusp. Sampling proceeded toward the CEJ with each sample following the prism orientation and typically includ ed around 6 samples (Figure 4-1A).
93 Pretreatment After both sampling protocols were completed, enamel powder was placed directly into pre-weighed stainless steel capsules and loaded into a multiprep inlet system attached to a VG PRISM II mass spectrometer. The enamel samples were not chemically pretreated. Conventional sampling includes chemical pretreatme nt of bone and enamel with a 2% sodium hypochlorite (NaOHCl) soluti on to remove organics fo llowed by an acetic acid (CH3COOH) solution to remove adsorbed labile or secondary carbonates. However, these treatments result in significant sample loss. A recent trend is to avoid strong chemical treatments (Garvie-Lok 2004; Koch et al. 1997), especially for enamel, which is less likely to be diagen etically altered because it is more thermodynamically and kinetically stable than bone (Nielson-Marsh and Hedges 2000). Passey et al. (2002) found no predictabl e difference between untreated and pretreated samples of modern and fossil fauna from various localities in Nebraska and Texas. Such findings, along with the geologica lly young age of the samples (less likely to have absorbed secondary carbonates) argue agai nst pretreatment. In additio n, not conducting pretreatment allows for very small samples to be analyzed. Larger samples mix more growth bands and result in more time-averaging per sample. Results Fifteen individuals from diverse environmen ts and time periods were sampled in this research. Series of between 6 and 9 samples were removed from the external surface of each tooth for a total of 104 samples, 97 of which re turned acceptable result s. The micromill was used to collected series of between 5 and 10 samples from the interior of the teeth. Of 109 samples collected with the micromill, 103 produced suitable values. Combined, 213 enamel samples were collected and 200 returned satisf actory results. Summar y statistics for each individual are shown in Table 4-1.
94 Discussion Effects of Secondary Mineralization Fifteen human molars were sampled for 18O and 13C (Table 4-1). Combined the 18O and 13C samples provide 30 cases of paired data to assess the variability produced by each sampling method. The possible outcomes can be expressed as a null hypothesis and two alternatives. H0: There will be no significant difference in the amount of variation recovered from the growth structures a nd the mineralization wave. H1: Sampling growth structures w ill recover more isotopic variation. H2: Sampling parallel to the enamel prisms will recover more variation. Of 15 sets of 18O values, sampling the visible growth structure, recovered more isotopic variation in 10 of the 15 cases (Figure 4-4). Sa mpling roughly parallel to the enamel prisms recovered more variation in 4 of the 15 cases. One sample recovered similar variation with both methods. The average difference between the two methods was .69. The 15 sets of 13C values produced results similar to the 18O values. Of the 15 cases, sampling the perikymata recovered more isotopic variation in 13 of the 15 cases (Figure 4-5). Sampling parallel to the enamel prisms only recovered more variation in only 2 of the 15 samples. The average difference in 13C between the two methods is .54. In total, sampling parallel to the perikymata recovered more variation in 23 of 30 cases. The average difference between sampling along the en amel prisms and the growth structures for all 30 samples was .62. Though the difference is relatively small, the results overwhelmingly support alternative hypothesis H1sampling growth structures reco vers more isotopic variation.
95 Source of Isotopic Variation The results of the external sampling of mode rn and prehistoric teeth are compared to determine if the isotopic variation recovered is due to climatic or physiological factors. If the intratooth variation in 18O ratios primarily reflects the consumption of isotopically fluctuating precipitation and surface water, then we expect lit tle to no variation in the modern samples. The pooling and storage of drinking water in reservoi rs, the consumption of bottled and/or canned beverages, and possible buffering from secondary mineralization is expected to erase any seasonal 18O variation in modern samples. If the variation is primarily the result of a physiological process then the variation between the modern and prehistoric teeth should be similar. The modern sample consists of two molars collected from Gainesville, FL and two molars collected from New Castle-on-Tyne, U.K. The results from the modern teeth reflect thei r places of origin (Table 4-6). The mean 18O ratios differ by over 2.0 between the Florida and New Castle specimens as is expected based on the global di stribution of 18O ratios. Interestingly, the 13C ratios also differ between the two locations; both individuals from Gainesville were enriched compared to the two New Castle individuals. The enrichment of the Gainesville specimens relative to the New Castle individuals is probably due to the U.S.s reliance on co rn based products. According to the Food and Agriculture Organization of the United Nations, in 2003 the average kg of corn consumed per captia annually in the U.K. was 3.0kg. Conversely, the average consumption of corn per year per person in the U.S. in 2003 was 13.0kg; this cultural preference in food choice likely accounts for the difference in the 13C ratios. As expected, the modern teeth e xhibit less isotopic variation in 18O than the prehistoric teeth; the average difference in min/max values of 18O for the modern samples is 0.98 compared to 1.4 for prehistoric samples. However, the difference of only .42 between the
96 prehistoric samples and the modern samples is not as large as expected. This difference suggests a considerable amount of buffering is occurring in the prehistoric samples or that modern teeth show more of a seasonal signa l than initially expected. Estimates of modern seasonal variation fr om Gainesville and New Castle-on-Tyne predict that precipitation in New Castle-on-Tyne w ill exhibit more seasonal variation than in the Gainesville area (Bowen, 2005; Bowen & Re venaugh, 2003; Bowen et al., 2005; Bowen & Wilkinson, 2002) but the mean difference in 18O for each site is almost equivalent. The similar mean difference in 18O ratios for each locality suggests that the variation in modern teeth is not accurately reflecting a seasonal signal and likely results from isotopic buffering of the signal prior to ingestion. Since the majority of 18O present in mature enamel is incorporated during enamel deposition, this averaging does not resu lt from secondary mineralization. The lower variation in the modern samples probably result s from the buffering and mixing of the drinking water consumed by modern populations. Buffering occurs as groundwater pools and is mixed in rivers, streams, and lakes, as well as within the reservoir of body water befo re it is incorporated into enamel. Though the difference between the m odern and prehistoric samples is not as large as expected, there is little vari ation in the modern teeth and no pa ttern of variati on that could be attributed to a biological rhythm (Figure 4-7). When the isotopic series are averaged, the isotopic samples from the modern teeth accurately reflect average differences in climate a nd diet, but the individual serial samples do not show a seasonal pattern of precipi tation or evidence a pattern attri buted to an internal biological rhythm. Instead they establish a baseline of variation for two highly buffered populations of around 1.0. The prehistoric samples exceeded this average only by ca. 0.4 which suggests the archaeological populations from Southeast As ia, may have also experienced considerable
97 environmental buffering. Based on the four modern teeth, no pattern of vari ation is apparent that might be caused by an internal biological rhythm, instead the prehistoric teeth reflect an attenuated record of seasonal variation in 18O values. Conclusion Interpreting isotopic time-ser ies data remains challenging. The isotopic signals are subject attenuation from a variet y of source prior to and after i ngestion by an organism. Despite these factors, many serial sampling studies pr oduce results which indicate an environmental signal is recorded in tooth enamel. This study o ffers evidence which indicates why these studies work. Sampling parallel to the visible growth stru ctures recovers more isotopic variation than does sampling parallel to the enamel prisms r oughly along the projected pa th of mineralization. These results do not suggest secondary minera lization is a significant source of isotopic attenuation. Though enamel is only light mine ralized when secreted, ca. 20% (Avery, 2000; Schroeder, 1991), it already c ontains 85-90% of the carbonate present in mature enamel. Additionally, this 85-90% of the total carbonate prefer entially occupies the PO4 position which is assayed during isotopic study. The remaining 10-15% of the total carbonate in mature enamel is introduced during the maturation phase but this enamel preferentially substituted into the OH position which is not assessed durin g isotopic study. Therefore, ne arly 100% of the carbonate sampled in most isotopic studies of tooth enamel is directly associated with the visible growth structure. In these cases secondary mineralizat ion will not result in attenuation of the isotopic signal. However, the relatively small amount of variation found in the pr ehistoric teeth sampled here suggest other forms of isotopic buffering st ill result in significant attenuation. Continued work is needed to model these factors in the absence of attenua tion caused by secondary mineralization.
98 These results indicate secondary mineralization is not a concern for researchers who sample carbonates, but secondary mineralization remains an obstacle when sampling phosphates. Though the majority of the carbonate present in mature enamel is present at deposition, this is not true of phosphate. The amount of phosphate in a sample appears to greatly increase over the maturation phase and as such, concerns over attenu ation still apply. In these cases, models such as those by Passey and Cerling (2002), Passey and colleagues (2005) and Zazzo and colleagues (2005) may help to recover the original amplitude and structure of the isotopic signal. When sampling phosphate the port ion of the tooth sampled is also a concern. Previous studies have suggested the inner most layer of enamel immediately adjacent to the EDJ would produce superior results because it is more heavil y mineralized at deposition and is therefore less affected by the effects of secondary mineralization (Balasse, 2003; Sakae & Hirai, 1982; Suga et al., 1987). However, this inner layer is thin, only around 8-10um thick and is located in the interior of the tooth re quiring the tooth to be sectioned, a potentially expensive and destructive process (Sakae & Hirai, 1982; Suga et al., 1987). Another approach might be to sample the outer layer of enamel; this layer is thicker than the inner layer ca. 15um and more easily accessible. The outer layer remains lightly mineralized until late in develo pment when it undergoes a rapid mineralization to become the most mineralized tissue in the human body (Suga, 1989; Suga et al., 1987). If this mineralization front follows a predictable path, then the outer layer might provide a source of phosphate samples that encaps ulate a relatively small amount of time due to the rapid mineralization of the layer. Further re search may reveal when other isotopes, like lead and strontium, are incorporated in to enamel and if they are subj ect to the effects of secondary mineralization.
99 Table 4-1. Summary statistics for indi viduals sampled in this research. N e w C a s t l e 2 18O External Internal Diff. 13C External Internal Diff. Mean -5.8 -4.3 -1.4 Mean -13.0 -12.5 -0.5 Std. Dev. 0.5 0.2 0.3 Std. Dev. 0.3 0.2 0.1 Median -5.5 -4.3 -1.3 Median -13.0 -12.5 -0.5 Max -5.1 -4.1 -1.1 Max -12.3 -12.2 -0.2 Min -6.7 -4.5 -2.2 Min -13.3 -12.7 -0.7 Diff. 1.6 0.4 1.1 Diff. 1.0 0.5 0.5 New Castle 3 18O External Internal Diff. 13C External Internal Diff. Mean -4.7 -4.9 0.2 Mean -12.3 -12.5 0.2 Std. Dev. 0.1 0.1 0.0 Std. Dev. 0.2 0.1 0.1 Median -4.7 -4.9 0.2 Median -12.3 -12.5 0.3 Max -4.6 -4.8 0.2 Max -12.0 -12.5 0.5 Min -4.9 -5.0 0.1 Min -12.6 -12.7 0.1 Diff. 0.3 0.1 0.1 Diff. 0.6 0.2 0.4 Gainesville G-3 18O External Internal Diff. 13C External Internal Diff. Mean -2.9 -2.6 -0.3 Mean -9.1 -9.2 0.2 Std. Dev. 0.1 0.3 -0.1 Std. Dev. 0.1 0.2 0.0 Median -2.9 -2.6 -0.3 Median -9.1 -9.3 0.3 Max -2.7 -2.2 -0.5 Max -8.8 -9.0 0.2 Min -3.2 -3.0 -0.2 Min -9.3 -9.4 0.1 Diff. 0.5 0.8 -0.3 Diff. 0.4 0.4 0.0 Gainesville G-4 18O External Internal Diff. 13C External Internal Diff. Mean -3.0 -2.9 -0.1 Mean -8.1 -8.8 0.7 Std. Dev. 0.7 0.1 0.6 Std. Dev. 0.5 0.1 0.4 Median -2.7 -2.9 0.2 Median -8.0 -8.8 0.8 Max -2.4 -2.8 0.4 Max -7.3 -8.6 1.3 Min -4.0 -3.0 -1.0 Min -8.5 -8.9 0.4 Diff. 1.6 0.2 1.4 Diff. 1.2 0.3 1.0 N i a h N 3 6 18O External Internal Diff. 13C External Internal Diff. Mean -6.6 -5.5 -1.1 Mean -13.3 -13.5 0.3 Std. Dev. 1.4 0.9 0.5 Std. Dev. 0.5 0.2 0.3 Median -6.2 -5.9 -0.3 Median -13.4 -13.6 0.2 Max -5.5 -3.9 -1.6 Max -12.5 -13.2 0.7 Min -9.8 -6.2 -3.6 Min -13.9 -13.7 -0.1 Diff. 4.3 2.3 2.0 Diff. 1.3 0.5 0.8
100 Table 4-1. Continued N i a h N 8 3 18O External Internal Diff. 13C External Internal Diff. Mean -4.7 -4.9 0.2 Mean -14.7 -15.9 1.2 Std. Dev. 0.5 0.4 0.1 Std. Dev. 0.3 0.1 0.3 Median -4.8 -4.8 0.0 Median -14.8 -15.9 1.1 Max -4.0 -4.4 0.5 Max -14.2 -15.8 1.6 Min -5.5 -5.7 0.2 Min -15.1 -16.0 0.9 Diff. 1.6 1.3 0.3 Diff. 0.8 0.1 0.7 N i a h N 1 6 0 18O External Internal Diff. 13C External Internal Diff. Mean -6.4 -6.6 0.2 Mean -12.0 -12.7 0.7 Std. Dev. 0.6 0.4 0.2 Std. Dev. 0.2 0.4 -0.2 Median -6.2 -6.7 0.5 Median -12.1 -12.5 0.5 Max -5.6 -5.8 0.1 Max -11.6 -12.3 0.7 Min -7.2 -6.9 -0.2 Min -12.4 -13.6 1.3 Diff. 1.5 1.1 0.4 Diff. 0.8 1.3 -0.5 Niah N-173 18O External Internal Diff. 13C External Internal Diff. Mean -7.7 -6.5 -1.2 Mean -13.0 -13.3 0.3 Std. Dev. 0.3 0.3 0.0 Std. Dev. 0.5 0.3 0.2 Median -7.6 -6.6 -1.0 Median -13.0 -13.4 0.4 Max -7.3 -5.8 -1.6 Max -12.3 -12.7 0.4 Min -8.1 -6.8 -1.4 Min -13.6 -13.6 0.0 Diff. 0.8 1.0 -0.2 Diff. 1.3 1.0 0.3 Lobang Jeragan B1 18O External Internal Diff. 13C External Internal Diff. Mean -7.1 -6.9 -0.2 Mean -11.0 -12.4 1.4 Std. Dev. 0.2 0.7 -0.5 Std. Dev. 0.6 0.5 0.1 Median -7.1 -7.1 0.0 Median -11.2 -12.6 1.4 Max -6.9 -5.5 -1.5 Max -9.9 -11.5 1.6 Min -7.3 -7.7 0.4 Min -11.6 -12.9 1.2 Diff. 0.4 2.2 -1.8 Diff. 1.7 1.4 0.3 Lobang Jeragan B12 18O External Internal Diff. 13C External Internal Diff. Mean -6.7 -6.2 -0.4 Mean -11.6 -12.0 0.4 Std. Dev. 0.2 0.1 0.1 Std. Dev. 0.5 0.3 0.2 Median -6.7 -6.2 -0.5 Median -11.6 -12.1 0.5 Max -6.2 -6.0 -0.2 Max -10.8 -11.5 0.7 Min -6.9 -6.3 -0.6 Min -12.1 -12.2 0.1 Diff. 0.7 0.3 0.4 Diff. 1.3 0.7 0.6
101 Table 4-1. Continued Lobang Jeragan B28 18O External Internal Diff. 13C External Internal Diff. Mean -6.5 -7.0 0.5 Mean -10.9 -12.4 1.5 Std. Dev. 0.2 0.3 -0.1 Std. Dev. 0.6 0.3 0.3 Median -6.6 -7.2 0.6 Median -10.7 -12.5 1.8 Max -6.2 -6.6 0.5 Max -10.2 -12.1 1.9 Min -6.8 -7.3 0.5 Min -11.7 -12.7 1.0 Diff. 0.7 0.7 0.0 Diff. 1.6 0.7 0.9 Lobang Jeragan B33 18O External Internal Diff. 13C External Internal Diff. Mean -7.1 -6.5 -0.6 Mean -11.0 -12.8 1.9 Std. Dev. 0.3 0.4 -0.1 Std. Dev. 0.8 0.4 0.4 Median -7.1 -6.5 -0.5 Median -10.9 -13.0 2.1 Max -6.7 -6.0 -0.7 Max -10.2 -12.1 1.9 Min -7.5 -7.0 -0.5 Min -12.1 -13.2 1.1 Diff. 0.8 1.0 -0.2 Diff. 1.9 1.1 0.8 Gua Cha B13 18O External Internal Diff. 13C External Internal Diff. Mean -6.3 -6.2 -0.1 Mean -11.6 -12.9 1.2 Std. Dev. 1.5 0.6 0.8 Std. Dev. 0.8 0.6 0.2 Median -5.6 -6.1 0.4 Median -11.8 -12.7 1.0 Max -5.2 -5.6 0.4 Max -10.6 -12.3 1.7 Min -8.8 -7.7 -1.1 Min -13.1 -14.4 1.3 Diff. 3.6 2.1 1.5 Diff. 2.5 2.1 0.4 Gua Cha B19B 18O External Internal Diff. 13C External Internal Diff. Mean -7.7 -6.6 -1.1 Mean -14.4 -15.3 0.9 Std. Dev. 0.4 0.2 0.2 Std. Dev. 0.4 0.1 0.3 Median -7.8 -6.7 -1.2 Median -14.3 -15.4 1.0 Max -7.2 -6.3 -0.9 Max -13.9 -15.1 1.2 Min -8.1 -6.8 -1.3 Min -14.9 -15.4 0.5 Diff. 0.9 0.5 0.4 Diff. 1.0 0.3 0.7 Gua Cha B21B 18O External Internal Diff. 13C External Internal Diff. Mean -6.4 -6.8 0.4 Mean -13.9 -14.4 0.5 Std. Dev. 0.2 0.1 0.1 Std. Dev. 0.4 0.4 0.0 Median -6.4 -6.8 0.4 Median -13.9 -14.5 0.6 Max -6.1 -6.7 0.6 Max -13.5 -13.8 0.4 Min -6.5 -6.9 0.4 Min -14.3 -14.8 0.6 Diff. 0.4 0.2 0.2 Diff. 0.8 1.0 -0.2
102 Figure 4-1. Serial sampling me thods employed in this study. Samples following prism orientation Samples following prism orientation Samples following perikymata
103 Molar Layers of enamel Molar Layers of enamel Figure 4-2. Appositional nature of enamel formation. Adapted from Hilson S. 1996. Dental Anthropology. Cambridge, Cambridge University Press.
104 Figure 4-3. Picture of tooth micr ostructure showing perikymata, enamel prisms and Striae of Retzius. Adapted from Guatelli-Steinberg D ., Reid D. J., Bishop T. A. & Larsen C. S. 2005. Anterior tooth growth periods in Neandertals were comp arable to those of modern humans. Proceedings of the Nationa l Academy of Sciences 102:14197-14202. Perikymata Enamel Prisms Striae of Retzius
105 Figure 4-4. Difference in 18O ratios recovered from modern and prehistoric human molars. Prehistoric Teeth0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5Ni a h N-3 6 N i ah N-83 Niah N-160 Ni a h N-1 7 3 LJ B 1 LJ B 1 2 Lu b a n g J B 28 Lubang J B 3 3 Gu a C ha B 1 3 Gua Cha B19B Gu a Ch a B 2 1 B18O difference Growth Structure Prisms Modern Teeth0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 5New Castle 2New Castle 3Gainesville G-3Gainesville G-418O difference Growth Structure Prisms
106 Figure 4-5. Difference in 13C ratios recovered from modern and prehistoric human molars. Prehistoric Teeth0 0.5 1 1.5 2 2.5 3Ni ah N-3 6 Ni a h N-83 Niah N-160 N iah N1 7 3 L J B 1 LJ B12 Lubang J B 28 L ub a n g J B3 3 Gua C h a B 1 3 Gua Ch a B 1 9B Gua Cha B21B13C difference Growth Structure Prisms Modern Teeth0 0.5 1 1.5 2 2.5 3New Castle 2New Castle 3Gainesville G-3Gainesville G-413C difference Growth Structure Prisms
107 Figure 4-6. Distribution of 18O ratios for Western Eur ope and the Eastern US. Adapted from www.waterisotopes.org
108 Figure 4-7. Series of 18O ratios from each sampling method for the modern teeth. Very little variability is present except for New Castle 2. New Castle 3-6 -5.5 -5 -4.5 -4 -3.5 -3 -2.5 -2 0123456789 sample18O Growth Structures Prisms New Castle 2-7 -6.5 -6 -5.5 -5 -4.5 -4 -3.5 -3 012345678 samples18O Growth Structures Prisms Gainesville G-3-6 -5.5 -5 -4.5 -4 -3.5 -3 -2.5 -2 02468 samples18O Growth Structures Prisms Gainesville G-4-6 -5.5 -5 -4.5 -4 -3.5 -3 -2.5 -2 02468 samples18O Growth Structures Prisms
109 CHAPTER 5 PIERCING THE SEASONAL RO UND: USING STABLE ISOTOPES TO RECONSTRUCT HUMAN DIET BY SEASON AT WINDOVER AND HARRIS CREEK Introduction Mobility is a universal part of human existen ce that stretches back to our earliest human ancestors (Brantingham, 1998; Close, 2000). Cha nges in mobility are of ten associated with changes in food storage, trade, territoriality, social and gender inequa lity, sexual division of labor, subsistence and demography, as well as cu ltural notions of mate rial wealth, privacy, individuality, cooperation, and competition (Kel ly, 1992:44). Mobility allows hunter-gatherers to manage risk by moving away from scarce reso urces as part of their yearly seasonal round (Cashdan, 1992:237; Halstead & O'Shea, 1989; Kelly, 1983; 1992; Steward, 1938). Noneconomic factors also influence mo bility, the build up of trash a nd vermin or the depletion of local resources like firewood can necessitate movement (Smith & McNees, 1999; Wandsnider, 1992). Social factors are equally important, mobility may occur if there is a need to seek spouses, allies, shamans, or to move away from sorcery, death, or poli tical concerns (Griffin, 1989; Vickers, 1989). Traditionally, increasing socia l/political complexity is asso ciated with the emergence of agriculture and a high degree of sedentism (Christenson, 1980:52-53; Cohen, 1977; Rindos, 1983). However, areas with bountiful food resour ces allow for relatively sedentary societies with large populations and comple x political and social systems in the absence of agriculture. Many of these complex hunter-gatherer societie s subsist on aquatic resources (Erlandson, 2001; Moseley, 1975; 1992; Rick et al., 2001; Russ o, 1996). Aquatic resources can support large populations (Binford, 1968:332-333; Moseley, 1975; Sealy, 2006) and the richness and availability of aquatic resour ces foods may obviate many of th e shortages that necessitate residential mobility (Bailey & Milner, 2002; Bailey & Parkington, 1988; Binford, 1968; Cohen,
110 1977). The St Johns region of Florida is a rich en vironment where coastal, esturine and riverine environments meet producing ecotones which support a diverse fauna rich in food resources. These environments can support a variety of soci eties with levels of mobility and sedentism ranging from seasonally mobile to fully sede ntary settlements (Perlman, 1983:293; Widmer, 1988). Delineating the types of mobility and the rela tive level of sedentis m practiced by groups in the St Johns region through time could allow th e relationship between settlement pattern and social/political complexity to be examined. The St Johns River basin and the Atlantic co ast have been subject to archaeological study for over 100 years (Moore, 1892a; b; c; 1894; Wyman, 1868). Archaeological sites in the area suggest a continuous occupation of the region fr om the Paleoindian to the Mission Period. The Archaic Period in the region has received extensive study due to th e presence of large man-made freshwater shell mounds along the St Johns Ri ver and coquina middens along the coast. Researchers were quick to draw a connection between the riverine and coastal sites, by the early 1900s researchers suggested people migrated between the coastal and interi or sites on a seasonal basis occupying the coast in the winter and the river in the summer months. Subsequent researchers adopted these ideas, ofte n incorporating additional justific ations such as 'oysters taste better in the winter' (Bullen and Bullen 1961). This scenario of seasonal transhumance was adopted by Milanich and Fairbanks when they published Florida Archaeology in 1980. In their model, Archaic groups occupied base camps along the St Johns River du ring the warm months and dispersed to the coast and/or interior highlands during the cooler months (Figure 5-1). In the early 1990s, researchers began to test the model of seasonal transhumance with data from flora and fauna recovered from sites along the coast a nd river. Based on floral and faunal data, Russo et al. (1992a, 1992b) suggest that se parate ethnically distinct gr oups occupied the coast and the
111 St Johns basin year-round. Though these groups may not have been fully sedentary, they maintained separate territories a nd remained in their respective co stal or riverine zone (1992a). Regional settlement patterns and mobility are prerequisites to the study of ethnicity, immigration, trade, social complex ity or the role of monumental architecture. Unfortunately, the archaeological study of mobility is complicated by an imperfect relationship between movement and material culture (Kelly, 1992:21) and the epheme ral nature of movement that seldom leaves any material trace in the arch aeological record (Close, 2000). Archaeological studies typically assess mobility through patterns of mobility or mobility strategies and evaluating group mobility based on the organization of technology or the sourcing of raw materials (Amick, 1996; Close, 2000; Simms et al., 1997). These techniques lack the analytical power to identify the movement of individuals in th e archaeological record because the movement of material is independent from the movement of people. Ma terials may curated, cach ed, or traded down-theline. Mobility is also evaluated by the study of human remains which, unlike artifacts, cannot be decoupled from the individual. Two methods ar e common in biological studies of mobility; biomechanical markers in human bones and stable isotope analysis of human bones and teeth. Bone is a living tissue and reacts to stresses fr om repeated activities leaving evidence of the activity in its morphology. These markers can pr ovide life history information including the degree of mobility (Carlson et al., 2007; Larsen et al., 1991; Merbs, 1983; Wanner et al., 2007). For instance, in a study of pre and post-contact Native Americans in the Georgia Bight, Larsen and Ruff (1991) interpret change s in the cross-sectional geomet ry of femora as evidence of reduced mobility.
112 Stable isotopes are a common way to asse ss mobility in many animals including humans (Balasse et al., 2002; Hoogewe rff et al., 2001; Knudson & Price, 2007; Knudson et al., 2005; Koch et al., 1992; Koch et al., 1995; Prowse et al., 2003; Richards et al., 2008; Ri chards et al., 2001). Differences in light stable isotope ra tios of carbon and oxygen ca n distinguish broadly dissimilar subsistence regimes and environments, un like the analysis of ar tifacts, flora or fauna, and settlement patterns, stab le isotopes provide evidence of individual mobility (e.g. Montgomery et al., 2005; Price et al., 2000; Whit e et al., 1998). Stable isotopic studies of mobility are not subject to same problems of e quifinality as are archaeological studies based on material culture or subsistence remains. Unlike ar tifacts, stable isotope ratios are locked in tooth enamel and may not be dec oupled from the individual. The masses of the elements assessed in stable isotopic are so small that stable isotope abundances are reported in delta ( ) notation which expresses the ratio of heavy to light isotopes in a sample compared to a standard measure in parts per thousand. The two light stable isotopes most often employed in isotopic studies of diet and environment are carbon, 13C, and oxygen 18O. Differences in the fractionation of carbon isotopes in terrestrial and marine ecosystems render them useful in diet ary studies. The ratio of 13C to 12C varies in terrestrial systems due to differences in plants photosynthetic pathways Each type of photosynthetic pathway, C3, C4, and CAM (Crassulacean Acid Metabolism) produces a different range of carbon isotope values. C3 resources include almost all flora in the southeastern US, while edible C4 resources are limited primarily to maize and a few starchy seed plants like amaranths. Edible CAM plants in the Florida largely are limited to prickly pear cactus. Carbon stab le isotope ratios also differ between terrestrial and marine sy stems, the carbon values of marine systems overlap those of C4 plants and are considerably le ss negative than terrestrial C3 systems.
113 In this study 18O ratios provide the time line against which the 13C ratios are interpreted. Stable isotope ratios of 18O from surface water are determined in large part by temperature, precipitation, and evapotransporation. The 18O of precipitation results from the interaction of the ambient temperature and the amount of preci pitation. Generally, warmer weather results in enriched 18O values and cooler weather in decreased 18O values (Bryant and Froelich 1996). In temperate regions, which have relatively constant rainfall paired with varied temperatures throughout the year, 18O values are enriched during the summer months (Gonfiantini 1985). In tropical regions, which have varied rainfa ll and more constant annual temperatures, 18O values track the amount of rainfall (Njitchoua et al. 1999). Environmentally similar regions have similar seasonal 18O averages which fractionate in a predicta ble fashion according to latitude, altitude, and distance from the sea. The 18O values of precipitation are in corporated into the surface water of lakes and rivers and eventually into the biological tissues of the animals and humans (Longinelli 1984; Luz et al. 1984). Significant differences in 18O ratios between regions have been used to assess migration and immigration in archaeological contexts (Balasse et al. 2002; Bentley and Knipper 2005; Evans et al. 2006; Hoogewerff et al 2001; Prowse et al. 2003; Turner et al. 2005; White et al. 1998, 2001) This research recovers isotopic signatures from human teeth by serially sampling the dentition along growth layers. Removing multiple samples from a single tooth along an axis of growth provides a view of dietary and envir onmental change through time. Serial sampling studies are well established for a variety of fauna (Fricke and ONeal 1996; Gadbury et al. 2000; MacFadden et al. 2004; Wiedema nn et al. 1999) but are infreque ntly applied to humans. However, the few serial sampling studies of hum ans and human ancestors have been successful (Fuller et al., 2003; Richards et al., 2008; Sponheimer et al., 2 006). With this technique each
114 sample represents a period of mont hs rather than the year or mo re provided by traditional bulk techniques. This precision allows researchers to examin e differences in seasonal food consumption and to address the issue of a seasonal round using stable isotopes. Human molars from two well known Archaic s ites in the St Johns region of Florida, Windover (8BR246) and Harris Creek (8VO24), are serially sampled to examine human diet diachronically (Figure 5-3). For the first time, stable isotopes provide human dietary data with sub-annular resolution allowing is otopic assessment of a seasonal round. These data are used to test settlement models for the region outlined above (e.g. Milanich & Fairbanks, 1980; Russo et al., 1992; Russo & Ste. Claire, 1992). If populati ons were moving out the river basin to exploit marine/estuarine resources on a seasonal basis, th en evidence of this dietary change should be evident in stable carbon and oxygen isot opes recorded in their tooth enamel. Background Site Descriptions This study is based on a sample of human dentition recovered from Windover (8BR246) and Harris Creek/Tick Island (8VO24). Both sites are considered to be good representatives of their time period and are well known in the archaeol ogical literature of th e Southeastern US and of the Archaic Period. Windover (8BR246) The Windover site is an Early Archaic cemetery located near Titusv ille, Florida (Doran 2002). The site was discovered in 1982 while a small pond was being filled in preparation for the construction of a subdivision (Doran 2002). The site was excavated by Dr. Glen Doran and colleagues over three field seasons fro m 1984 to 1986 (Doran 2002). The anaerobic environment afforded excellent preservation which resulted in an extensive artifact assemblage including textiles, floral remains, and tools made of bone, antler, and marine shell (Doran 2002).
115 At least 168 human skeletons were recovered from the shallow but persistent pond (Doran 2002). Radiocarbon dates on the human sk eletal material indicate the interments occurred between 7,100 and 7,330 rcybp (Doran 2002). Burials were fl exed on their side and held down with matting or fabric and wooden stakes in shallow hollows underwater (Doran 2002). Harris Creek (8VO24) Prior to its destruction, Harris Cr eek, also referred to as Tick Island, was a series of shell middens and ridges with a substantial bean-sha ped platform mound constr ucted of freshwater shell estimated to have been 5.5 to 10.7 m above ground level (Aten 1999). The site was first investigated by Moore in 1891. Bu llen conducted excavations of the northeastern slope of the mortuary mound in 1961 and recovered more than 175 burial features. Fi rsthand accounts from people involved in shell mining operations assert that mining activities destroyed hundreds of burials prior to Bullens mitigation (Aten 1999). Th is suggests the original number of interments at the site greatly exceeded t hose recovered by Bullen (Aten 1999). Jahn and Bullen (1978) reported 4 conventiona l radiocarbon dates from samples taken from his Harris Creek excavations two of which bracket the mort uary layers (Table 5-1). Charcoal at the base of layer 3 produced a radiocarbon date of 5450 rcybp (M-1264) and charcoal from layer 8 produced a date of 5320 rcybp (M-1265) (Aten 1999). Recently, four new AMS dates on human bone, two from layer 3 and two from layer 4, produced older dates than those reported by Jahn and Bullen. Two indivi duals from layer 3 produced dates of 6125 rcybp (X-9110) and 6053 rcybp (X -9112RA). Two individuals from layer 4 date are slightly younger in age at 5904 rcybp (X-9109A) and 5825 rcybp (X-9111A). There is considerable variation present in the origin of the individuals buried at Harris Creek/Tick Island. Recent work by Quinn et al. (2 008) identified four isotopically distinct immigrants in a sub-sample of the burial population. Is otope ratios of 13C and SR87/SR86
116 indicate two immigrants were from a coastal z one. Two other immigrants were identified based on their 18O values which suggest they originated as far north as Tennessee or Virginia. Previous Studies of Paleodiet at Windover and Harris Creek In an unpublished study, Tucker and colleague s (2007) examined paleodiet at Windover and Harris Creek and found both populations had enriched 13C ratios which indicated the consumption of marine resources. Bone collagen 13C ratios from Windover (Tuross et al., 1994) and Harris Creek are compared to Woodland Period hunter-gatherers fr om the interior of Georgia (Tucker, 2007), Late Arch aic inhabitants of Horrs Island, Florida (Kelly et al., 2006), and St. Johns Period remains from the nearby Ross Hammock site (Tucker et al. 2007) (Figure 54). The Georgia population did not have access to a significant amount of marine resources, confirmed by their average bone collagen 13C ratio of -19.5. Ross Hammock and Horr's Island have both been interprete d as year-round coastal occupati ons and their populations have 13C collagen ratios of -10.6 and -10.2, resp ectively. The averag e bone collagen 13C ratio from Windover is -15.6 (Tuross et al., 1994) and Harris Creek has an average value of 16.5. These ratios are between the interior Ge orgia population and the coastal populations at Ross Hammock and Horrs Island. The interm ediate position of Windover and Harris Creek indicates a marine component in the diet. Other is otopic indicators including 15N ratios and collagen-to-apatite spacing support the pres ence of a substantial marine component. Season of Site Occupation Establishing the time season of occupation for riverine and coastal sites is critical to understanding Archaic settlement patterns. Archaic interior sites with seasonality data include Windover, Groves Orange Midden, and to a lesser extent Munroe Outlet Midden (Table 5-2). The Early Archaic Windover site predates Harris Creek by ca. 2000 radiocarbon years. Botanical remains including wax myrtle, cabbage palm, holly, hickory, bottle gourd, oak, black
117 gum, hackberry, grape, maypop, pe rsimmon, nightshade, elderber ry, and prickly pear, were recovered from Windover and indicate a summer to fall use of the mortuary site (Tuross et al. 1994). Supporting evidence comes from the stakes used to keep the bodies submerged in the lake; study of the growth rings in these stakes indicate they were cut dur ing the late summer or fall. Groves Orange Midden is a larg e freshwater Middle to Late Archaic shell midden (3100 to 5700 BP) partially submerged in Lake Monroe (Vol usia County). The remains of hickory nuts, acorns, persimmons and grapes suggest occupation of the site occurred in the summer and fall (Russo et al. 1994). Lake Monroe Outlet Midden, like Groves Orange Midden, is a Middle Archaic site partly submerged in Lake Monroe. Botanical remains recovered from the site, including blueberries, hickory nuts and acorns, indicate a late summer/fall occupation. However, Ruhl (2001) cautions that mast are often stored and that their se ason of ripening is not a good indicator of site occupation. Coastal sites with seasonality data include Cotten, Tomoka Stone, Crescent Beach, Rollins, Spencers Midden, and the Guana Shell Ring. Se asonality data for Tomoka and Cotten are derived primarily from the size classes of fish and shellfish while the data from Crescent Beach, Rollins, Spencers Midden, and Guana are the result of isotopic studies and the sclerochronology of shellfish recovered from the middens at these sites. The Cotten site is a Late Archaic coquina mi dden on the west bank of the Halifax River in Ormond Beach, Florida. Investigations by Hale (1984) suggest coquina was collected in the summer months, the size classes of the coquina coincided with modern collections from June.
118 Hale (1984) further suggests the site may be used in the winter months based on the presence of migratory birds. Tomoka Stone is primarily a Late Archaic c oquina midden located just north of the Cotten site. Excavations uncovered a s ubstantial amount of fiber temper ed pottery and several Archaic stemmed and side notched points. Russo and Ste. Claire (1992) assessed the season of capture of coquina, quahog, pinfish, and croakers. All of these, except the quahog, suggest a summer or early fall occupation. The qua hogs are distributed throughout the midden in small amounts and appear to have been collected in the winter and/or spring mont hs. The only botanical remains recovered from the site are hi ckory nuts, which tentatively support a summer/fall occupation. Spencers Midden, Crescent Beach, Rollins, a nd Guana were surveyed by Jones et al. (2005) who used oxygen isotopes recovered from c oquina to estimate paleotemperature and the season of collection. Of these site Spencer's Midden is the oldest, ca. 5570 rcybp, and is likely contemporaneous with Harris Creek. Crescent Beach is a late Middle Archaic site which dates to around 4240 rcybp. Rollins and Guana Shell Ring are both Late Archaic sites and date to 3760 and 3600 rc ybp respectively. Though Jones et al. (2005) only evaluated a single shell from each of the four archaeological sites, all the results suggest the coquina were collected during the fall. With the exception of a few quahog from Tomo ka Stone and the scattered remains of migratory birds, the floral and faunal remains from coastal and riverine sites in the region suggest occupation in the summer and fall; a pattern which holds from the Early to Late Archaic. Data suggest both the coast and river were occupi ed during the summer and fall. The floral and faunal data do not indicate where the Archaic groups spent the winter and spring. However, this may be due to inherent difficulties involved in floral and faunal assessments of seasonality rather
119 than suggesting the Archaic inhabitants wintered el sewhere. Most floral resources ripen during the warm months but many, like mast, might be stored and consumed during the winter and spring limiting their usefulness as markers of seas onality. Many of the estimates for coastal sites rely on isotopic studies of coquina, however, co quina have a growth season limited to the warm months making the winter months isotopically invisible. These limitations make flora and fauna ill-suited to identify winter occupations. Serial Sampling A study of human diet by season can detect the consumption of marine vs. non-marine foods throughout the year, including the winter months, possibly indicat ing the season of occupation. Unlike coquina, tooth enamel conti nuously records the isotopes ratios of food and water consumed during enamel formation. Temperat ure and rainfall vary in a predictable pattern with the season, allowing 18O variation recorded in tooth enam el to indicate the season that unit of enamel was deposited. Ratios are also record ed in enamel and are recovered from the same samples as the 18O ratios. The 18O ratios can provide a seasonal time scale against which the 13C can be evaluated. If significant amount of ma rine foods are consumed during a portion of the year, then this difference should be visible in the isotopic ratios record ed in the tooth enamel. This research employs a serial samp ling methodology to recover variation in 13C and 18O ratios recorded in human tooth enamel. In se rial sampling or isotopi c zoning studies (Kohn & Cerling, 2002) small units of enamel are removed from a tooth along the axis of growth. Fauna are frequently sampled using this method (Fricke & O'Neil, 1996 ; Gadbury et al., 2000; MacFadden et al., 2004; Wiedema nn et al., 1999), and these studies range from identifying seasonal seaweed consumption by sheep (Balas se et al., 2006; Bala sse et al., 2005) to paleoclimatic reconstructions based on ty rannosaurid teeth (Straight et al., 2004).
120 Many factors complicate serial sampling human teeth including their relatively small size and concerns over isotopic buffering common to a ll serial sampling studies of tooth enamel. Isotopic signals can be attenuated due to many factors before and after inge stion by an organism. These factors include a reservoir effect as precipi tation is collected, mixed, and stored in lakes and rivers (Balasse, 2003; Kohn & Cerling, 2002) A similar effect occurs once water is consumed by an organism when oxygen isotopes are stored and mixed in body water before they are locked into tooth enamel (B alasse 2003; Kohn and Cerling 2002). The primary concern in serial sampling huma n and faunal teeth is isotopic attenuation caused by secondary mineralization, a process by which the mineral content of enamel slowly increases after initial deposition prolonging the period of enamel formation and divorcing the isotopic signal for a given unit of enamel from the time period of its formation (Balasse, 2003; Kohn & Cerling, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). However, a recent study of mammalian enamel demonstrates th at isotopic sampling of growth layers between the inner and outter enamel recovers isotopic variation whic h remained intact afte r the maturation phase (Humphrey et al., 2007). Humphrey et al. (2007) identified cha nges in strontium/calcium ratios in the human infant enamel before and after birth and these changes were retained after mineralization was complete. Additionally, dental studies indicate the ma jority of the carbonate in mature enamel is present during deposition, therefore, secondary mineraliz ation is not likely to complicate isotopic assessments of carbonate (Aoba & Moreno, 1990; Elliott et al., 1985). The incremental nature of human tooth gr owth allows sampling of enamel which encapsulates isotopic ratios representing discrete units of time. Human teeth begin to form as enamel is deposited on a few millimeters of den tin at the tip of the cusp. (Avery 2000). Additional layers of dentin and enamel are laid down as the tooth grows from the cusp
121 downward until the apex of the root is complete (Dean 1989). Sequential layers of enamel appear as stacked cones on the surface of the tooth after formation is comp lete (Figure 5-5). The edges of these layers form a system of ridges and furrows on the tooth surface called perikymata which are the surface manifestation of brown striae of Retzius; each perikymata takes ca. 9 days to form (Dean et al. 2001 ; FitzGerald 1998). Methods A series of six to eight enamel samples, between .25 and .5mg were removed from each tooth under magnification with a low speed dental dr ill parallel to the peri kymata (Figure 5-7). Samples were not chemically pretreated and were placed directly into pre-weighed stainless steel capsules and loaded into a multiprep inlet system attached to a VG PRISM II mass spectrometer. Results A total of 60 samples were obtained from 10 individuals from Windover and the fifteen individuals sampled from Harris Creek produced a total of 103 samples. The average number of samples per individual from Wi ndover was 6.0 and 6.4 from Harris Creek. Summary statistics for the samples are included in Table 5-3. The amount of buffering in the samples wa s greater than expected. The average difference between the minimum and maximum 18O values for Windover was 1.0 and the average difference in minimum and maximum 13C ratios was only 1.7. Similarly, the average difference between minimum and maximum scores from Harris Creek was 1.1 for 18O and 1.5% for 13C. These values suggest a large amount of time averaging is occurring in these samples. This averaging may be due to the am ount of time represented by each sample. Perikymata counts were obtained for a subset of the dentition (Table 5-4). In some cases the perikymata were visible on the tooth surface under magnification. When possible, the number of
122 perikymata per sample were noted. These c ounts were performed under magnification, but are considered estimates of the act ual number of perikymata. Enamel deposition does not occur at a consta nt rate and as a result perikymata are not spaced evenly across the surface of the tooth. Th e perikymata are packed more tightly near the cervical region of each tooth and grow wider near the crown. Therefore, the first few samples in each series span a greater number of days than the last few samples in the series. In many cases the first sample in a series spans an estimated dur ation of ca. four months while the last sample may only encompass a month. The large amount of time incorporated by the first half of each series results in the greater amount of tim e averaging and isotopic buffering. The unequal distribution of time distribute across the tooth also complicates the interpretation of the 18O ratios as a reflection of seasonality. Though sinusoidal patterns result from the sampling of these teeth, the structure of these signals cannot be interpreted directly as a seasonal pattern. Discussion To overcome the difficulties imposed by the unequal amount of time encompassed by each sample, correlation coefficients were calculated for each data se ries. If the diets of these individuals varied in a predic table fashion by season, then the 18O and 13C values should be related in a positive or negative correlation regard less of the time encompassed by each sample. Pearson Product Moment Correlation Coefficients were calculated for each data series in SAS Enterprises JMP 7.0.1 and the results and as sociated p-value are shown in Table 5-3. The 13C and 18O ratios are positively correlated at the 0.10 level in two individuals (92.76 and 99.5) from Windover. The 13C and 18O ratios from Harris Creek were also correlated by the correlation showed more varia tion. Of seven significant corr elations (p>.10), three were significantly negatively correlated (8, 100/ 101, 128) and four (108, 132, 141, 147/148) were significantly positively correlated. These correlat ions indicate there is a relationship between
123 diet and season. A positive correlation indicates the 13C values are enriched in tandem with the 18O values, which suggest increased marine f ood intake during the summer months. The negative correlation found at Harr is Creek indicates ju st the opposite, that more marine food was consumed in the colder months. However, closer inspection of the negati vely correlated signals from Harris Creek reveals that they appear to be positively correlated over much of the sequence, suggesting the correlation could be an artifact of sampling or an atypical event recorded in the tooth enamel. These discrepancies might also result from the correlation methodology itself. Correlation, like regression, quantifies the linear relationship between two variables (Bewick et al., 2003). If a relationship is non-linear, as thes e sinusoidal signals sugg est, then correlation might not detect or adequately describe all re lationships (Bewick et al., 2003). Consequently, more qualitative visual inspection of the structure of the isotopic series may be better suited to these data. When inspected visual ly, it is apparent the 13C and 18O ratios from Windover also vary in concert with one another; th is relationship is especially ap parent in individuals 99.5 and 69.1 (Figure 5-7). Additionally, if the first two to three data points are excl uded from each sequence (on average these points span mu ch more time then the latter pa rts of the series) then 81.11 also appears highly correlated. Visual inspection also suggests the sequences from Harris Creek are positively correlated (Figures 5-8 and 5-9). In most cases the shape of each line mirrors the other to a greater or lesser degree. This trend is especially apparent in individuals B108, B132, B141, and B170. Like the Windover sequences, if the first few data points are dropped, other sequences become similar as well, e.g. B128, B91, and B36.
124 The visual inspection of the data reinfo rces the positive correlation suggested by the Pearson Product Moment Correlation Coefficients. In terms of diet, th ese results indicate the populations interred at both sites co nsumed foods more enriched in 13C during the warmer months when surface waters were more enriched in 18O: they ate more marine foods in the summer. Though previous work has suggested the presence of local and n on-local individuals at Harris Creek, there is no clear difference between the signals of these individuals. The results of this study, like those of previous more tr aditional isotopic studies of paleodiet, do not support the presence of separa te ethnically distinct groups occupying the river drainage and the coast (Russo et al., 1992; Russo & Ste. Claire, 1992). Neither Windover or Harris Creek is found on the coast, yet individual s at both sites included a significant portion of marine/estuarine foods in their di et suggesting they had direct inte raction with the coast. This interaction is also supported by the presence of marine artifacts at Windover and Harris Creek. Marine shells ( Busycon carica, Busycon contrarium, Strombus gigas ) and shark's teeth have been recovered from most, if not all, Mt. Taylor sites, in cluding Windover and Harris Creek (Aten, 1999; Doran, 2002; Wheeler et al., 2000). The presence of marine resources at interior riverine sites suggests Mt. Taylor people had access, at least indirect ly, to coastal resources. The results of this study, both in terms of quantitativ e correlations and qualitativ e visual observations, indicate the marine porti on of their diet was consumed during the warm portion of the year. Interestingly these results are opposite to the Milanich and Fairbanks (1980) model which suggested a winter occupation of the coast. In terms of the overall pattern and in sp ecific isotopic values, Windover and Harris Creek are virtually indistinguishable. This suggests that the pattern of transhumance and seasonal resource use exhibited by the population at Ha rris Creek was in place more than 2000 years
125 earlier at Windover. The continuity in resour ce scheduling supports th e previous study of paleodiet based on bone isotopes, which suggested there was l ittle to no dietary difference between the population at Windover a nd the population at Harris Creek. Conclusion For the first time isotopic data with sub-annual resolution was extracted from human teeth. These new data, though attenuated, provide an alternative means of evaluating Archaic settlement models for the St Johns region. Thes e data are interpreted to suggest the populations interred at Windover and Harris Creek consumed more marine food during warm months than during the cold months. These new data support previous studies which indicated a significant marine component in the diet of people from Windover and Harri s Creek. Based on these results and the distance of these sites from the sea, it is likely these hunt er-gatherer groups were residentially mobility and occupied the coast or an estuarine environment for a least a portion of each summer. Interestingly, the results from the populat ion from Windover and the population from Harris Creek who lived over 2000 years later are al most identical. These results suggest that contrary to conventional wisdom significant changes in subsis tence and resource scheduling did not occur between the Early and Middle Archai c despite the rise of new innovations, like freshwater shell mounds and long distance trade (Wheeler et al., 2000). These results provide another source of evidence suggesting innovation s like mound construction and establishing long distance trade networks could o ccur in mobile groups. Though s uggestive, these data must be viewed as preliminary, until more populations from the region are sampled and better control of the amount of time in each sample is achieved.
126 Table 5-1. Conventional and AM S radiocarbon dates from Harri s Creek. The new AMS dates are calibrated with CALIB 5.0.1 (Stuiver & Reimer, 1993). Lab No. Layer Sample Uncorrected 14C years B.P. Calibrated Dates, 2 sigma BP Reference M1264 3 charcoal 5450300 [5590 BP:6912 BP] 0.999 [6921 BP:6929 BP] 0.001 Aten 1999 M1265 8 charcoal 53200 [5645 BP:6494 BP] 1.00 Aten 1999 X9110 3 human bone (Burial 7) 6125 [6793 BP:7179 BP] 0.96 [7195 BP:7244 BP] 0.04 New date, this paper X9112R A 3 human bone (Burial 31) 6053 [6742 BP:7030 BP] 0.90 [7043 BP:7069 BP] 0.02 [7077 BP:7086 BP] 0.01 [7096 BP:7156 BP] 0.07 New date, this paper X9109A 4 human bone (Burial 3) 5904 [6563 BP:6592 BP] 0.03 [6598 BP:6891 BP] 0.97 New date, this paper X9111A 4 human bone (Burial 9) 5825 [6482 BP:6758 BP] 0.98 [6761 BP:6782 BP] 0.02 New date, this paper
127 Table 5-2. Season of site occupation of Archaic sites in the St. Johns Region. Site Interior or Coastal Period Season of Occupation Reference Windover Interior Early Archaic Summer/Fall Tuross et al. 1994 Groves Orange Midden Interior Middle to Late Archaic Summer/Fall Russo et al. 1994 Monroe Outlet Midden Interior Middle Archaic Summer/Fall ACI and JANUS 2001 Cotten Coastal Late Archaic Summer, possible Winter Hale 1984 Tomoka Stone Coastal Late Archaic Summer/early Fall Russo and Ste. Claire 1992 Crescent Beach Coastal Middle Archaic Fall Jones et al. 2005 Rollins Coastal Late Archaic Fall Jones et al. 2005 Spencers Midden Coastal Middle Archaic Fall Jones et al. 2005 Guana Shell Ring Coastal Late Archaic Fall Jones et al. 2005
128 Table 5-3. Summary statis tics from Windover (8BR246) and Harris Creek (8VO24).Site Individual Correlation # of samples Significance mean 18O Minimum Maximum Difference mean 13C Minimum Maximum Difference 8BR246 184.108.40.206 0.41 7 0.36 -0.7 -1.1 -0.3 -0.8 -9.7 -10.3 -9.0 -1.3 8BR246 57.3 0.22 5 0.72 -1.6 -2.5 -1.1 -1.4 -11.6 -12.0 -11.3 -0.7 8BR246 69.1 0.4 7 0.38 -1.2 -1.8 -0.7 -1.1 -11.2 -12.0 -10.0 -2.0 8BR246 81.11 0.34 6 0.50 -1.0 -1.4 -0.6 -0.8 -10.9 -11.9 -9.6 -2.3 8BR246 83.26 0.55 5 0.33 -1.0 -1.3 -0.8 -0.5 -9.3 -9.9 -8.8 -1.1 8BR246 92.76 0.97 6 0.00 -1.1 -1.5 -0.6 -0.9 -10.5 -11.6 -8.8 -2.8 8BR246 95.1 0.46 8 0.25 -0.6 -1.0 -0.3 -0.7 -9.5 -10.7 -8.3 -2.4 8BR246 99.5 0.98 4 0.02 -1.0 -1.5 -0.2 -1.2 -11.0 -11.6 -10.1 -1.5 8BR246 130.28 0.56 7 0.19 -1.0 -1.4 0.6 -2.0 -10.6 -11.3 -10.0 -1.3 8BR246 143.40 0.76 5 0.14 -1.1 -1.5 -0.4 -1.1 -11.0 -11.7 -10.4 -1.2 Mean 0.57 6.00 0.29 -1.0 -1.5 -0.4 -1.0 -10.5 -11.3 -9.6 -1.7 8VO24 4 -0.0001 7 1.00 -0.5 -1.1 -0.1 -1.0 -10.3 -11.2 -9.0 -2.2 8VO24 8 -0.7 7 0.08 -1.4 -1.8 -0.9 -0.9 -11.2 -11.5 -10.8 -0.6 8VO24 35 0.6 6 0.22 -2.2 -2.6 -1.4 -1.1 -10.2 -10.9 -9.7 -1.2 8VO24 35 -0.4 6 0.43 -2.7 -3.2 -2.4 -0.8 -11.0 -11.4 -10.6 -0.8 8VO24 36 0.12 6 0.82 -0.8 -1.1 -0.4 -0.7 -10.7 -11.1 -10.1 -1.0 8VO24 38 0.49 7 0.26 -1.2 -1.8 -0.7 -1.1 -9.2 -10.5 -8.6 -2.0 8VO24 46 -0.07 7 0.88 -1.1 -1.4 -0.8 -0.6 -11.0 -11.6 -10.5 -1.1 8VO24 91 0.29 6 0.58 -1.1 -1.6 -0.6 -1.0 -9.3 -10.5 -8.2 -2.3 8VO24 100/101 -0.73 7 0.06 -1.6 -2.3 -1.2 -1.1 -11.6 -12.4 -10.5 -1.9 8VO24 108 0.96 5 0.01 -1.7 -2.8 -0.7 -2.2 -11.3 -12.3 -10.4 -1.9 8VO24 128 -0.88 6 0.02 -1.4 -2.7 -0.7 -2.0 -11.1 -11.6 -10.3 -1.3 8VO24 132 0.92 6 0.01 -1.3 -2.0 -0.4 -1.6 -10.4 -11.8 -9.3 -2.6 8VO24 135 0.45 6 0.37 -1.3 -1.6 -1.0 -0.6 -11.3 -11.7 -10.9 -0.8 8VO24 141 0.8 10 0.01 -1.0 -1.7 -0.4 -1.3 -11.4 -12.8 -10.9 -1.9 8VO24 147/148 0.87 5 0.05 -1.2 -2.2 -0.6 -1.7 -9.4 -10.0 -9.0 -1.0 8VO24 170 0.2 6 0.70 -0.5 -0.7 -0.2 -0.4 -10.1 -10.6 -9.6 -1.0 Mean 0.18 6.44 0.34 -1.3 -1.9 -0.8 -1.1 -10.6 -11.4 -9.9 -1.5
129 Table 5-4. Perikymata counts for select ed human molars from Harris Creek. Individual Sample # # of perikymata Avg # of days Individual Sample # # of perikymata Avg # of days 8VO24-4 1 10 90 8VO24-128 1 14 126 8VO24-4 2 0 8VO24-128 2 9 81 8VO24-4 3 0 8VO24-128 3 5 45 8VO24-4 4 0 8VO24-128 4 4 36 8VO24-4 5 0 8VO24-128 5 4 36 8VO24-4 6 7 63 8VO24-128 6 0 8VO24-4 7 0 8VO24-132 1 15 135 8VO24-8 1 6 54 8VO24-132 2 15 135 8VO24-8 2 7 63 8VO24-132 3 10 90 8VO24-8 3 5 45 8VO24-132 4 8 72 8VO24-8 4 5 45 8VO24-132 5 0 8VO24-8 5 3 27 8VO24-132 6 0 8VO24-8 6 4 36 8VO24-135 1 11 99 8VO24-8 7 0 8VO24-135 2 7 63 8VO24-35 1 9 81 8VO24-135 3 6 54 8VO24-35 2 0 8VO24-135 4 5 45 8VO24-35 3 6 54 8VO24-135 5 4 36 8VO24-35 4 4 36 8VO24-135 6 0 8VO24-35 5 5 45 8VO24-141 1 13 117 8VO24-35 6 0 8VO24-141 2 11 99 8VO24-36 1 12 108 8VO24-141 3 10 90 8VO24-36 2 6 54 8VO24-141 4 8 72 8VO24-36 3 4 36 8VO24-141 5 5 45 8VO24-36 4 3 27 8VO24-141 6 6 54 8VO24-36 5 0 8VO24-141 7 4 36 8VO24-36 6 0 8VO24-141 8 3 27 8VO24-38 1 11 99 8VO24-141 9 0 8VO24-38 2 8 72 8VO24-141 10 0 8VO24-38 3 7 63 8VO24-147.148 1 8 72 8VO24-38 4 4 36 8VO24-147.148 2 6 54 8VO24-38 5 3 27 8VO24-147.148 3 5 45 8VO24-38 6 3 27 8VO24-147.148 4 0 8VO24-38 7 0 8VO24-147.148 5 0 8VO24-46 1 10 90 8VO24-170 1 0 8VO24-46 2 8 72 8VO24-170 2 6 54 8VO24-46 3 5 45 8VO24-170 3 7 63 8VO24-46 4 4 36 8VO24-170 4 0 8VO24-46 5 4 36 8VO24-170 5 0 8VO24-46 6 4 36 8VO24-170 6 0 8VO24-46 7 0 8VO24-100.101 1 20 180 8VO24-100.101 2 16 144 8VO24-100.101 3 11 99 8VO24-100.101 4 10 90 8VO24-100.101 5 6 54 8VO24-100.101 6 0 8VO24-100.101 7 0
130 Figure 5-1. Milanich and Fairba nks (1980) model of movement between the river and coast. S t J o h n s R i v e rA t l a n t i c C o a s t S t J o h n s R i v e rA t l a n t i c C o a s t
131 Figure 5-2. Russo and Ste. Claires (1992) model of movement up and down the river and coast by separate ethnic groups. S t J o h n s R i v e rA t l a n t i c C o as t S t J o h n s R i v e rA t l a n t i c C o as t
132 Figure 5-3. The location of Windover (8BR246), Harri s Creek (8VO24) and othe r Archaic Period sites in the region. N
133 Figure 5-4. The 13C ratios of bone collagen from selected pr ehistoric sites. Note the intermediate position of Windover and Harris Creek. 13C ratios from bone collagen -20-19-18-17-16-15-14-13-12-11-10 Horrs Island Ross Hammock Georgia Hunter-Gatherers Harris Creek/Tick Island Windover13C Terrestrial C3 Marine or C4
134 Figure 5-5. Appositional nature of enamel fo rmation. Adapted from Hilson S. 1996. Dental Anthropology. Cambridge, Cambridge University Press. Molar Layers of enamel Molar Layers of enamel
135 Figure 5-6. Picture of tooth microstr ucture showing perikymata, enamel prisms and Striae of Retzius. Adapted from Guatelli-Steinberg D., Reid D. J., Bishop T. A. & Larsen C. S. 2005. Anterior tooth growth periods in Nea ndertals were comparable to those of modern humans. Proceedings of the National Academy of Sciences 102:14197-14202. Perikymata Enamel Prisms Striae of Retzius
136 Figure 5-7. Sampling tracks on a human molar.
137 Figure 5-7. Series of 18O and 13C ratios from Windover. 1234567 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-220.127.116.11 18O 13C 12345 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-57.300 18O 13C 1234567 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-69.1 18O 13C 123456 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-81.11 18O 13C 12345 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-83.26 18O 13C 1234567 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-92.76 18O 13C 1234 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-99.5 18O 13C 12345678 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-95.1 18O 13C 1234567 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-130.28 18O 13C 12345 -13.00 -12.00 -11.00 -10.00 -9.00 -8.00 -4.00 -3.00 -2.00 -1.00 0.00 1.00 8BR246-143.40 18O 13C
138 Figure 5-8. Series of 18O and 13C ratios from Harris Creekindividuals 4, 8, 35, 36, 38, 46, 91, 100/101. 1234567 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B4 18O 13C 1234567 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B8 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B35 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B36 18O 13C 1234567 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B38 18O 13C 1234567 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B46 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B91 18O 13C 1234567 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B100/101 18O 13C
139 Figure 5-9. Series of 18O and 13C ratios from Harris Creekindividuals 108, 128, 132, 153, 141, 147/148, 170. 12345 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B108 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B128 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B132 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B135 18O 13C 12345678910 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B141 18O 13C 12345 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B147/148 18O 13C 123456 -13 -12 -11 -10 -9 -8 -4 -3 -2 -1 0 1 8VO24-B170 18O 13C
140 CHAPTER 6 CONCLUSION This research uses stable isotope analysis and models of human mobility to investigate three common themes in the study of prehistoric hunter-gatherers: egalitarianism, subsistence, and settlement. The revisionist critique of hunter-gatherer studies dismissed simple causal relationships between mobility, social ranking, food production, and land use. However, many aspects of cultural historical models constructed pr ior to this critique have not been reexamined. Before new models of ethnicit y, immigration or landscape archaeology can be applied to the St Johns region of Florida, many a ssumptions based on dated culture hi stories need to be assessed. This research examines various aspects the cultural history of the region w ith data unavailable to archaeologists. Stable isotopes are robust proxies for diet and environment that provide independent data which can help to avoid much of the equafinality found in archaeological studies. Stable isotopes can directly assess m obility and diet at the individual level. Though mobility, social complexity and food production are no longer linked in cultural evolutionary models, the study of the different ways in wh ich these variables interact can enhance our understanding of the diversity of hunter-gatherer social organizati ons that existed in the past. In Chapter 2 Isotopic Investigations of Middle Archaic Mortuary Practices at the Harris Creek Site, Tick Island, Florida e xplanations for the presence of primary and secondary burials at Harris Creek are tested with st able isotopes. Two models common in mortuary archaeology could explain the different types of burials f ound at the site. The first model explains the differences in burial type as reflec ting status distinctions of the i ndividuals in life. According to Tainter (1975) the complexity of body treatment a nd the duration of mortuary ritual is associated with social rank (Tainter 1975). If this inte rpretation holds true, th en these burials would represent the earliest ex ample of social ranking in North America (Aten, 1999). Alternatively,
141 the differences in burial treatment could be a re sult of temporal/geo-spa tial considerations of corpse disposal. For seasonally mobile groups of hunter-gatherers deat h of a member might occur a considerable distance from the desired si re of interment. In such cases, the deceased must be processed and stored until the group arri ved at the appropriate place and time for burial (Buikstra & Charles, 1999; Hofman, 1985). To test these models, 13C and 18O ratios were sampled from the tooth enamel of the burials. Analysis of the 13C and 18O ratios revealed two groups we re present at the site, one with local 18O values and one with enriched 18O values indicating a non-local southern origin. All of the local burials were s ubjected to post-mortem processing as indicated by their secondary (tightly flexed) burial but only a portion of the nonlocal southern group received similar secondary treatment. The differences in bur ial treatment among the non-local southern group may be evidence of rank or status among the burials, but this is not supported by isotopic differences in diet or by the in clusion of grave goods with these burials. The presence of primary and secondary burials at Harris Cr eek does not appear to be relate d to curation or transportation of the dead. Instead, the differences may be ev idence of multiple ethnicities or religions among those people who joined the lo cal population during adulthood. The 18O ratios of the people interr ed at Harris Creek suggest that this population was composed of people from up and down the length of the St Johns River. Additionally, Quinn et al. (2008) used 18O and 86Sr/87Sr ratios to show several indivi duals interred at Harris Creek originated far from the St Johns River basin, with two from as far away as Tennessee or Virginia. These studies indicate the populat ion buried at Harris Creek is an amalgamation of individuals from across Florida and further to the north. This suggests buri al mounds like Harris Creek may have been points of contact where people from different groups came together for social,
142 economic, and ideological reasons. Rather than le gitimizing a single groups right to resources or serving as a territorial marker, Harris Creek may be viewed as a point of articulation for multiple groups of people spread acr oss the Archaic landscape. Chapter 3 Dietary Continuity and Change durin g the Florida Archaic uses stable isotopes from human bone to assess the contribution of marine foods and riverine res ources in the diets of populations at Windover and Harris Creek. A new and intense focus on riverine resources is a defining characteristic of the Middle Archaic Period in Florida (Milan ich, 1994; Milanich & Fairbanks, 1980). If this shif t in subsistence occu rred then it should be apparent in the 13C and 15N ratios from these sites. Supporting evidence fo r a shift to riverine resources comes from the appearance of large freshw ater shell mounds along the river. Ho wever, there is some question as to whether the shell mounds represent food refuse or the use of shell as a building material, or both. 13C and 15N ratios are used to assess the relative c ontribution of freshw ater shellfish to the diets of both populations. Finally, artifacts made from marine reso urces are common at Mt. Taylor sites but marine food resources are not. Settlement models for th e region suggest people occupied the coast on a seasonal basis (Milanic h & Fairbanks, 1980) or th at people restricted their range to the St Johns river drainage and did not occupy the coast (Russo & Ste. Claire, 1992). Based on these models any evidence of marine food consumption is interpreted as a seasonal occupation of the coast. The results of the isotopic assay of the populations from Harris Creek and Windover were remarkably similar. There was no evidence of a massive shift to freshw ater aquatic resources, though the individuals interred at Harris Creek may have consumed slightly more riverine resources than did the people at Windover. Cont rary to other sources of data (Russo et al., 1992), the isotopic data do not indica te that freshwater shellfish were a significant component of
143 their diet. Surprisingly, both populat ions appear to derive a signifi cant portion of their diet from marine or estuarine sources supporting a coastal occupation. Based on isotopic data from these two sites, no major change in subsistence o ccurred between the Early and Middle Archaic and innovations like the construction of large fres hwater shell mounds must be explained by something other than a broad scale change in diet and settlement. In Chapter 4 Evaluating a Serial Sampling Methodology, modern and prehistoric teeth from Southeast Asia are used to evaluate di fferent serial sampling methodologies. Serial sampling or isotopic zoning stud ies (Kohn & Cerling, 2002) have proven effective in a number of cases but have been hampered by a concer n about isotopic attenu ation produced by the process of secondary mineralization (Balasse 2003; Kohn & Cerling, 2002; Passey & Cerling, 2002; Zazzo et al., 2005). After the initial de position of enamel, s econdary mineralization slowly increases the mineral content of the en amel during the maturati on phase. This process introduces mineral into the sample not associated with the original growth structure. Since the majority of mineral content is in troduced after deposition, the isot opic signal associated with the growth structures could be writ ten over or swamped by the subs equent waves of mineralization (Balasse, 2003; Kohn & Cerling, 2002; Pa ssey & Cerling, 2002; Zazzo et al., 2005). To assess the effects of secondary minerali zation, 15 human molars were sampled along two axes. The method that returned the most is otopic variation was assumed to have the least amount of isotopic attenuation. Th e results of the study indicate th at more isotopic variation is recovered from sampling the visible growth structures than along the path of the enamel prisms. A review of carbonate substitutions in enamel supports the assertion that secondary mineralization introduces little additional carbonat e into the enamel after deposition (Elliott et al., 1985). These results explain why many studies of serially sample enamel carbonates have
144 produced valid results but suggest secondary mi neralization remains a concern for researchers sampling phosphates. In addition to investigating the effects of secondary mineralization, the modern and prehistoric teeth were compared to evaluate whet her the signals recovered from tooth enamel are the result of environmental input or the result of internal bi ological rhythms which produce signals that mimic those expected from climatic s ources. If the variation is environmental, then the pooling and storage of drinki ng water in reservoirs and the consumption of bottled and/or canned beverages is expected to buffer the 18O ratios of modern teeth. The results demonstrate that modern teeth show less variation than prehistoric teeth, though th e difference was small suggesting prehistoric teeth undergo more enviro nmental buffering than expected. Still the results of this research are promising and indicate that environmental signals recorded in human tooth enamel are not swamped by secondary mine ralization and can be extracted by sampling the visible growth layers on the surface of the tooth. Finally in Chapter 5 Piercing the Seasonal R ound: Using Stable Isotopes to Reconstruct Human Diet by Season at Windover and Harris Creek, the isotopic methods developed in Chapter 4 are applied to the archaeological populations from Windover and Harris Creek. Ten human molars from Windover and 15 molars from Harris Creek are serially sampled in this research. Previous paleodietary studies in Chap ter 2 have indicated ther e is a marine component in the diet of the populations interred at Windover and Harris Creek. This research uses serially sampled 13C and 18O ratios from human tooth enamel to assess the season of consumption of marine foods. This is the first time stable isotop es have been used to re construct human diet with a sub-annular resolution.
145 The isotopic data derived from the individua ls interred at Windover and Harris Creek are used to evaluate the settlement models proposed for the region. The first model posits that Archaic groups had base camps along the St John s River during the warm months and dispersed to the coast and/or interior hi ghlands during the cooler winter months (Milanich & Fairbanks, 1980). Based on data from flora and fauna recove red from sites along the Atlantic coast and St Johns River, Mike Russo, in collaboration with other researchers, deve loped a new settlement model for the region. Russo et al. (1992) suggest that separate, ethnical ly distinct groups occupied the coast and the St J ohns basin year-round. According to Russo et al. (1992) these groups may have moved along the river or coast, but not between them. Serially sampled isotopic data collected from Windover and Harris Creek are used to evaluate these models. The time series data collected from Windover and Harris Creek support the first model and support the bone isotope data presented in Chapte r 2 which indicates a marine component in the diet. Though different amounts of time incorporat ed in the samples prevent an exact estimation of 13C ratios at any given point in the y ear, a comparison of the shape of the 13C series and the 18O series from each individual suggests that more marine foods were more often consumed during the summer and fall. These estimates mesh well with the floral and faunal data from the region. Combined the isotopic and archaeological evidence sugge st a summer/fall occupation of the coast that is opposite to the original Minanch/Fairbanks model. This trend appears the same for both Windover and Harris Creek, again supporting the results of the bone data in Chapter 2 which suggest no significant difference in the type s of foods consumed. Additionally, these data suggest there was no detectable change in the resource schedu ling between Windover and Harris Creek, indicating the basic scheduling of the seasonal round in place during the Early Archaic was still in place over 2000 rcybp later.
146 These studies provide new data to assess cult ure historical assessments of the Archaic Period in the St Johns region of Florida. Da ta presented here suggest that many of the adaptations or innovations thought to have developed in the Middle Archaic or later had roots in the Early Archaic. Contrary to popular wis dom, based on isotopic data from Harris Creek and Windover, there was no major change in subsiste nce between the Early and Middle Archaic. Additionally, both populations explo ited marine resources though not yet at the levels achieved by later Orange Period and St Johns groups. The people interred in the massive shellworks at Harris Creek were not subsisting primarily on freshwater shellfish, which supports the assertion that much of the freshwater shell was a building material (Ra ndall & Sassaman, 2005b). The dietary data from the bone isotope and serial sampling studies of tooth enamel suggest that the people interred at Harris Creek were seasonally mobile. Though the St. Johns River and the surrounding ecosystem could likely support a population of relatively sedentary complex hunter-gathers, the society that cr eated Harris Creek did not simply substitute aquatic foods in place of agriculture. For these people, sedentism was not a prerequisite for the construction of massive monumental mound complexes or the forma tion of long-distance so cial networks which moved people as well as goods. Finally, based on the population at Harri s Creek, Middle Archaic mounds had large number of immigrants interred in them making it unlik ely that the mounds were the property of any one group. The presence of individuals from across the region bur ied at Harris Creek significantly weakens interpretations of the mound as a territorial marker for a local group or as a means of legitimizing a single groups rights to cr itical but restricted resources (Goldstein, 1976 ; Saxe, 1970 ). The mound may have functioned as a seasonal aggregation site which served economic, political, social, and ideological func tions for very widely distributed groups of
147 seasonally mobile hunter-gatherers (Hofman 1985). Ultimately, even this interpretation may fall short of explaining and contextu alizing the complex role of mortuary mounds on the social landscape. More research is needed to esta blish number of immigr ants and total distance encompassed by the Archaic social networks. Once the scale and scope of these interactions are known, models and theories can be developed to explain the role of Archaic Period shell mounds and cemeteries.
148 LIST OF REFERENCES Ambrose S. H. 1990. Preparation and characteri zation of bone and toot h collagen for isotopic analysis. Journal of Archaeo logical Science 17:431451. Ambrose S. H., Buikstra J. F. & Krueger H. W. 2003. Status and gender di fferences in diet at Mound 72, Cahokia, revealed by isotopic analys is of bone. Journal of Anthropological Archaeology 22:217-226. Ambrose S. H. & DeNiro M. J. 1986. Reconstructi on of African Human Diet using Bone Collage Carbon and Nitrogen Isotope Ratios. Nature 319:321-324. Ambrose S. H. & Norr L. 1993. Experimental evid ence for the relationship of the carbon isotope ratios of whole diet and dietary protein to those of bone co llagen and carbonate. In: (J. B. Lambert and G. Grupe, Eds) Prehistoric Human Bone: Archaeology at the Molecular Level. Springer-Verlag, Berlin pp. 1-37. Ames K. M. 1991. Sedentism: a temporal shift or a transitional change in hunter-gather mobility patterns. In: (S. A. Gregg, Ed) Between ba nds and states. Center for Archaeological Investigations, Carbondale. Amick D. S. 1996. Regional patterns of Folsom mobility and land use in the American Southwest. World archaeology 27:411-426. Aoba T. 1996. Recent observations on enamel crystal formation during mammalian amelogenesis. Anatomical Record 245:208-218. Aoba T. & Moreno E. 1990. Changes in the nature and composition of enamel mineral during porcine amelogenesis. Calcified Tissue Internat ional 47:356-364. Aoba T. & Moreno E. C. 1987. The Enamel Flui d in the Early Secretory Stage of Porcine Amelogenesis Chemical-Composition and Satura tion with Respect to Enamel Mineral. Calcified Tissue International 41:86-94. Arnold B. & Wicker N. L., (Eds) 2001 Gender a nd the Archaeology of D eath. AltaMira Press, Walnut Creek. Aten L. E. 1999. Middle Archaic Ce remonialism at Tick Island, Fl orida: Ripley P. Bullen's 1961 Excavation at the Harris Creek Site. The Florida Anthr opologist 52:132-200. Avery J. K. 2000. Essentials of oral histology an d embryology: a clinical approach. Baltimore, Mosby. Bailey G. & Milner N. 2002. Coastal hunter-gatherers and social evolution: marginal or central? Before Farming: The Archaeology an d Anthropology of HunterGatherers. ( http://www.waspjournals.com/ ) 2002/3-4(1). Bailey G. & Parkington J ., (Eds) 1988. The arch aeology of prehistoric coastlines. Cambridge University Press, Cambridge. Balasse M. 2003. Potential biases in sampling de sign and interpretation of intra-tooth isotope analysis. International Journa l of Osteoarchaeology 13:3-10. Balasse M., Ambrose S. H., Smith A. B. & Pr ice D. T. 2002. The Seasonal Mobility Model for Prehistoric Herders in the Southwestern Ca pe of South Africa assessed by Isotopic Analysis of Sheep Tooth Enamel. Jour nal of Archaeological Science 29:917-932. Balasse M., Tresset A. & Ambrose S. H. 2006. Stab le isotope evidence (delta C-13, delta O-18) for winter feeding on seaweed by Neolithic sheep of Scotland. Journal of Zoology 270:170-176. Balasse M., Tresset A., Dobney K. & Ambrose S. H. 2005. The use of isotope ratios to test for seaweed eating in sheep. Journal of Zoology 266:283-291.
149 Bamforth D. B. 1988. Ecology and human organization on the Great Plains. New York, Plenum Press. Barker G. 2002. Prehistoric foragers and farmers in south-east Asia: rene wed investigations at Niah Cave, Sarawak. Proceedings of the Prehistoric Society 68:147-164. Barrett J. C. 1990. The monumentality of death: the character of early Bronze Age mortuary mounds in southern England. World Archaeology:179-89. Bellwood P. 1997. Prehistory of the Indo-Malays ian Archipelago. Honolulu, University of Hawai'i Press. Bentley R. A. & Knipper C. 2005. Geographical pa tterns in biologically available Strontium, Carbon and Oxygen isotope signatures in prehistoric SW Germany. Archaeometry 47:629-644. Benyon A. D., Clayton C. B., Ramirez Rozzi F. V. & Reid D. J. 1998. Radiological aspects of the developing dentition in modern humans and great apes. Journal of human evolution 35:351-370. Bewick V., Cheek L. & Ball J. 2003. Statistics review 7: Correlation and regression. Critical Care London 7:451-459. Binford L. R. 1968. Post-Pleistocence Adaptations. In: (S. R. Binford and L. R. Binford, Eds) New Perspectives in Archeology. Aldi ne Publishing Co., New York, pp. 313-341. Binford L. R. 1971. Mortuary practices: Their st udy and their potential. In: (J. A. Brown, Ed) Approaches to the social dimensions of mortuary practices Society for American Archaeology, Washington, D.C., pp. 6-29. Binford L. R. 1980. Willow Smoke and Dog's Tails : Hunter-Gatherer Settlement Systems and Archaeological Site Formation. American Antiquity 45:4-20. Binford L. R. 1983. Long-term land use patterns: some implications for archaeology. In: (R. C. Dunnel and D. K. Grayson, Eds) Lulu Linear Punctate: Essays in H onor of George Irving Quimby. University of Michigan, Museum of Anthropology, Ann Arbor, pp. 27-54. Binford L. R. 1990. Mobility, Housing, and Environment: A Comparative Study. Journal of Anthropological Research 46:119-152. Binford L. R. 2001. Constructing frames of reference: An analytical method for archaeological theory building using ethnographic and environmental data sets. Berkeley, University of California Press. Bocherens H. & Drucker D. 2003. Trophic level isot opic enrichment of carbon and nitrogen in bone collagen: case studies from recent and an cient terrestrial ecosystems. International Journal of Osteoarchaeology 13:46-53. Bonzani R. M. 1997. Plant Diversity in the Arch aeological Record: A Means Toward Defining HunterGatherer Mobility Strategies. Jour nal of Archaeologi cal Science 24:1129-1139. Bowen G. 2005. Online Isotopes in Precipitation Calculator (OIPC). ( http://www.waterisotopes.org/) Bowen G. J. & Revenau gh J. 2003. Interpolating the isotopic composition of modern meteoric precipitation. Water Res ources Research 39:1299. Bowen G. J., Wassenaar L. I. & Hobson K. A. 2005. Global application of stable hydrogen and oxygen isotopes to wildlife fore nsics. Oecologia 143:337-348. Bowen G. J. & Wilkinson B. 2002. Spatial distribution of 18O in meteoric precipitation. Geology 30:315-318.
150 Boyde A. 1979. Carbonate concentration, crysta l centres, core dissolu tion, caries, cross striations, circadian rhythms, and comositiona l contrast in the SEM. Journal of Dental Research 58b:981-983. Brantingham P. J. 1998. Mobilit y, competition, and Plio-Pleistocence hominid foraging groups. Journal of Archaeological Method and Theory 5:57-98. Brown J. 1995 On Mortuary Analysis; with Spec ial Reference to the Saxe-Binford Research Program. In: (L. A. Beck, Ed) Regional Appro aches to Mortuary Analysis. Plenum Press, New York, pp. 3-26. Bryant J. D. & Froelich P. N. 1996. Oxygen isotope composition of human tooth enamel from medieval Greenland: Linking climate and society: Comment. Geology 24:477-478. Buikstra J. F. & Charles D. K. 1999. Centeri ng the Ancestors: Cemeteries, Mounds, and Sacred Landscapes of the Ancient North American Midcontinent. In: (W. Ashmore and B. Knapp, Eds) Archaeologies of Landscape: Contemporary Perspectives. Blackwell, Oxford, pp. 201-228. Buikstra J. F. & Milner G. R. 1991. Isotopic and Archaeological Interpretations of Diet in the Central Mississippi Valley. Journal of Archaeological Science 18:319-329. Burch E. S. 1998. The future of hunter-gatherer research. In: (J. Gowdy, Ed) Limited Wants, Unlimited Means: A Reader in Hunter-Gathe rer Economics and the Environment. Island Press, Washington, D.C., pp. 201-217. Cannon A. 2005 Gender and Agency in Mortuary Fash ion. In: (G. F. M. Rakita, J. E. Buikstra, L. A. Beck and S. R. Williams, Eds) Interacting with the Dead: Perspectives on Mortuary Archaeology for the New Millennium. University Press of Florida, Gainesville, pp. 4165. Carlson K. J., Grine F. E. & Pearson O. M. 2007. Robusticity and sexual dimorphism in the postcranium of modern hunter-g atherers from Australia. Amer ican Journal of Physical Anthropology 134:9. Carr C. 1995. Mortuary practices: Their social, philosophical-religious, circumstantial, and physical determinants. Journal of Archaeological Method and Theory 2:105-200. Cashdan E. 1983. Territoriality Among Human Foragers: Ecological Models and an Application to Four Bushman Groups. Current Anthropology 24:47. Cashdan E. 1992. The spatial organization of habitat use. In: (B. Winterhalder and E. A. Smith, Eds) Evolutionary ecology and human beha vior. Aldine de Gr uyter, New York, pp. 237266. Chapman R. 1995 Ten years after-Megaliths, mortua ry practices, and the te rritorial model. In: (L. A. Beck, Ed) Regional Approaches to Mo rtuary Analysis. Plenum Press, New York, pp. 29-51. Chapman R. 2003. Archaeologies of Complexity. London, Routledge. Chatters J. C. 1987. Hunter-gatherer adaptatio ns and assemblage structure. Journal of Anthropological Archaeology 6:336-375. Christenson A. L. 1980. Change in the human ni che in response to popul ation growth. In: (T. Earle and A. Christenson, Eds) Modeling ch ange in prehistoric subsistence economies. Academic Press, New York, pp. 32-72. Claassen C. P. 1991. New Hypotheses for the Demise of the Shell Mound Archaic. In: (C. H. McNutt, Ed) Proceedings of the 1989 Mid-South Archaeological Conference, Memphis, TN. Archaeological Report No. 24, Mississippi Department of Archives and History, Jackson, Mississippi, pp. 66-72.
151 Claassen C. P. 1996. A Consideratio n of the Social Organization of the Shell Mound Archaic. In: (K. E. Sassaman and D. G. Anderson, Eds) Ar chaeology of the Mid-Holocene Southeast. University Press of Florida, Gainesville, pp. 235-258. Close A. E. 2000. Reconstructing Movement in Prehistory. Journal of Archaeological Method and Theory 7:49-77. Cohen M. N. 1977. The food crisis in prehistory : Overpopulation and the or igins of agriculture. New Haven, Yale University Press. Coplen T. B., C. K. & Hopple J. 1983. Comparison of stable isotope reference samples. Nature 302:236-238. Cormie A. B., Luz B. & Schwarcz H. P. 1994. Relationship between the hydrogen and oxygen isotopes of deer bone and their use in the es timation of relative humidity. Geochimica et Cosmochimica Acta 58:3439-3449. Crane H. R. & Griffin J. B. 1965. University of Michigan Radiocar bon Dates X. Radiocarbon 6:1-24. Crown P. L. & Fish S. K. 1996. Gender and stat us in the Hohokam pre-Classic to Classic transition. American An thropologist 98:803-817. Dean C. & Benyon A. D. 1991. Histological reconstr uction of crown forma tion times and initial root formation times in a modern child. American Journal of Physical Anthropology 86:215-228. Dean C., Leakey M. G., Reid D., Schrenk F., Schwartz G. T., Stringer C. & Walker A. 2001. Growth processes in teeth di stinguish modern humans from Homo erectus and earlier hominins. Nature 414:628-631. Deetz J. 1968. Hunters in archaeological perspec tive. In: (R. Lee and I. DeVore, Eds) Man the hunter. Aldine, Chicago, pp. 281-285. DeLeon V. B. 1998. Stable Isotope Analysis and Pa leodiet at the Bay West site, Collier County, Florida. Unpublished M.A. Thesis, Un iversity of Florida, Gainesville. DeNiro M. J. 1985. Postmortem Preservation and Alteration of in vivo Bone Collagen Isotope Ratios in Relation to Paleodietary reconstruction. Nature 317:806-809. DeNiro M. J. & Epstein S. 1978. Carbon Isotopic Evidence for Different Feeding Patterns in Two Hyrax Species Occupying the Sa me Habitat. Science 201:906-907. DeNiro M. J. & Epstein S. 1981. Influence of Di et on the Distribution of Nitrogen Isotopes in Animals. Geochimica et Cosmochimica Acta 45:341-351. Donald L. & Mitchell D. H. 1994. Nature and cult ure on the nortwest coast of North America: The case of Wakashan salmon resources. In: (E. S. Burch and L. J. Ellanna, Eds) Key issues in hunter-gatherer research. Berg, Oxford, pp. 95-117. Doran G. H., (Ed) 2002. Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery. University of Florida Press, Gainesville. Drucker D. G. & Bocherens H. 2004. Carbon and Nitroge n stable isotopes as tracers of change in diet breadth during Middle a nd Upper Palaeolithic in Eur ope. International Journal of Osteoarchaeology 14:162-177. Dupras T. L. & Schwarcz H. P. 2001. Strangers in a Strange Land: Stable Isotope Evidence for Human Migration in the Dakhleh Oasis, Egypt. Journal of Archaeological Science 28:1199-1208. Dupras T. L., Schwarcz H. P. & Fairgrieve S. I. 2001. Infant feeding and weaning practices in Roman Egypt. American Journal of Physical Anthropology 115:204-212.
152 Dutton A., Wilkinson B. H., Welker J. M., Bowen G. J. & Lohmann K. C. 2005 Spatial Distribution and Seasonal Vari ation in 18O/16O of Modern Precipitation and River Water across the Conterminous USA. Hydrological Processes 19:4121-4146. Elliott J. C., Holcomb D. W. & Young R. A. 1985. Infrared Determination of the Degree of Substitution of Hydroxyl by Carbonate Ions in Human Dental Enamel. Calcified Tissue International 37:372-375. Erlandson J. M. 2001. The Archaeology of A quatic Adaptations: Paradigms for a New Millennium. Journal of Arch aeological Research 9:287-350. Evans J. A., Chenery C. A. & Fitzpatrick A. P. 2006. Bronze Age childhood migration of individuals near Stonehenge, revealed by St rontium and Oxygen isotope tooth enamel analysis. Archaeometry 48:309-321. Faure G. 1986. Principles of Isotope Geology. New York, John Wiley and Sons. Fawcett D. W. 1986. A Textbook of Histology, WB Saunders Company. FitzGerald C. M. 1998. Do enamel microstructu res have regular time dependency? Conclusions from the literature and a large-scale st udy. Journal of Human Evolution 35:371-386. Fricke H. C. & O'Neil J. R. 1996 Interand intratooth variation in the oxygen isotope composition of mammalian tooth enamel phosphate: implications for palaeoclimatological and palaeobiol ogical research. Palaeogeography, Palaeoclimatology, Palaeoecology 126:91-100. Fuller B. T., Richards M. P. & Mays S. 2003. Stab le carbon and nitrogen isotope variations in tooth dentine subsections from Wharram Percy. Journal of Archaeological Science 30:1673-1684. Gadbury C., Todd L., Jahren A. H. & Amundson R. 2000. Spatial and temporal variations in the isotopic composition of bison tooth enamel from the Early Holocene Hudson-Meng Bone Bed, Nebraska. Palaeogeography, Palaeoclimatology, Palaeoecology 157:79-93. Gallivan M. D. 2002. Measuring sedentariness and settlement population: Accumulations research in the Middle Atlantic re gion. American Antiquity 67:535-557. Garazhain O. & Yazdi L. P. 2008 Mortuary Practices in Bam afte r the Earthquake: An Ehnoarchaeological Study. Journa l of Social Archaeology 8:94. Gibson J. L. 2006. Navels of the earth: sedentis m in early mound-building cultures in the lower Mississippi valley. Worl d Archaeology 38:311-329. Gillespie S. D. 2001. Personhood, Agency, and Mortuary Ritual: A Case Study from the Ancient Maya. Journal of Anthropological Archaeology 20:73-112. Gluckman M. 1937. Mortuary Customs and the Be lief in Survival after Death among the SouthEastern Bantu. Bantu 11:117-136. Goldstein L. 1976 Spatial Structure and Social Organization: Regiona l Manifestations of Mississippian Society. Unpublis hed Ph.D. dissertation, Northwestern University, Evanston. Goldstein L. 2006. Mortuary Analysis and Bioarch aeology. In: (J. E. Buikstra and L. A. Beck, Eds) Bioarchaeology: The Contextual Analys is of Human Remains. Elsevier, New York, pp. 375-388. Gonfiantini R. 1985. On the isotopic composition of precipitation in tropical stations. Acta Amazonica 15:121-139. Griffin P. B. 1989. Hunting, farmi ng, and sedentism in a rain forest foraging society. In: (S. Kent, Ed) Farmers as Hunters. Cambridge University Press, Cambridge, pp. 60-70.
153 Halstead P. & O'Shea J. 1989. Introduction: cultur al responses to risk and uncertainty. In: (P. Halstead and J. O'Shea, Eds) Bad year economics: Cultural responses to risk and uncertainty. Cambridge Universi ty Press, Cambridge, pp. 1-7. Hrke H. 2002. Interdisciplinarity and the arch aeological study of death. Mortality 7:340-341. Hedges R. E. M. 2004. Isotopes and red herrings: comments on Milner et al. and Lidn et al. Antiquity 78:34-37. Hiller C. R., Robinson C. & Weatherell J. A. 1975. Variations in Composition of Developing Rat Incisor Enamel. Calcified Tissue Research 18:1-12. Hitchcock R. K. 1982. Patterns of sedentism am ong the Basarwa of eastern Botswana In: (E. Leacock and R. Lee, Eds) Politics and hist ory in band society. Cambridge University Press, Cambridge, pp. 223-267. Hodder I. 1984. Burials, houses, women and men in the European Neolithic. In: (D. Miller, and Tilley,C., Ed) Ideology, Power, and Prehistor y. Cambridge University Press, New York, pp. 51-68. Hodder I. 1986. Reading the pas t: Current Approaches to in terpretation in archaeology. Cambridge, Cambridge University Press. Hofman J. L. 1985. Middle Archaic Ritual a nd Shell Midden Archaeology: Considering the Significance of Cremations. In: (T. Whyte, C. Boyd and B. Riggs, Eds) Exploring Tennessee Prehistory: A Dedication to Alfred K. Guthe. Department of Anthropology, University of Tennessee, Knoxville, pp. 1-21. Hoogewerff J., Papesch W., Kra lik M., Berner M., Vroon P., Miesbauer H., Gaber O., Kunzel K.-H. & Kleinjans J. 2001. The last domoc ile of the Iceman from Hauslabjoch: A geochemical approach using Sr, C, and O Isotopes and trace element signatures. Journal of Archaeological Science 28:983-989. Hoppe K. A., Stover S. M., Pascoe J. R. & Amundson R. 2004. Tooth enamel biomineralization in extant horses: implicaitons for isotopic microsampling. Palaeogeography Palaeoclimatology Palaeoecology 206:355-365. Horsley P. A. 2008 Death dwells in spaces: Bodies in the hospital mortuary. Anthropology & Medicine 15:133-146. Howell T. L. 1995. Tracking Zuni Gender and Leadership Roles across the Contact Period. Journal of Anthropologic al Research 51:125-147. Humphrey L. T., Dean M. C. & Jeffries T. E. 2007. An evaluation of changes in strontium/calcium ratios across the neonatal line in human deciduous teeth. In: (S. Bailey and J.-J. Hublin, Eds) Dental Perspectives on Human Evolution: State of the Art Research in Dental Paleoanthropology pp. 303-319. Hutchinson D. L., (Ed) 2002. Foraging, Farming, and Coastal Biocultural Adaptation in Late Prehistoric North Carolina. University of Florida Press, Gainesville. Hutchinson D. L., (Ed) 2004. Bioarchaeology of the Florida Gulf Coas t: Adaptation, Conflict, and Change. University Press of Florida, Gainesville. Hutchinson D. L., Larsen C. S., Norr L. & Sc hoeninger M. J. 2000. Agricultural Melodies and Alternative Harmonies in Florid a and Georgia. In: (P. M. La mbert, Ed) Bioarchaeological Studies of Life in the Age of Agriculture: A View from the Southeast. The University of Alabama Press, Tuscaloosa, pp. 96-115. Iacumin P., Bocherens H., Mariotti A. & Longi nelli A. 1996. Oxygen isotope analyses of coexisting carbonate and phosphate in biogeni c apatite: a way to monitor diagenetic alteration of bone phosphate? Earth and Planetary Science Letters:1-6.
154 Jahn O. L. & Bullen R. P. 1978. The Tick Island S ite, St. Johns River, Florida. Gainesville, Florida Anthropological Society Publications. Jochim M. A. 1976. Hunter-gathere r Subsistence and Settlement: A Predictive Model, Academic Press. Jones D. S., Quitmyer I. R. & Andrus F. T. 2005. Oxygen isotopic evidence for greater seasonality in Holocene shells of Donax variabilis from Florida. Palaeogeography, Palaeoclimatology, Palaeoecology 228:96-108. Katzenberg M. A. 1989. Stable is otope analysis of archaeological faunal remains from southern Ontario. Journal of Archaeo logical Science 16:319. Katzenberg M. A. & Saunders S. R. 2000. Stable Isotope Analysis: A Tool for Studying Past Diet, Demography and Life History. In: (M. A. Katzenberg and S. R. Saunders, Eds) Biological Anthropology of the Human Sk eleton. Wiley-Liss, New York, pp. 305-327. Katzenberg M. A., Schwarcz H. P., Knyf M. & Melbye F. J. 1995. Stable-Isotope Evidence for Maize Horticulture and Paleodiet in Sout hern Ontario, Canada. American Antiquity 60:335-350. Keenleyside A., Schwarcz H. & Panayotova K. 2006. Stable isotopic evidence of diet in a Greek colonial population from the Black Sea. Journal of Archaeological Science 33:12051215. Kellner C. M. & Schoeninger M. J. 2007. A simp le carbon isotope model for reconstructing prehistoric human diet. American Jo urnal of Physical Anthropology 133:1112. Kelly J. A., Tycot R. H. & Milanich J. T. 2006. Evidence for Early Use of Maize in Peninsular Florida. In: (J. E. Staller, R. H. Tykot and B. F. Benz, Eds) Histories of Maize. Academic Press, New York, pp. 249-262. Kelly J. F. 2000. Stable isotopes of carbon and nitrogen in the study of avian and mammalian trophic ecology. Canadian Journal of Zoology 78:1-27. Kelly R. L. 1983. Hunter-gatherer mobility stra tegies. Journal of Anthropological Research 39:277-306. Kelly R. L. 1990. Marshes and mobility in the Wester n Great Basin. In: (J. C. Janetski and D. B. Madsen, Eds) Wetlands adaptations in the Great Basin. Museum of People and Cultures Occasional Papers. Brigham Young University, Provo, pp. 259-276. Kelly R. L. 1992. Mobility/Seden tism: Concepts, Archaeological Measures, and Effects. Annual Review of Anthropology 21:43-66. Kelly R. L. 1995. The Foraging Spectrum: Diversity in Hunter-Gatherer Lifeways. Washington, Smithsonian Institution Press. Knudson K. J. & Price T. D. 2007. Utility of mu ltiple chemical techniques in archaeological residential mobility studies: Case studies from Tiwanakuand Chiribaya-affiliated sites in the Andes. American Journal of Physical Anthropology 132:25-39. Knudson K. J., Tung T. A., Nystrom K. C., Price T. D. & Fullagar P. D. 2005. The origin of the Juch'uypampa Cave mummies: strontium isot ope analysis of archaeological human remains from Bolivia. Journal of Archaeological Science 32:903-913. Koch P. L., Fogel M. J. & Tuross N. 1994. Traci ng the diets of fossil animals using stable isotopes. In: (K. Lajtha and R. H. Michener Eds) Stable Isotopes in Ecology and Environmental Science. Blackwe ll Scientific, Oxford, pp. 63-92. Koch P. L., Halliday A. N., Walter L. M., Stear ley R. F., Huston T. J. & Smith G. R. 1992. Sr Isotopic Composition of Hydroxyapatite from Recent and Fossil Salmon the Record of Lifetime Migration and Diagenesis. Earth and Planetary Science Letters 108:277-287.
155 Koch P. L., Heisinger J., Moss C., Carlson R. W., Fogel M. L. & Behrensmeyer A. K. 1995. Isotopic Tracking of Change in Diet and Ha bitat Use in African Elephants. Science 267:1340-1343. Koch P. L., Tuross N. & Fogel M. L. 1997. The e ffects of sample treatm ent and diagenesis on the isotopic integrity of carbonate in biogenic hydroxylapatite. Journal of Archaeological Science 24:417-429. Kohn M. J. & Cerling T. E. 2002. Stable isotope compositions of biological apatite. Phosphates: Geochemical, Geobiological, and Materials Importance 48:455-488. Kohn M. J., Schoeninger M. J. & Valley J. W. 1998. Variability in oxygen isotope compositions of herbivore teeth: reflections of seasonality or developmental physiology? Chemical Geology 152:97-112. Komar D. 2008 Patterns of Mortuary Practice Associated with Genocide. Current Anthropology 49:123-133. Krigbaum J. 2005. Reconstructing human subsistence in the West Mouth (Niah Cave, Sarawak) burial series using stable isotopes of carbon. Asian Perspectives 44:73-89. Kroeber A. L. 1927. Disposal of the D ead. American Anthropologist 29:308-315. Larsen C. S. 1997. Bioarchaeology: Interpreting behavior from the human skeleton. Cambrige, Cambridge University Press. Larsen C. S. & Ruff C. B. 1991. Biomechanical Adaptation and Behavior on the Prehistoric Georgia Coast. What Mean These Bones? St udies in Southeastern Bioarchaeology:102113. Larsen C. S. & Ruff C. B. 1994. The stresses of conquest in Spanish Florida: structural adaptation and change before and after c ontact In the Wake of Contact: Biological Responses to Conquest. Wiley-Liss, New York, pp. 21. Larsen C. S., Ruff C. B. & Kelly R. L. 1991. Sk eletal structural adaptations in prehistoric Western Great Basin hunter-gatherers American Association of Physical Anthropologists Wisconson. Leakey M. D. & Hay R. L. 1979. Pliocene footprints in the Laetolil Beds at Laetoli, northern Tanzania. Nature 278:317-323. Lee-Thorp J. A. 1989. Stable carbon isotopes in deep time: the diet s of fossil fauna and hominids. Unpublished Ph.D. Dissertation, Un iversity of Cape Town, Cape Town. Lee R. & De Vore I. 1968. Ma n the Hunter. Chicago, Aldine. Lewin R. 1988. New views emerge on Hunters and Gatherers. Science 240:1146-1148. Lidn K., Eriksson G., Nordqvist B., Gtherstr m A. & Bendixen E. 2004. The wet and the wild followed by the dry and the tame" or did they occur at the same time? Diet in Mesolithic -Neolithic southe rn Sweden. Antiquity 78:23-33. Lightfoot K. G. & Jewett R. 1986. The shift to sedentary life: A consider ation of the occupation duration of Early Mogollon p ithouse villages. In: (C. Benson and S. Upham, Eds) Mogollon Variability. New Mexico Stat e University, La Cruces, pp. 9-44. Longinelli A. 1984. Oxygen isotopes in mammal bone phosphate: A new tool for paleohydrological and paleoclim atological research? Geochi mica et Cosmochimica Acta 48:385-390 Lull V. 2000 Death and Society: a Ma rxist approach. Antiquity 74:576-580. Luz B., Cormie A. B. & Schwarcz H. P. 1990 Oxyge n Isotope Variations in Phosphate of Deer Bones. Geochimica et Cosmochimica Acta 54:1723-1728.
156 Luz B. & Kolodny Y. 1985. Oxygen isotope variati ons in phosphate of biogenic apatites, IV, mammal teeth and bones. Earth and Planetary Science Letters 75:29-36. Luz B., Kolodny Y. & Kovach J. 1984. Oxygen Isot ope Variations in Phosphate of Biogenic Apatites, 3 Conodonts. Earth and Pl anetary Science Letters 69:255-262. MacFadden B. J., Higgins P., Clementz M. T. & Jones D. S. 2004. Diets, habitat preferences, and niche differentiation of Cenozoic sirenians from Florida: evidence from stable isotopes. Paleobiology 30:297-324. McGee R. M. 1995. Fired-Clay Objects: Cooking Technology from the Florida Shell Mound Archaic at Groves' Orange Midden (8VO2601) Florida. Unpublished M.A. thesis, University of Florida, Gainesville. Merbs C. F. 1983. Patterns of Activity-Induced Pathology in a Canadian Inuit Population. Ottawa, Archaeological Survey of Canada. Milanich J. T. 1994. Archaeology of Precolumbian Florida. Gainesville, University Press of Florida. Milanich J. T. 1998. Florida's Indians from Ancien t Times to the Present. Gainesville, University Press of Florida. Milanich J. T. & Fairbanks C. H. 1980. Flor ida Archaeology. New York, Academic Press. Milner N., Craig O. E., Bailey G. N., Pedersen K. & Anderson S. H. 2004. Something fishy in the Neolithic? A re-evaluation of stable is otope analysis of Mesolithic and Neolithic coastal populations. Antiquity 78:9-22. Minagawa M. & Wada E. 1984. Stepwise enri chment of 15 N along food chains: Further evidence and the relation between 15 N a nd animal age. Geochimica et Cosmochimica Acta 48:1135-1140. Montgomery J., Evans J. A., Powlesland D. & Robe rts C. A. 2005. Continuity or colonization in Anglo-Saxon England? Isotope evidence for mobility, subsistence practice, and status at West Heslerton American Journal of Physical Anthropology 126:123-138. Moore C. B. 1892a. A Burial Mound of Florida. American Naturalist 26:128-143. Moore C. B. 1892b. Certain Shell Heaps of the St. John's River, Florida, Hitherto Unexplored. American Naturalist 27:8-13, 113-117, 605-624, 708-729. Moore C. B. 1892c. Supplementary Investigations at Tick Island. American Naturalist 26:568579. Moore C. B. 1894. Certain sand Mounds of the St. Johns River, Florida. Journal of the Academy of Natural Sciences of Philadelphia 10. Morris I. 1991 The archaeology of ancestors: The Saxe/Goldstein hypothesis revisited. Cambridge Archaeological Journal 1:147-169. Moseley M. E. 1975. The maritime foundations of Andean civilization. Menlo Park, California, Cummings Publishing Co. Moseley M. E. 1992. The Incas and Their Ances tors: The Archaeology of Peru, Thames and Hudson. Moss-Salentijn L., Moss M. L. & Yuan M. S.-t. 1997. The ontogeny of mammalian enamel. In: (W. V. Koenigswald and P. M. Sander, Eds) Tooth Enamel Microstructure. Balkema, Rotterdam, pp. 5-30. Murdock G. P. 1967. The ethnographic atlas: a summary. Ethnology 6:109-236. Murdock G. P. & White D. R. 1969. Standa rd Cross-Cultural Sample. Ethnology 8:329-69. Nelson S. V. 2005. Paleoseasonality inferred from e quid teeth and intra-toot h isotopic variability. Palaeogeography Palaeoclim atology Palaeoecology 222:122-144.
157 Newsom L. 1994. Archaeobotanical Data from Groves Orange Midden (8VO2601), Volusia County, Florida. The Florida Anthropologist 47:404-417. Newsome S. D., Phillips D. L., Culleton B. J., Guilderson T. P. & Koch P. L. 2004. Dietary reconstruction of an early to middle Ho locene human population from the central California coast: insights from advanced stable isotope mixing models. Journal of Archaeological Science 31:1101-1115. Nielson-Marsh C. M. & Hedges R. E. M. 2000. Patte rns of diagenesis in bone: 1. The effect of site environments. Journal of Archaeological Science 27:1139-1150. Njitchoua R., Sigha-Nkamdjou L., Dever L ., Marlin C., Sighomnou D. & Nia P. 1999. Variations of the stable isotopic compositions of rainfall events from the Cameroon rain forest, Central Africa. Journal of Hydrology 223:17-26. Norr L. 1995. Interpreting dietary maize from bone isotopes in the New World tropics: the state of the art. In: (P. W. Stah l, Ed) Archaeology of the Lowland American Tropics: Current Analytical Methods and App lications. Cambridge University Press, New York, pp. 198 223. Parker-Pearson M. 1993 The powerful dead: Archaeo logical relationships between the living and the dead. Cambridge Archaeo logical Journal 3:203-229. Parker-Pearson M. 1999. The Archaeology of Deat h and Burial. College Station, Texas A&M University Press. Passey B. H. & Cerling T. E. 2002. Tooth enamel mineralization in ungulates: Implications for recovering a primary isotopic time-series. Geochimica Et Cosmochimica Acta 66:32253234. Passey B. H., Cerling T. E., Perkins M. E., Voor hies M. R., Harris J. M. & Tucker S. T. 2002. Environmental change in the Great Plains: An isotopic record from fossil horses. Journal of Geology 110:123-140. Passey B. H., Cerling T. E., Schuster G. T., Robi nson T. F., Roeder B. L. & Krueger S. K. 2005. Inverse methods for estimating primary input signals from time-averaged isotope profiles. Geochimica Et Cosmochimica Acta 69:4101-4116. Pauketat T. R. 1989. Monitoring Mississippian ho mestead occupation span and economy using ceramic refuse. American Antiquity 54:288-310. Perlman S. M. 1983. An optimum diet model, co astal variability, and hunter-gatherer behaviour. In: (M. B. Schiffer, Ed) Advances in archaeo logical method and theory. Academic Press, New York, pp. 257-310. Phillips D. L. & Gregg J. W. 2003. Source partiti oning using stable isotopes: coping with too many sources. Oecologia 136:261-269. Piatek B. J. 1994. The Tomoka Mound Complex in Northeast Florida. So utheastern Archaeology 12:109-118. Plog S. 1990. Agriculture, sedentism, and environm ent in the evolution of political systems. In: (S. Upham, Ed) The evolution of political systems. Cambridge University Press, Cambridge, pp. 177-199. Price D. T., Burton J. H. & Bentley R. A. 2002. The characterisation of biologically-available strontium isotope ratios for i nvestigation of prehistoric migration. Archaeometry 44:117135. Price D. T., Manzanilla L. & Middleton W. D. 2000. Immigration and the Ancient City of Teotihuacan in Mexico: a Study Using Str ontium Isotope Ratios in Human Bone and Teeth Journal of Archaeological Science 27:903-913.
158 Price T. D. & Brown J. A., (Eds) 1985. Prehistori c Hunter-Gatherers: The Emergence of Cultural Complexity. Academic Press, New York. Pritchard G. T. & Robbins C. T. 1990. Digestive a nd metabolic efficiencies of grizzly and black bears. Canadian Journal of Zoology 68:1645-1651. Prowse T., Schwarcz H., Macchiarelli R. & Bondi oli L. 2003. Isotopic evidence of migration at the imperial port of Portus Romae, Italy. American Journal of Physical Anthropology:172-172. Prowse T., Schwarcz H. P., Saunders S., Macchia relli R. & Bondioli L. 2004. Isotopic paleodiet studies of skeletons from the imperial Roman-age cemetery of Isola Sacra, Rome, Italy. Journal of Archaeologi cal Science 31:259-272. Quinn R. 1999. Evidence for Diet and Climate in Archaic Florida: Analyses of the Tick Island Human and Faunal Skeletal Samples. Unpublishe d M.A. Thesis, University of Florida, Gainesville. Quinn R. L., Tucker B. D. & Krigbaum J. 2008. Di et and mobility in Middle Archaic Florida: stable isotopic and faunal evidence from th e Harris Creek archaeological site (8Vo24), Tick Island. Journal of Arch aeological Scie nce 35:2346-2356. Quitmyer I. R., Jones D. S. & Arnold W. S. 1997. The Sclerochronology of Hard Clams, Mercenariaspp., from the South-Easter n USA: A Method of Elucidating the Zooarchaeological Records of Seasonal Re source Procurement and Seasonality in Prehistoric Shell Middens. Journal of Archaeological Science 24:825-840. Rafferty J. E. 1985. The archaeological record on sedentariness: Recognition, development, and implication. In: (M. B. Schiffer, Ed) Adva nces in Archaeological Method and Theory. Academic Press, New York. Rakita G. F. M. & Buikstra J. F. 2005. Theories, Time, and Space. In: (G. F. M. Rakita, J. F. Buikstra, L. A. Beck and S. R. Williams, Eds) Interacting with the Dead: Perspectives on Mortuary Archaeology for the New Millennium. University Press of Fl orida, Gainesville, pp. 13-15. Randall A. & Sassaman K. E. 2005a. (Re)Placin g Archaic History. Paper invited to the symposium "Theorizing Place in Archaeology: Pr ospects and Potentialities," 70th Annual Meeting of the Society for American Archaeology Salt Lake City, UT. Randall A. & Sassaman K. E. 2005b. St. Johns Archaeological Field School 2003-2004: Hontoon Island State Park. Laboratory of Sout heastern Archaeology, Technical Report 6. Department of Anthropology, University of Florida Gainesville. Reid D. J. & Dean M. C. 2006. Variation in modern human enamel formation times. Journal of Human Evolution 50:329-346. Richards M., Harvati K., Grimes V., Smith C ., Smith T., Hublin J. J., Karkanas P. & Panagopoulou E. 2008. Strontium isotope evidence of Neanderthal mobility at the site of Lakonis, Greece using laser-ablation PIMMS Journal of Archaeological Science 35:1251-1256. Richards M. P., Fuller B. T. & Hedges R. E. M. 2001. Sulphur isotopic va riation in ancient bone collagen from Europe: implications for hu man palaeodiet, residence mobility, and modern pollutant studies. Earth and Planetary Science Letters 191:185-190. Richards M. P. & Hedges R. E. M. 1999. Stable is otope evidence for similarities in the types of marine foods used by Late Mesolithic humans at sites along the Atlantic coast of Europe. Journal of Archaeologi cal Science 26:717-722.
159 Richards M. P., Jacobi R., Cook J., Pettitt P. B. & Stringer C. B. 2005. Isotope evidence for the intensive use of marine foods by Late Upper Palaeolithic humans. Journal of Human Evolution 49:390-394. Richards M. P., Mays S. & Fuller B. T. 2002. Stab le carbon and nitrogen isotope values of bone and teeth reflect weaning age at the Medi eval Wharram Percy site, Yorkshire, UK. American Journal of Physical Anthropology 119:205-210. Richards M. P., Schulting R. J. & Hedges R. E. M. 2003. Archaeology: Sharp shift in diet at onset of the Neolithic. Nature 425:366 Rick T. C., Erlandson J. M. & Vellanoweth R. L. 2001. Paleocoastal marine fishing on the Pacific coast of the Americas: Perspectives from Daisy Cave, California. American antiquity 66:595-613. Rindos D. 1983. The origins of agriculture. New York, Academic Press. Russo M. 1992. Chronologies and Cultures of the St Marys Region of Northeast Florida and Southeast Georgia. The Flor ida Anthropologist 45:107-126. Russo M. 1994. Why We Don't Believe in Arch aic Ceremonial Mounds and Why We Should: The Case from Florida. Sout heastern Archaeology 13:93-108. Russo M. 1996. Southeastern Mid-Holocene Coastal Settlements. In: (K. E. Sassaman and D. G. Anderson, Eds) Archaeology of the Mid-Holocene Southeast. University Press of Florida, Gainesville, pp. 177-199. Russo M. 1998 Measuring Sedentism with Fauna: Archaic Cultures Along the Southwest Florida Coast. In: (T. R. Rocek and O. Bar-Yos ef, Eds) Seasonality and Sedentism: Archaeological Perspectives from Old and New World Sites. Peabody Museum Bulletin Cambridge. Russo M., Purdy B., Newsom L. A. & McGee R. M. 1992. A Reinterpretation of Late Archaic Adaptations in Central-East Florida: Grove's Orange Midden (8vo2601). Southeastern Archaeology 11:95-108. Russo M. & Ste. Claire D. 1992. Tomoka Stone: Archaic Period Coastal Settlement in East Florida. The Florida Anthropologist 45:336-346. Sakae T. & Hirai G. 1982. Calcific ation and crystalliza tion in Bovine Enamel. Journal of Dental Research 61:57-59. Sassaman K. E. 2004. Complex Hunter-Gatherers in Evolution and History: A North American Perspective. Journal of Arch aeological Research 12:227-280. Saunders R. 1994. The Case for Archaic Period Mounds in Southeastern Louisiana. Southeastern Archaeology 13:118-133. Saxe A. A. 1970 Social dimensions of mortuary practices. Ph.D. dissert ation, University of Michigan, Ann Arbor. Schoeninger M. J. 1985. Trophic Level Effects on 15N/14N and 13C/12C Ratios in Bone Collagen and Strontium Levels in Bone Mineral. Journal of Hu man Evolution 14:515525. Schoeninger M. J. 1989. Reconstructing Prehistori c Human Diet. In: (T. D. Price, Ed) The Chemistry of Prehistoric Bone. Cambridge University Press, Cambridge, pp. 38-67. Schoeninger M. J. & DeNiro M. J. 1984. Nitrogen and Carbon Isotopic Composition of BoneCollagen from Marine and Terrestrial Anim als. Geochimica Et Cosmochimica Acta 48:625-639.
160 Schoeninger M. J., Deniro M. J. & Tauber H. 1983. Stable Nitrogen Isotope Ratios of BoneCollagen Reflect Marine and Terrestrial Com ponents of Prehistoric Human Diet. Science 220:1381-1383. Schoeninger M. J. & Moore K. 1992. Bone Stable Isotope Studies in Archaeology. Journal of World Prehistory 6:247-296. Schoeninger M. J. & Schurr M. R. 1998. Human Subsistence at Moundville: The Stable Isotope Data. In: (V. J. Knight and V. P. Ste ponaitis, Eds) Archaeology of the Moundville Chiefdom. Smithsonian Institution Press, Washington, D.C., pp. 120-132. Schrire C. 1980. An inquiry into the evolutionary status and apparent idenitity of San huntergatherers. Human Ecology 8:9-32. Schrire C., (Ed) 1984. Past and present in hunte r-gatherer studies. Academic Press, Orlando. Schroeder H. E. 1991. Oral Structural Biology. New York, Thieme Medical Publishers, Inc. Schroeder S. 2001. Secondary Disposal of the Dead: Cross-Cultural Codes. World Cultures 12:77-93. Schwarcz H. P., Buchner S. & Walker P. L. 2005. Isotopic evidence of consumption of marine foods by ancestral Chumash. American Journal of Physical Anthropology:190. Schwarcz H. P. & Schoeninger M. J. 1991. Stable Isotope Analyses in Human Nutritional Ecology. Yearbook of Physical Anthropology 34:283-321. Sealy J. 2006. Diet, mobility, and settlement pattern among Holocence hunter-gatherers in southernmost Africa. Current Anthropology 47:569-595. Semple S. & Williams H. 2007 Early Medieval Mortuary Practices. Oxford, University of Oxford. Sharp Z. D. & Cerling T. E. 1998. Fossil isot ope records of seasona l climate and ecology: Straight from the horse's mouth. Geology 26:219-222. Shearer G. & Kohl D. H. 1986. N2 -fixation in field settings: Es timations based on natural 15N abundance. Australian Journal of Plant Phys iology 13:699-757. Shellis R. P. 1998. Utilization of periodic marki ngs in enamel to obtain information on tooth growth. Journal of Hu man Evolution 35:387-400. Simms S. R., Bright J. R. & Ugan A. 1997. Plain-Ware Ceramics and Residential Mobility: A Case Study From the Great Basin. Journa l of Archaeological Science 24:779-792. Smith B. H. 1991. Standards of human tooth form ation and dental age assesment. In: (M. A. Kelly and C. S. Larsen, Eds) Advances in dental anthropology. Wiley-Liss, New York, pp. 143-168. Smith B. N. 1972. Natural Abundance of the Stab le Isotopes of Carbon in Biological Systems BioScience 22:226-231. Smith B. N. & Epstein S. 1971. Two Categories of 13C/12C Ratios for Higher Plants. Plant Physiology 47:380-384. Smith C. S. 2003. Hunter-gatherer mobility, storag e, and houses in a marginal environment: an example from the mid-Holocence of Wyomi ng. Journal of Anthropological Archaeology 22:162-189. Smith C. S. & McNees L. M. 1999. Facilities an d hunter-gatherer long-term land use patterns: an example from southwest Wyomi ng. American Antiquity 64:117-136. Sponheimer M., Passey B. H., de Ruiter D. J., Guat elli-Steinberg D., Cer ling T. E. & Lee-Thorp J. A. 2006. Isotopic evidence for dietar y variability in the early hominin Paranthropus robustus Science 314:980-982.
161 Ste. Claire D. 1990. The Archaic in East Florida: Archaeological Evidence from Early Coastal Adaptations. The Florida Anthropologist 43:189-197. Steward J. H. 1938. Basin-Pl ateau aboriginal sociopoliti cal groups. Washington, D.C., Smithsonian Institution Press. Stiles D. 1992. The Hunter-Gatherer 'Revis ionist Debate'. An thropology Today 8:13-17. Straight W. H., Barrick R. E. & Eberth D. A. 2004. Reflections of surface water, seasonality and climate in stable oxygen isotopes from ty rannosaurid tooth enamel. Palaeogeography Palaeoclimatology Palaeoecology 206:239-256. Stuiver M. & Reimer P. J. 1993. Extended 14C data base and revised CALIB 3.0 14C age calibration program. Radiocarbon:215-230. Suga S. 1982. Progressive mineraliz ation pattern of developing en amel during the mineralization stage. Journal of Dent al Research 61:1532-1542. Suga S. 1989. Enamel hypomineralization viewed from the pattern of progressive mineralization of human and monkey devel oping enamel. Advances in Dental Research 3:188-198. Suga S., Aoki H., Yamashita Y., Tsuno M. & Ogawa M. 1987. A comparative study of disturbed mineralization of rat incisor enamel induced by strontium and flouride administration. Advances in Dental Research 1:339-355. Sullivan L. P. 2001 Those Men in the Mounds: Gender, Politics, and Mortuary Practices in Late Prehistoric Eastern Tennessee. In: (J. M. E. a. C. B. Rodning, Ed) Archaeological Studies of Gender in the Southeastern United States. University Press of Florida, Gainesville, pp. 101-126. Sydney-Zax M., Mayer I. & Deutsch D. 1991. Ca rbonate content in developing human and bovine enamel. Journal of Dental Research 70:913-916. Tainter J. A. 1973 Structure and organization of Middle Woodland societies in the lower Illinois River Valley. Masters thesis, Northwestern University, Tainter J. A. 1975. Social Inference and Mortua ry Practices: An Experiment in Numerical Classification. World Archaeology 7:1-15. Tainter J. A. 1978. Mortuary practices and the study of prehistoric social systems. Advances in Archaeological Method and Theory 1:105-141. Tieszen L. L. & Fagre T. 1993. Effect of diet quality and composition on the isotopic composition of respiratory CO 2, bone collag en, bioapatite, and so ft tissues. In: (J. Lambert and G. Grupe, Eds) Prehistoric human bone: archaeology at the molecular level. Springer-Verlag, Berlin, pp. 121. Tucker B. D. 2007. The Ocmulgee/Blackshear peopl e and the middleman hypot hesis: an isotopic evaluation. Southeastern Archaeology 26:124-133. Tucker B. D. & Krigbaum J. 2005. Was there Social Ranking at Harris Cr eek? Using light stable isotopes to address status, residence, and sedentism during the Middle Archaic 62nd Southeastern Archaeology Conference (SEAC) Columbia, South Carolina. Turner B. L., Kingston J. D. & Milanich J. T. 2005. Isotopic Evidence of Immigration Linked to Status during the Weeden Island and Suwa nnee Valley Periods in North Florida. Southeastern Archaeology 24:121-136. Tuross N., Fogel M. L., Newsom L. & Doran G. H. 1994. Subsistence in the Florida Archaic: The Stable-Isotope and Archaeobotanical Evidence from the Windover Site. American Antiquity 59:288-303. van der Merwe N. J. 1982. Carbon isotopes, photos ynthesis and archaeology. American Scientist 70:596-606.
162 Vanderklift M. A. & Ponsard S. 2003. Sources of variation in consumer-diet d 15 N enrichment: a meta-analysis. Oecologia 136:169-182. Varien M. D. 1999. Sedentism and mobility in a social landscape: Mesa Verde and beyond. Tucson, University of Arizona Press. Vickers W. T. 1989. Patterns of foraging and farm ing in semi-sedentary Amazonian community. In: (S. Kent, Ed) Farmers as Hunters. Cambridge University Press, Cambridge, pp. 46-59. Vogel J. C. & Van de Merwe N. J. 1977. Isotopic Evidence for Early Maize Cultivation in New York State. American Antiquity 42:238-242. Wada E. 1980. Nitrogen isotope fractionation and its significance in biogeochemical processes occurring in marine environments. Isotope Marine Chemistry:375. Wandsnider L. 1992. The spatial dimension of time In: (J. Rossignol and L. Wandsnider, Eds) Space, time, and archaeological landscapes. Plenum Press, New York, pp. 257-282. Wanner I. S., Sosa T. S., Alt K. W. & Blos V. T. 2007. Lifestyle, occupation, and whole bone morphology of the pre-Hispanic Maya co astal population from Xcambo, Yucatan, Mexico. International Journa l of Osteoarchaeology 17:253. Warrick G. A. 1988. Estimating Ontario Iroquoian village duration. Man in the Northeast 36:2161. Weatherell J. A. & Robinson C. 1973. The inorga nic composition of teeth. In: (I. Zipkin, Ed) Biological Mineralization. John Wiley and Sons, Inc., New York, pp. 43-74. Wheeler R. J. & McGee R. M. 1994. Report of Pr eliminary Zooarchaeological Analysis: Groves' Orange Midden. The Florida Anthropologist 47:393-403. Wheeler R. J., Newman C. L. & McGee R. M. 2000. A New Look at the Mount Taylor and Bluffton Sites, Volusia County, with an Outlin e of the Mount Taylor Culture. The Florida Anthropologist 53:132-157. White C. D., Pendergast D. M., Longstaffe F. J. & Law K. R. 2001. Social Complexity and Food Systems at Altun Ha, Belize: The Isotopic Ev idence. Latin American Antiquity 12:371393. White C. D. & Schwarcz H. P. 1989. Ancient Maya Diet as Infe rred from Isotopic and Elemental Analysis of Human-Bone. Jour nal of Archaeological Science 16:451-474. White C. D., Spence M. W., Stuart-Williams H. L. Q. & Schwarcz H. P. 1998. Oxygen Isotopes and the Indentification of Geographical Origins: the Valley of Oxaca versus the Valley of Teotihuacan. Journal of Ar chaeological Science 25:643-655. Widmer R. J. 1988. The Evolution of the Calusa : A Nonagricultural Chiefdom on the Southwest Florida Coast, University of Alabama Press. Wiedemann F. B., Bocherens H., Mariotti A., Driesch A. v. d. & Grupe G. 1999. Methodological and Archaeological Implications of Intra-tooth Isotopic Variations ( 13C, 18O) in Herbivores from Ain Ghazal (Jordan, Neolith ic). Journal of Archaeological Science 26: 697-704. Wilmsen E. 1983. Land Filled with Flies: a Po litical Economy of the Kalahari. Chicago, University of Chicago. Wobst H. M. 1978. The archaeo-ethnology of hunter-gatherers, or The tyranny of the ethnographic record in archaeol ogy. American Antiquity 43:303-309. Woodburn J. 1980. Hunters and gatherers today and reconstruction of the past. In: (E. Gellner, Ed) Soviet and western anthropology. Duckworth, London, pp. 95-117.
163 Wright L. E. & Schwarcz H. P. 1998. Stable ca rbon and oxygen isotopes in human tooth enamel: Identifying breastfeeding and weaning in prehistory. (vol 106, pg 1, 1998). American Journal of Physical Anthropology 106:1-18. Wyman J. 1868. An Account of the Fresh-Water Shell Heaps of the St. Johns River, East Florida. American Naturalist 2:393-403, 440-463. Yellen J. E. 1977. Archaeological approaches to the Present: Models for r econstructing the past. New York, Academic Press. Zazzo A., Balasse M. & Patterson W. P. 2005. High-resolution 13C intratooth profiles in bovine enamel: Implications for minerali zation pattern and isotopic attenuation. Geochimica et Cosmochimica Acta 69:3631-3642. Zazzo A., Balasse M. & Patterson W. P. 2006. The reconstruction of mammal individual history: refining high-resolution isotope record in bovine tooth dentin e. Journal of Archaeological Science 33:1177-1187.
164 BIOGRAPHICAL SKETCH Bryan Tucker began his college career at Berry College in 1994. After a year at Berry, Tucker transferred to Clayton State College a nd University and eventually to Georgia State University, where he received a BA in anthropo logy in 1998. Tucker continued his education at Louisiana State University, where he received a MA in anthropology in 2003. Finally, Tucker attended the University of Florida, where he graduated with a PhD in anthropology in 2009.