ARCHAEOLOGICAL EVIDENCE OF OYSTER MARICULTURE IN THE LOWER SUWANNEE REGION OF GULF COASTAL FLORIDA By JESSICA APRIL ROSE JENKINS A THESIS PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMEN T OF THE REQUI REMENTS FOR THE DEGREE OF MASTER OF ARTS UNIVERSITY OF FLORIDA 2016
2016 Jessica A. Jenkins
To Naomi Marrow, whose never ending support and encouragement has inspired me in this and every en deavor I have undertaken
4 ACKNOWLEDGMENTS The successful completion of this thesis is thanks, in large part, to the support of all those surrounding me during this process. I would first like to thank my chair, Dr. Kenneth Sasssaman for his guidance a nd patience throughout the entire process of research and writing. I would also like to thank my other committee members, Dr. Neill Wallis of the Florida Museum of Natural History, and Dr. Andrew Kane. I would like to acknowledge Dr. Kane specifically for inspiring the concept of mariculture as the direction of my research after inviting me to join him on a research project with oystermen in Apalachicola Bay, and Dr. Wallis for his confidence in my methods by conducting a similar analysis on oyster shells f rom Garden Patch. I would like to acknowledge Hyatt and Cici Brown for their endowment to the Laboratory of Southeastern Archaeology, because of which this research was made possible. I would also like to thank the Lower Suwannee Archaeological Field Scho ol students and staff of the 2014 and 2015 field seasons for their efforts in excavation at Shell Mound and help with the processing of archaeological materials. Specifically, I would like to acknowledge Megan Lisle for her help in the Laboratory of Southe astern Archaeo logy Furthermore, I would like to acknowledge my colleagues in the LSA who continue to inspire me and provide an unfailing source of support and encouragement. I would also like to acknowledge Dr. Martin Gallivan and Dr. Frederick Smith of the College of William and Mary who have never stopped supporting me as I continu e to pursue my academic career. Finally, I would like to thank my family and fri ends, specifically Naomi Marrow, Karen Mar row, and Virginia Jenkins, who have stood by me throu ghout the process.
5 TABLE OF CONTENTS page ACKNOWLE DGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 8 LIST OF FIGURES ................................ ................................ ................................ ........ 11 ABSTRACT ................................ ................................ ................................ ................... 12 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .... 13 2 OYSTER ECOLOGY AND MARICULTURE PAST AND PRESENT ...................... 21 Introduction ................................ ................................ ................................ ............. 21 General Oyster Biology and Ecology ................................ ................................ ...... 24 Biology ................................ ................................ ................................ .............. 24 Temperature and Salinity ................................ ................................ ................. 25 Substrate ................................ ................................ ................................ .......... 26 Location in the Water Column: Subtidal Versus Intertidal ................................ 27 Optimal Ecological Conditions ................................ ................................ .......... 28 Oyster Mariculture ................................ ................................ ................................ .. 28 Relaying ................................ ................................ ................................ ........... 29 Culling ................................ ................................ ................................ .............. 30 Shelling ................................ ................................ ................................ ............. 30 Size/Age Selection ................................ ................................ ........................... 31 Selective Harvest Location ................................ ................................ ............... 31 Off Bottom Techniques ................................ ................................ ..................... 31 Why Mariculture? ................................ ................................ ............................. 32 Present Day Ecology and Mariculture in Florida Gulf Coast Estuaries ................... 32 Gulf Coast Ecology ................................ ................................ ........................... 32 The Suwannee Sou nd ................................ ................................ ...................... 33 Present ................................ ............. 34 Past Environmental and Ecological Variation of the Lower Suwannee ................... 35 4500 4300 BP ................................ ................................ ................................ .. 36 3200 2500 BP ................................ ................................ ................................ .. 38 2500 BP ................................ ................................ ................................ ............ 39 A.D. 200 650 ................................ ................................ ................................ .... 39 Archaeological Evidence of Mariculture ................................ ................................ .. 41 Examples of Ancient Mariculture World Wide ................................ .................. 41 Methods in the Archaeological Literature ................................ ......................... 45 Discussion ................................ ................................ ................................ .............. 46 3 ARCHAEOLOGY AND CULTURE HISTORY AT SHELL MOUND ........................ 50
6 Introduction ................................ ................................ ................................ ............. 50 Overview of the Archaeology of the Lower Suwannee ................................ ........... 51 Culture History ................................ ................................ ................................ ........ 52 Archaeology at Shell Mound (8LV42) ................................ ................................ ..... 55 Early Excavation ................................ ................................ ............................... 55 Recent Excavation ................................ ................................ ........................... 56 Excavation on the Outside Perimeter of the Ridge ................................ ........... 58 Excavation on the Apex of the Ridge ................................ ............................... 58 Excavation on the Interior of the Slope ................................ ............................. 60 Excavation of the Central Open Area ................................ ............................... 61 Conclusion ................................ ................................ ................................ ........ 62 Palmetto Mound on Hog Island (8LV2) ................................ ................................ ... 62 Discussion ................................ ................................ ................................ .............. 64 4 SAMPLING, HYPOTHESES AND METHODS ................................ ....................... 66 Premises ................................ ................................ ................................ ................. 68 Hypotheses and Impl ications ................................ ................................ .................. 69 Sampling ................................ ................................ ................................ ................. 72 Methods and Techniques ................................ ................................ ........................ 75 Harvesting Locati on ................................ ................................ .......................... 75 Shelling ................................ ................................ ................................ ............. 78 Relaying ................................ ................................ ................................ ........... 78 Off Bottom Techniques ................................ ................................ ..................... 79 Culling ................................ ................................ ................................ .............. 80 Size/Age Selection ................................ ................................ ........................... 81 Discussion ................................ ................................ ................................ .............. 81 5 RESULTS AND ANALYSIS ................................ ................................ .................... 83 Height, Length, and Height to Length Ratio ................................ ............................ 84 Presence/Absence and Type o f Attachment Scars ................................ ................. 88 Presence/Absence of Sponge Parasitism ................................ ............................... 91 Presence/Absence of Parasitism on the Attachment Scar ................................ ...... 94 Left versus Right Shells ................................ ................................ .......................... 98 Left Valve Concavity ................................ ................................ ............................. 100 Biofouling ................................ ................................ ................................ .............. 102 Hypotheses 1 and 2: Location of Harvest ................................ ............................. 104 Hypotheses 3 and 4: Shelling ................................ ................................ ............... 106 Hypotheses 5 and 6: Relaying ................................ ................................ .............. 107 Hypotheses 7 and 8: Culling ................................ ................................ ................. 107 Hypotheses 9 and 10: Size Selection ................................ ................................ ... 108 Hypotheses 11 and 12: Off Bottom Growing ................................ ........................ 109 Palmetto Mound ................................ ................................ ................................ .... 109 Discussion ................................ ................................ ................................ ............ 110 Lower Macrounit (Samples 14 20) ................................ ................................ 111 Upper Macrounit 2 (Samples 7 13) ................................ ................................ 112
7 Upper Macrounit 1 (Samples 1 6) ................................ ................................ .. 113 Palmetto Mound ................................ ................................ ............................. 114 6 SUMMARY AND CONCLUSION ................................ ................................ .......... 117 APPENDIX T TESTS ................................ ................................ ................................ .. 122 LIST OF REFERENCES ................................ ................................ ............................. 126 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 137
8 LIST OF TABLES Table page 2 1 Methods for determining mariculture in the archaeological record. .................... 45 4 1 Inferences regarding environment of harvest and associated attributes evident on archaeological shell ................................ ................................ .......... 77 5 1 Descriptive statistics for oyster height by subsistence column sample. .............. 85 5 2 Number and percent of oysters in each macrounit by size range. ...................... 86 5 3 Descriptive statistics for oyster length by subsistence column sample. .............. 86 5 4 Descriptive statistics for oyster HLR by subsistence column sample. ................ 87 5 5 Presence or absence of attachment scar by subsistence column sample. ......... 89 5 6 Presence or absence of attachment scars Upper and Lower Macrounits. .......... 90 5 7 Presence or absence of attachment scars by Upper Macrounits 1 an d 2 and the Lower Macrounit. ................................ ................................ .......................... 90 5 8 Type of substrate by subsistence column sample. ................................ ............. 91 5 9 Presence or absence of parasitism by su bsistence column sample. .................. 93 5 10 Presence or absence of parasitism by Upper and Lower Macrounits. ................ 93 5 11 Presence or absence of s ponge parasitism by Upper Macrounits 1 and 2 and the Lower Macrounit. ................................ ................................ .......................... 93 5 12 Presence or absence of sponge parasitism on the attachment scar by subsistence column sample. ................................ ................................ .............. 96 5 13 Presence or absence of sponge parasitism on attachment scars by Upper and Lower Macrounits. ................................ ................................ ....................... 96 5 14 Presence or absence of sponge parasitism on attachment scars by Upper Macrounits 1 and 2 and the Lower Macrounit. ................................ .................... 96 5 15 Presence or absence of sponge parasitism on the attachment scar on shells with attachment scars by subsistence col umn sample. ................................ ...... 97 5 16 Presence or absence of sponge parasitism on attachment scars on shells with attachment scars by Upper and Lower Macrounits. ................................ .... 98
9 5 17 Presence or absence of sponge parasitism on attachment scars on shells with attachment scars by Upper Macrounits 1 and 2 and the Lower Macrounit. ................................ ................................ ................................ ........... 98 5 18 Number and percen tage of left and right oyster valves by subsistence column sample. ................................ ................................ ................................ ............... 99 5 19 Percent of right and left valves by Upper and Lower macrounit. ....................... 100 5 20 Percent of right and left valves by Upper Macrounits 1 and 2 and Lower Macrounit. ................................ ................................ ................................ ......... 100 5 21 Number and percentage of shells with each left valve concavity value by subsistence c olumn sample. ................................ ................................ ............ 1 01 5 22 Number and percentage of shells with each left valve concavity value by Upper and Lower macrounit. ................................ ................................ ............ 102 5 23 Number and percentage of shells with each left valve concavity value by Upper Macrounits 1 and 2 and the Lower Macrounit. ................................ ....... 102 5 24 Number and percentage of shells with biofouling present and a bsent by subsistence column sample. ................................ ................................ ............ 103 5 25 Number and percentage of shells with biofouling present and absent by Upper and Lower Macrounit. ................................ ................................ ............ 104 5 26 Number and percentage of shells with biofouling present and absent by Upper Macrounit 1 and 2, and Lower Macrounit. ................................ .............. 104 5 27 Descriptive statistics for oyster height, lengt h, and HLR for Palmetto Mound. 109 5 28 Comparison of mean height, length, HLR, and percentage of parasitism between five sites in the Lower Suwannee research area. ............................... 110 A 1 t Test: Two Sample Assuming Unequal Variances Height .............................. 122 A 2 t Test: Two Sample Assuming Unequal Variances Height .............................. 122 A 3 t Test: Two Sample Assuming Unequal Variances Height .............................. 123 A 4 t Test: Two Sample Assuming Unequal Variances Length ............................. 123 A 5 t Test: Two Sample Assuming Unequal Variances Length ............................. 124 A 6 t Test: Two Sample Assuming Unequal Variances Length ............................. 124 A 7 t Test: Two Sample Assuming Unequal Variances Height .............................. 124
10 A 8 t Test: Two Sample Assuming Unequal Variances Height .............................. 125 A 9 t Test: Two Sample Assuming Unequal Variances Height .............................. 125
11 LIST OF FIGURES Figure page 1 1 Map of the Lower Suwannee Archaeologic al Survey study area (image adapted from Sassaman et al. 2016). ................................ ................................ 17 3 1 LiDAR topographic map of Shell Mound showing excavation units (from Sassaman et al. 2015). ................................ ................................ ....................... 57 3 2 Topographic map showing the location of TU8 (Sassaman et al. 2015:38). ....... 60 3 3 LiDAR topographic map of Palmetto Mound and Shell Mound (Sassaman et al. 2015). ................................ ................................ ................................ ............ 63 4 1 Drawing and photograph of TU8 east profile (Sassaman et al. 2015) ................ 73 4 2 Drawing and photograph of TU8 east profil e (Sassaman et al. 2015: 41). ......... 74 4 3 Photograph of two left valves showing sponge parasitism on attachment scars. ................................ ................................ ................................ .................. 80 5 1 Mean an d range oyster height by subsistence column sample. ......................... 85 5 2 Mean and range oyster length by subsistence column sample. ......................... 87 5 3 Mea n and range oyster HLR by subsistence column sample. ............................ 88 5 4 Percentage of oyster shells with parasitism by subsistence column sample. ..... 90 5 5 Percent of oyster with sponge parasitism by subsistence column. ..................... 92 5 6 Percent of oyster with sponge parasitism on attachment scar by subsistence column. ................................ ................................ ................................ ............... 95 5 7 Percent of shells with attachment scars that have sponge parasitism on the attachment scar by subsistence column sample. ................................ ............... 97 5 8 Percentage of left and right oyster valves by subsistence column sample. ........ 99 5 9 Percent of oyster shells with each left valve concavity value by subsistence column sample. ................................ ................................ ................................ 102
12 Abstract of Thesis Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Master of Arts ARCHAEOLOGICAL EVIDENCE OF OYSTER MARICULTURE IN THE LOWER SUWANNEE REGION OF GULF COASTAL FLORIDA By Jessica April Rose Jenkins May 2016 Chair: Kenneth E. Sassaman Major: Anthropology Archaeological evidence from oyster shells recovered from Shell Mound (8LV42), that maricultural practices were used by coastal inhabitants when the scale and intensity of oyster procurement increased during a time of ritual economy and mound building activity from A.D. 550 650. Shell Mound was transformed from a quotient village si te to a coastal civic ceremonial center, likely related to the burial complex on Palmetto Mound (8LV2), during a time of environmental and sea level uncertainty. The mound was constructed in less than two centuries, accumulating as many as 1.2 billion oyst ers. Mariculture, which has been in practice since at least 2,000 years ago, was likely practiced at Shell Mound as a mechanism to prevent over exploitation of the resource and to sustain the local oyster reefs for daily subsistence, feasting, and mound co nstruction. Analysis of 3,252 oyster shells from a single test unit that encompasses three phases of occupation shows that culling and shelling were practices used when the scale and intensity of oyster harvesting increased.
13 CHAPTER 1 INTRODUCTION People began harvesting oysters on the Gulf Coast of Florida 7,200 years ago, and at about 2,000 years ago, the region was teeming with aboriginal communities that harvested oysters and collected the inedible remains in huge mounds and middens. The importance of oysters to ancient communities in the Lower Suwannee region of Florida is evidenced by the monumental architecture constructed from shell, numerous subsistence refuse middens, and the potential role of this resource in the regional political and ritual ec onomy during feasting or pilgrimage events related to burial complexes such as Palmetto Mound (8LV2). Given the intensity of harvesting at this time, oyster populations may have declined or become depleted in certain areas, forcing communities unwilling or unable to relocate to find ways to sustain production, potentially through the use of maricultural practices. In the broadest sense, mariculture can be defined as the manipulation of a marine resource for economic gain by humans. Mariculture incorporates management of marine resources or making any conscious decisions that could influence the resource, including choices about location, timing, catch limits, habitat enhancement, or transplantation the resource (Lepofsky and Caldwell 2013:2; Lepofsky et al. 2015:237). Maricultural practices can be as simple as shelling the bottom, or returning dead shell to known reefs to encourage recruitment of larval oyster, or spat, or as complex as creating a massive industry outfitted with specialized technology for th e successful reproduction of marine resources. Based on archaeological evidence, maritime management techniques are known to have been in place at least 2,000 years ago on the Northwest coast of North
14 America, supporting ethnographic accounts that managem ent has long been practiced (Brown and Brown 2009; Lepofsky et al. 2015). Marine culturing and management techniques have also been described and documented in other parts of the world; such global archaeological evidence, combined with historical and ethn ographic accounts, indicate that some sort of management of aquatic and marine resources was practiced in the past by coastal communities where resources were heavily exploited. Despite this, investigation into shellfish management, culturing, and cultivat ing practices has not been previously investigated on the eastern and Gulf coasts of North America, where numerous massive oyster shell mounds and middens populate the coast. In the past few decades, archaeological shell of the oyster Crassostrea virgini ca has been analyzed for various other purposes including as an indicator of seasonality of site use (Custer and Doms 1990; Russo 1991), environmental changes such as drought (Harding et al. 2010), subsistence economies, including caloric and protein consi derations (Erlandson 1988), harvesting strategies (Bird et al. 2004; Crook 1992), harvesting location (Schmidt and Haven 2004), and to further the goal of ecological reconstruction (Kirby 2004; Rick and Lockwood 2013). Some general conclusions reached are that oysters are not necessarily a seasonal food, but can be harvested and eaten year round (Russo 1991), and that oysters have a greater protein and caloric value in coastal subsistence economies than previously assumed (Erlandson 1988). While these findi ngs further support the idea that oysters were an important subsistence resource, it is unclear how oyster populations, which are vulnerable to overharvesting due to their ease of accessibility, were maintained during times of intensive and sustained harve st.
15 There have been multiple studies in the archaeological literature that have dealt with the reality of overharvesting and declining oyster populations in estuarine settings (e.g. Erlandson et al. 2008; Lightfoot et al. 1993). Given the abandonment of co astal sites at various points in the past, some researchers have hypothesized that over exploitation of natural oyster populations caused widespread economic collapse (e.g. Dame 2009; Manni n o and Thomas 2002). Despite the evidence of resource decline in ma ny estuarine settings, research has shown, though the analysis of archaeological oyster shell, that over exploitation may not necessarily be the motivating factor of relocation (Doucet 2012). Furthermore, this hypothesis ignores the agency and resiliency o f coastal communities in the face of challenging circumstances, especially as shellfish are ideal candidates for mariculture. Oysters are easy to manipulate by humans as they are resilient, hardy, sessile, suspension feeding organisms that consume food lo w on the food chain (Castagna et al. 1996:676). For example, one culturing technique called relaying can be used to move oysters from intertidal to subtidal conditions, so that oysters grow faster and have a better meat quality. A similar culturing techniq ue is to transplant oysters from areas of high salinity, where they are vulnerable to parasitic and predatory attack, to areas of lower salinity where many oyster predators and parasites cannot survive. Possible reasoning for the lack of research into oys ter culturing practices of coastal peoples who heavily harvested oysters is that signs of oyster maricultural practices in the archaeological record may be ephemeral. Whereas in other parts of the world there is tangible archaeological evidence of culturin g practices (such as the clam gardens on the Northwest coast and the fishponds of the Native Hawaiian
16 agriculturalists), no research has endeavored to establish if similar management processes are archaeologically evident on the eastern and Gulf coasts of North America. Archaeological oyster shell offers a unique avenue of assessment into pre Columbian subsistence practices, as they represent almost perfect indicators of the environment in which the harvested oysters grew. This is important because, in ord er to productivity and success during different life stages. By careful analysis, archaeologists are able to infer ecological conditions under which oysters grew, such a s salinity levels, tidal conditions (intertidal versus subtidal), and substrate. Other attributes of oyster shells allow assessment of the relative health of the oyster when harvested or whether the oyster was alive or dead upon collection. With the use of maricultural techniques, these ecological variables can be manipulated by humans not only to sustain oyster terms of nutrition and taste. The hypothesis of this thesi s is that if coastal populations in the Lower Suwannee regularly and intensively harvested oysters during times of increased sedentism 2,000 years ago, maricultural practices were employed to sustain the resource. To test this hypothesis, archaeological sh ell from Shell Mound (8LV42) in the Lower Suwannee region of Florida was evaluated for indications of oyster harvesting and culturing practices. The area of focus for this study is the Lower Suwannee region of the northern Gulf Coast of Florida (Figure 1 1). The study location encompasses a 47 kilometer
17 Figure 1 1. M ap of the Lower Suwannee Ar chaeological Survey study area (image adapted from Sassaman et al. 2016)
18 stretch of Gulf coastal lands and surrounding islands of Dixie and Levy counties, and wa s designed as part of a research endeavor known as the Lower Suwannee Archaeological Survey conducted by the Laboratory of Southeastern Archaeology at the University of Florida. The research area is bounded by Horseshoe Beach to the north and Cedar Key to the south, with the Suwannee River in the middle of the coast. This coastline is particularly attractive for sustained settlement because of the plethora of exploitable resources produced by a rich estuarine and intertidal environment. Notable among the re sources of economic value to humans is the Eastern Oyster ( Crassostrea virginica ). Although the specifics of environmental change are debated, the Gulf Coast has been subject to periods of sea level rise and transgression along with periods of drier condit ions during the late Pleistocene and wetter conditions during the Holocene (Wright et al. 2005:621). Sea levels have risen ~100 meters since the end of the Ice Age, which has resulted in the flooding of about half of the original Florida peninsula (Sassama n et al. 2011:1). Environmental changes such as this have affected the type and availability of marine resources, including oysters. Sea level rise decelerated during the middle to late Holocene, with an average rise of ~0.16 cm/year between 7500 and 5500 cal yr B.P., to ~0.07 cm/year between 5500 and 2500 cal yr B.P., and slowing to ~0.05 cm/year between 2500 and 750 cal yr B.P. (Wright et al. 2005:631). Hine et al. (1988) propose that this sea level rise deceleration established the modern coastline of th e Lower Suwannee and allowed for the establishment of oyster reefs, or bioherms. From what can be understood archaeologically about the region from sites that are not submerged due to rising sea level, the Lower Suwannee has been a region of
19 occupation for people, perhaps in greater numbers than that of today, starting at least 4,500 years ago (Sassaman et al. 2011). The study area once consisted of at least 20 earthen mounds, two large shell mounds, smaller mounds, linear ridges, and large U shaped middens many of which are now destroyed because of sea level transgression (Sassaman et al. 2011:12). Of particular importance to this study is the Woodland period site of Shell Mound (8LV42), 10 kilometers north of Cedar Key. Shell Mound is the largest and mos t formalized above ground mound in the research area, being 190 x 180 meters in plan and about 7 meters tall. The mound is comprised primarily of oyster shell, with an estimate of between 420 million and 1.2 billion oysters being deposited in the mound in less than two centuries (manuscript on file at LSA). This mound is one of about two dozen archaeological sites in the vicinity, many of which are arcuate and linear shell ridges. Oyster shell from bulk samples recovered from a deeply stratified test unit a t the apex of Shell Mound provide the samples used in this research. In terms of the structure of this thesis, Chapter 2 describes aspects of mariculture from the present to archaeological evidence of past aquatic management practices from various places a emphasized, especially among ancient non agricultural societies, in order to propagate the regular harvesting of important marine resources. In this chapter, relevant biological and e cological factors affecting oyster populations and mariculture are described. An intimate understanding of these variables is crucial for the successful management of oyster populations as well as a nuanced interpretation of archaeological data. Furthermo re, due to the dynamic nature of estuarine settings, the ecology of the Lower
20 Suwannee estuary is described. Chapter 3 outlines the archaeological and cultural history of the study area as well as the recent excavations in the Lower Suwannee, focusing pri marily on Shell Mound and Palmetto Mound. Within this chapter, the chronology of the Shell Mound is established by century (A.D. 200 700) based on radiocarbon dates and excavation results. Chapter 4 explains the sampling strategy and methodology employed t o collect relevant data pertaining to oyster harvesting and possible management practices. Chapter 5 reveals the results of the research as well as discusses the results in a broader context. Chapter 6 concludes the thesis with recommendations for future r esearch based data that can be extrapolated from archaeological oyster shell.
21 CHAPTER 2 OYSTER ECOLOGY AND MARICULTURE PAST AND PRESENT Introduction specifies the husbandry or culturing of marine animals and plants, such as shellfish and finfish (National Resource Council 1992:9). According to the National Research Council on Marine Aquaculture, there is evidence of aquaculture from the earliest records of human history, and it is a rapidly growing industry in many parts of the world today controlling the envi ronment in order to obtain a greater yield of a desirable product from some human intervention and manipulation, however primitive, Mariculture is being used today in many parts of the world as important natural marine populations have been rapidly decl ining. For example, oyster reefs have declined 85 percent from historic levels worldwide, and the oyster industry has been in steady decline for over 70 years (National Research Council 1992; Seavey et al. 2011). Among the reasons for such decline are dise ase epidemics, pollution, and blooms of toxic algae, but the primary reasons are overfishing and habitat loss (National Research Council 1992:29). To respond to these issues, especially those concerning public health, many oyster beds have been closed to h arvesting, although people are now taking advantage of the opportunity to use mariculture in order to revitalize oyster
22 production. Management efforts to restore oyster reefs include conservation of available shell stock for replanting as cultch, protectio n of spat through cull laws, seasonal closures or the protection of productive areas, and restrictions on harvesting gear (Castagna et al. 1996:675). Despite these efforts, oyster populations have not fully rebounded, and as of a 1992 report, the U.S. mari ne aquaculture industry had not sufficiently demonstrated long term economic viability (National Research Council 1992:9). not immune to the overall global decline and collap se of oyster reefs. For example, Apalachicola Bay, the Suwannee Sound, and Tampa Bay have all been suffering from natural and anthropogenic effects on local reefs, hindering large scale commercial production of oysters in these areas (Seavey et al. 2011). Recently, local community mariculture endeavors as well as harvesting laws have been enacted to protect and sustain what is left natural reefs for hopeful future proliferation (Castangna et al. 1996). Although reasons for the declining oyster production i n the past may be different from that of today, maricultural practices may have been very similar For example, although oysters did not suffer from the same pollution that is destroying reefs today, reefs of the ancient past were affected by ecologically stressful events and large scale environmental change, including periods of drought and flooding, climate change, and sea level rise and regression (McFadden 2015; Sas saman and Wallis 2015; Wright et al. 2005). All of these changes affect the type and availability of marine resources and create differentiated access for humans to economically viable oyster reefs through
23 time. Unstable conditions in the Gulf create stres sful conditions for oysters to live in; the stress on their populations is exacerbated by regular, heavy, and at times intensified, exploitation, further weakening oyster reefs, making them susceptible to collapse. Cultivating oysters is not only importan t in order to sustain exploitation by coastal peoples, but also because oysters are a keystone species, meaning that they provide many essential ecological services such as maintaining food webs, regulating water quality, creating habitat for numerous fish crustacean, bivalve, mollusk, and gastropod species to survive, and protecti on coastal lands (Beck 2011; Kennedy 1996; Kilgen and Dugas 1989). Oyster shells were also used by pre Columbian people as building material, in mortuary practices, and as tools and ornamentation. Because of their significance in many aspects of life, the maintenance of oyster populations is a priority for long term coastal occupation by ancient and modern coastal peoples alike. Archaeological evidence of mariculture is apparent in many parts of the world, where coastal communities of the past implemented management practices to sustain marine exploitation (Nash 2011; Lepofsky et al. 2015). Both shellfish and fish were cultivated, and the cultivation practices not only affected th e resource, but also had important cultural and political implications for non agricultural communities, making research into possible culturing practices valuable to broader archaeological interpretation (i.e., Grier 2014). Several points are discussed i n this chapter that are critical for interpreting oyster mariculture in the Lower Suwannee. First, I will discuss the relevant biological and ecological attributes of the Eastern oyster, Crassostrea virginica, which must be understood in order for people s uccessfully to cultivate the resource. Next, I describe
24 oyster mariculture, and review the various types of practices that are used to manipulate ecological conditions so that oysters prosper. Then I outline the specifics of those ecological conditions alo day ecological conditions and maricultural practices. I then use those same basic ecological variables to structure a discussion of changing environmental conditions in the Lower Suwannee region of F implication of the shifting environment on natural oyster populations. Finally, I use global archaeological examples of mariculture to lead to a concluding discussion of the hypothesis that maricu lture was possibly practiced by people 2,000 years ago in the Lower Suwannee region. General Oyster Biology and Ecology The study of oysters from an archaeological perspective first requires knowledge of basic biology and ecology. Understanding the natur al aspects of oysters allows for a nuanced interpretation of how oyster populations of the past were affected by dynamic estuarine conditions and long term environmental change, how human predation intervene to sustain oyster populations during times of intensive harvest. Biology Oysters are bivalves, meaning that they have two calcareous valves, or shells, that protect the soft body of the organism. The two valves are joined at the dorsal edge by a resilient hinge ligament and interlocking hinge teeth (Eble and Scro 1996:19). In order to filter feed, oysters open and close their shells by contracting and relaxing the pair of abductor muscles running between the valves (Kent 1988:3). The adductor mu scle scar is visible in the form of a dark purple half moon shape on the inside of the
25 valves. Oyster valves are asymmetrical, with the left valve being larger and more deeply cupped than the right, which tends to be flat. Oysters usually attach on the lef t valve, meaning that the right is generally uppermost. The shape and quality of oysters are influenced by four main ecological variables discussed below: temperature, salinity, substrate, and location in the water column (Supan 2002). Temperature and Sal inity While some argue that temperature is the most important ecological variable affecting the success of oysters (e.g., Gunter 1957), others argue that salinity is the most important variable (e.g., Butler 1949). In reality, the synergistic effect of the se two variables likely has the most profound effect on the success of oyster populations (Shumway 1996). Together, these variables affect feeding, respiration, utilization of food reserves, gonadal development and time of spawning, parasite disease intera ctions, predation rates, growth, and distribution (Shumway 1996:467). These factors are and salinity are more sensitive during the mobile larval stage or reproductive period than at any other stage in their lives (Dame 2012:44). Temperature is important as a determinant of oyster distribution and growth rates; as temperatures increase, growth rates generally increase and then gradually decrease towards the fringe of t heir tolerance (Dame 2012:45). Oysters are found in waters where the annual range is from 2C to 36C, and can often stand extremes for short amounts of time (Shumway 1996:468). The general rule is that warmer waters, such as in the Gulf of Mexico, facili tate the rapid growth of oysters, not only in height, but in volume of meat produced (Shumway 1996:484). Temperature not only affects oyster growth rates, but also respiration, feeding, excretion, and spawning, all of which
26 are crucial for the success of o yster populations. In most of these aspects, oysters are the most successful in the intermediate range of their temperature tolerance (~15C), with factors such as spawning being negatively affected by extremes (Shumway 1996:474 475). Much like temperatu re, salinity levels impact multiple aspects of the ability for oyster populations to survive and thrive, with the primary effect being population distribution. Salinity is also a highly dynamic variable, with variations in estuaries being diurnal, seasonal or spatial with gradual or abrupt changes (Shumway 1996:467). Salinity in estuarine and coastal environments varies along a gradient, generally in accordance with distance from freshwater inputs. Oysters are able to survive in estuarine environments wher e salinity ranges from oceanic levels of 35 ppt to areas where the influence of freshwater brings salinity to as low as 1 2 ppt. Despite the ability of C. virginica to survive in a wide range of salinity levels, there are conditions under which success is more likely. For example, there are optimal salinity conditions that ensure the organisms success in different life stages: 10 15 ppt for larval development, 10 29 ppt for larval growth, 16 22 ppt for spat settling, and 10 30 ppt for juveniles and adults (Patillo et al. 1997). Generally, the optimum salinity level for oyster populations is 14 make the oyster vulnerable to parasites that can only live in high salinity waters (Shumway 1996:468). Subs trate While temperature and salinity are instrumental to the success of oyster populations, the type and availability of suitable substrate and bottom conditions are also important factors. Oysters are gregarious, meaning that, while they are larvae, their
27 settlement preference for attachment is their own species, forming aggregations of conspecifics in the form of oyster beds or reefs (Kennedy 1996:396). Oyster reefs, composed of both live and dead oysters, tend to form in estuarine environments where oyst ers have already settled on muddy sand bottoms with a scattering of hard substrates, and where ecological conditions are favorable. Excessive siltation slows the development of extensive oyster beds, and on clean, well sorted sand, wave action regularly di slodges oysters and removes newly settled larvae and young spat (Kent 1988:8). Location in the Water Column: Subtidal Versus Intertidal The final ecological aspect of oysters that is relevant to this study is the depth of the water that oysters grow in an d where they fall in the water column. Because of the generally shallow nature of their habitat, oysters grow both subtidally and intertidally. This difference greatly affects how oysters grow, their success rates, and their economic value to human populat ions. Subtidal oysters often have ovate to subovate shell outlines and have regular refuge from the elements, as well as longer feeding times, increasing their growth rate to about 1.5 times that of intertidal oysters. Subtidal oysters in high salinity are as also often are affected by predators and parasites such as boring sponges, worms, and boring clams. The shells of subtidal oysters are usually thicker with increased valve cupping (Lawrence 1988:267). In contrast, intertidal oysters grow in tight clumps or burrs, causing their shells to be relatively small, thin, and elongate. Intertidal oysters benefit by having refuge from marine predators as well as freedom from competition.
28 Optimal Ecological Conditions Oysters have specific conditions under which th ey are naturally most prosperous. The optimal natural conditions for oysters to survive and thrive are subtidal environments with salinities between 14 28 ppt, temperatures close to 15 C and in areas with muddy sand bottoms. The four ecological variables discussed above can be manipulated by humans to affect oyster production as well as encourage desirable qualities such as taste and nutrition. In order for humans to manage the resource successfully, optimal conditions are slightly different than what they would be naturally; for example, although oysters naturally grow faster in high salinity waters, the most favorable conditions for oysters to be grown in are areas that allow seed beds to be bathed in water of reduced salinity, to control for the negative effects of parasites and predators in higher salinity waters (Korringa 1976; Shumway 1996). For oyster farmers today, it is also important to know how to grow oysters under conditions where they will have the fastest growth, the firmest meats, the best f lavor and the deepest shells. Balancing these specifications often makes the best places for growing oysters in the Gulf today to be near the shore, especially in the cooler months (Korringa 1976:64). Oyster Mariculture Oyster farming includes enhancing su bstrate for larval settlement, transferring oysters to sites where they do not settle naturally, and protecting the crop against predators, parasites, and competitors (Korringa 1976). Maricultural practices can be as simple as tending to known reefs, shell ing the bottom and creating habitat, or as complex as creating an industry with technology specifically geared toward the specialized production of oysters.
29 Mariculture is necessary to sustain oyster populations if natural stocks become depleted and need t o be enhanced by reseeding of the bottom. The modern process of can be accompli shed today by buying seed from hatcheries, which is often costly, or the collection of natural seed. The collected seed is then placed in the water by oyster farmers at strategic times and locations (Korringa 1976:2; National Research Council 1992:212). Th e six types of oyster mariculture being investigated for this study are relaying, culling, shelling, size/age selection, selective harvest location, and off bottom techniques. These maricultural practices were chosen because of their likelihood of being pr acticed in the Gulf Coast 2,000 years ago, based on archaeologically identified practices of shellfish management in the past, and also the feasibility of being able to determine these practices from the archaeological record in the Lower Suwannee. Relayi ng Relaying is the term used for moving live oysters from areas of an estuary that are suboptimal and transplanting them to areas of optimal conditions. For example, intertidal oysters can be relayed to subtidal locations in order to increase their growth rates. Also, oysters in high salinity areas can be moved to areas of lower salinity to decrease the number of predators and parasites. Although this method is effective, suddenly altering the conditions under which oysters are living could increase the mor tality rate. In culturing methods today, once oyster spat are about 2.5 3.8 cm in shell height, they are considered large enough to be transplanted. Oysters are transplanted to areas where they grow more rapidly and are intentionally planted at lower densi ties to allow for regularity in their growth patterns. After the oysters are
30 moved to new locations, it generally takes 13 months to five years before the oysters reach legal harvest size based on local environmental conditions (Castagna et al.:675 676). C ulling Separating individual oysters from larger aggregations (i.e., clumps or burrs) is called culling. Culling includes separating oysters from one another, as well as removing biofouling organisms and dead shell from healthy oysters. Oysters are often c ulled while oystermen are still on the water; the removed spat, dead shells, and oysters that are less than legal harvest size (3 inches in the US) are returned to the water. Oysters that are not yet large enough to be harvested are returned to the water a they can grow without hindrance of other organisms; this allows oysters to grow without stress of competition, making their shells more deeply cupped and rounded, which is considered ideal for the health, growth, and reproductive capabi lities of oysters. Shelling Shelling is the process of returning dead shell to existing oyster reefs or in areas of ideal ecological conditions that could support an oyster reef. Returning dead shell to the water is one step towards ameliorating habitat de pletion. Without suitable habitat for oysters to settle on, populations would collapse. Because oyster larvae have a natural tendency to settle on other oysters for attachment as spat, studies have shown that the best artificial substrate, or cultch, for c ulturing oysters is clean, seasoned (i.e., air dried for about 12 months) whole or partially crushed oyster shells (Crisp 1967; Castagna et al. 1992:684). The process of laying down cultch in areas where reefs are depleted or nonexistent is to encourage sp at settlement and growth in the most suitable environmental conditions. Today cultch is usually obtained from stockpiled shells from
31 shucking houses or dredged from fossil beds, and is put back into the water over reefs and estuary bottoms in areas where l arval settlement is generally high due to hydrography and other suitable environmental conditions (Castagna et al. 1992:675). The cultch is then left relatively undisturbed. Size/Age Selection The selective harvesting of oysters of a certain size or age is a maricultural practice that allows for populations to be sustained in a certain area. For example, by harvesting large numbers of oysters that have not lived through reproductive cycles, the population size will eventually diminish. Today this is practic ed with the enactment of under the legal limit (3 inches) must be returned to the water to continue growing before they can be harvested. Selective Harvest Location By selectively harvesting oysters from certain locations, others are left undisturbed so that populations can naturally rebound. If all reefs in an area are harvested regularly and at the same rate, it is likely that, should collapse happen, all of the ree fs would collapse at the same time. Due to resource depression in many estuaries today, oyster reefs have been subject, by law, to closures in hopes that populations will naturally replenish themselves without human predation pressures slowing the process. Off Bottom Techniques Oysters can be grown directly on estuary bottoms or using hanging culture techniques on ropes, rafts, nets, or in cages. Growing oysters in this three dimensional mode allows them greater access to food and protection from sediment ation and
32 benthic predators. Also, more oysters can be grown per unit area in this way than by bottom culture methods (National Research Council 1992:212). Why Mariculture? In 2009 the Nature Conservancy released a report that there have been oyster loss es worldwide of up to 85 percent, and both anthropogenic as well as ecological factors have been blamed for such a loss. Researchers have cited various forces for this loss including erosion and storm damage (Goodbred and Hine 1995; Seavey et al. 2011), po llution and disease (Beck et al. 2011), sea level rise (Wright et al. 2005), and overharvesting (Jackson et al. 2001; National Research Council 1992). Also, unstable salinity regimes, dredging, hurricanes, channelization, watershed projects, activities of the petroleum industry, and predators can all adversely impact oyster reef communities (Kilgen and Dugas 1989). Although there are many stresses that can devastate oyster communities, Jackson et al. (2001) argue that overharvesting is the primary contribut or, explaining that such unsustainable large scale harvesting is responsible for the majority of the 52 fold decline of oyster populations they studied in the Chesapeake Bay, with decline in water quality and disease as secondary contributing factors. Each of the six techniques described above help sustain oyster populations during times of resource decline and depression, and are especially impactful when used in conjunction with one another. Present Day Ecology and Mariculture in Florida Gulf Coast Estuar ies Gulf Coast Ecology lying terrain interspersed with swamps and marshes, and extensive offshore oyster reefs. The climate is generally mild to warm, with hot wet summers and drier winters. Along the coast are numerous
33 areas of freshwater inflow from small and large rivers and springs. Estuaries are produced in areas where freshwater inflow meets the higher salinity water of the Gulf of Mexico, and are highly productive environments for a variety of fa una to survive. Shumway (1996:468) describes Gulf estuarine environments to be prime locations for high oyster growth rates and reproductive capability because of generally optimal salinity and temperature conditions. Oysters are abundant in the shallow, brackish water of estuarine environments and seasonal variation of influential ecological factors that may have different ranges of water quality in different geographic locations. While there are some forty estuarine distributed within and between estuaries, creating areas with differential capabilities of natural oyster productivity (McNult y et al. 1972). The Suwannee Sound The Lower Suwannee has been characterized as the one of the largest low gradient open marine marsh shorelines in North America (Wright et al. 2005:621, 623). The Suwannee River, the main source of freshwater inflow to th e estuary, is located in the middle of the estuary and is the second largest river system in Florida (Mattson 2002). Salinity levels in the Suwannee Sound are less than 0.01 ppt in the Suwannee River, and reach near 35 ppt at sampling stations that were th e farthest offshore, approximately 31.5 kilometers from the coast (Bledsoe and Phlips 2000:461). The average salinity has been gaged in parts of the estuary as between 21 and 28 ppt, fluctuating with storm and drought events (Baker personal communication) The Suwannee estuary is unique in that the brackish transitional zone between freshwater
34 and seawater is much more extensive than in any other Florida Gulf Coast estuary (Mattson 2002:1337). Oyster habitats are the main structural habitat feature in the Suwannee Sound, with the best reef development in areas of reduced salinity (Mattson 2002:1338). Present practiced in conjunction, allowing for enhancement of substrate as well as letting less mature oysters to grow to harvestable size. Oystermen in Apalachicola Bay harvest oysters using oyster tongs or oyster rakes. These tools allow the oystermen to stand on their boats and harvest oysters at the bottom of the Bay without diving for them. The tongs are nonselective, and a single tong lick will pick up anything from large live oysters, to spat, jelly fish, crabs, mud, dead shells or substrate, or anything else in the general area. Once the to ngs are filled, they are raised out of the water and placed onto a board on the boat where they can be sorted and culled According to law, oystermen are allowed to only harvest oysters that are of legal size, which in this case is three inches. The smalle r oysters and all dead shell are returned to the water to continue to develop and act as substrate to attract spat settlement, thereby encouraging population rebounding, or at least maintain the current population. As well as oyster tongs, other important tools that were used included a boat of proper size and proportion, a small rake to be able to sort oysters, a sharp shooter shovel to easily return unwanted materials back into the water, a tool used for feeling the estuary bottom in order to find oyster reefs, buckets or bags for collection, and a tool that was used to break up clusters of oysters, or burrs, removing dead shell and separating single oysters. This tool also has another purpose: at the opposite end is a
35 measuring tool that is three inches, so that an oyster is easily measured to see if it is of legal harvest size before it is collected for distribution or consumption. All steps of this process can be done alone, but are much faster when done with at least one other person. Because the Suwan nee Sound is not a protected area like Apalachicola Bay, culturing practices are different. Instead of shelling the bottom for encouraging spat growth, spat are produced and harvested from protected areas, or grow houses, and when they are large enough the y are transferred to leases in the water where they are grown in bags. Every few months, depending on growth rates, the oysters are put into new bags until they reach harvestable size. Based on the location of the leases, oysters are exposed to different e growth than others. The oyster bags are regularly checked and flipped to avoid sedimentation and to discourage fouling. Here, the maricultural techniques being used include a mix of transplanta tion, size selection, and off bottom cultivation. The two examples discussed above exemplify how different culturing techniques can be used in conjunction, and that different culturing techniques are appropriate for different locations based on environmen tal factors. Not only do we as researchers need to have an understanding of the environment and ecology of oysters to be able to discern mariculture in the archaeological record, but people in the past who were manipulating oyster resources would too in or der for their efforts to be successful. Past Environmental and Ecological Variation of the Lower Suwannee From at least 4,500 years ago, the coastal dwellers of the Lower Suwannee have been subject to periods of dramatic environmental change which likely impacted important marine resources to a great extent. Aside from a steady supply of food
36 regulated by water flow and tidal action, the success of oyster populations is dependent on four main ecological factors (discussed above): water temperature, salinit y, substrate, and location in the water column (Supan 2002). Although oysters can withstand wide ranges of ecological conditions, sudden, drastic, or prolonged changes in these conditions, such as those that occurred throughout pre Columbian history in the Lower Suwannee, can cause reefs to collapse and subsistence practices to change. Below is a projection of major shifts in sea level and climate in the Lower Suwannee region. Because of the lack of fine grained detail on the exact aspects of how global or regional climatic events and sea level fluctuations were manifested in the Lower Suwannee, the following is the best interpretation based on the limited data to date. While acknowledging the short comings of some of the data, I believe this is a useful exe rcise in discussing the unstable nature of estuarine environments, the risk involved in coastal living, how oyster populations respond to major climate events and change, and how variable natural conditions coupled with social and cultural change through t ime may have instigated human intervention into marine resources. 4500 4300 BP level, during the time of occupation from about 4500 4300 years ago, conditions on st were likely drier and salinity levels were higher. Such observations of environmental conditions dating to this time (~4300 cal. yr. BP) are drawn from proxies from the Ehrbar site (8LV282) near Cedar Key. Excavations from two test units at the site sho w high proportions of high salinity species, such as scallop, to oyster in shell middens (McFadden and Palmiotto 2013). If the contents of the midden were collected from local resource patches, it would follow that the salinity of the surrounding
37 waters wa s high. This is important as salinity is arguably the most important variable for tolerance produce larger and denser oyster populations. Hopkins (1957:416) argues that oysters are more prolific, larger, and more prosperous in areas of high salinity, although high salinity predators and parasites, such as oyster drills, barnacles, and bioeroding sponges, can be a limiting factor, increasing mortality rates. Also, salinity and species diversity are positively correlated; therefore, if salinity levels were high during this early phase of occupation in the Lower Suwannee, there was likely a plethora of exploitable resources, distributing predation pressure on multiple species (Wells 1961:262). Shellfish ratios from Ehrbar show that salinity levels were higher during the earliest phase of occupation, and lessened during later occupations, indicating observable environmental change within this period (McFadden and Palmiotto 2013 :35). Using a single component site as the sole indicator of environmental conditions in a larger region is inherently problematic, especially in the absence of independent data on environmental change (Sassaman et al. 2010:139). As salinity is the main c ontrolling factor of the distribution of marine species, it follows that the species within a midden could be used as a proxy for local conditions. This, though, disregards human agency and preference; perhaps high salinity species were preferred, and thus inhabitants of the Ehrbar site travelled to high salinity waters to acquire resources which were later disposed of in the middens. More samples from midden sites within this time period need to be taken throughout the Lower Suwannee to see if the trend is
38 replicable, and independent paleoclimatic data should be established as a further means of verification. 3200 2500 BP The next notable time period of change in the region is from ca. 3200 2500 BP; this period is marked by structural change in the Southea st, and, in the Lower Suwannee, the absence of coastal sites (Thomas and Sanger 2011). This period is accompanied by a global climatic event called the Neoglacial, where precipitation increased, temperatures dropped, sea level rose, and instances of river flooding increased (Kidder 2006). If these global events also shifted the environmental conditions in the Lower Suwannee, salinity and temperature would have changed significantly, likely impacting oyster populations. Increased precipitation and instances of river flooding during this time would have significantly lowered the salinity of the Lower Suwannee estuary, although sea level rise may have offset this to an extent. During times of ecological stress, oysters have the ability to tightly close their s hells to protect them from sudden changes in ecological conditions. Oysters are able to adjust to salinity ranges from 3 35 ppt in only a few hours, but only have the ability to survive for up to two weeks when salinity levels drop below their normal thres hold (Pearse and Gunter 1957:135, 139 140). Therefore, where it has been shown that salinity levels were likely high in the preceding period, such a dramatic drop in salinity levels would have possibly killed off, depleted, or significantly altered oyster reefs in the area. If populations did rebound or stabilize, populations of small, roundish oysters, which are less desirable or economically beneficial to humans.
39 The other significant environmental change that would affect oyster populations is the drop in temperatures, specifically when combined with the regression of sea levels during this period. Low temperatures decrease growth rates, and if temperatures are regula replaced with more tolerant forms (Dame 2012:45). Furthermore, the effects of temperature tolerance are differentially experienced by oyster populations depending on the depth of w ater they are in; oysters in very shallow water will experience a wider range of temperatures following seasonal and daily climatic variations (Dame 2012:45). Therefore, as sea level was lower during this time, subtidal oyster populations, that flourished in the preceding period of high salinity and warmer temperatures, likely became intertidal and highly vulnerable to decreased temperature and salinity. 2500 BP At about 2500 BP, human populations returned to the coast in dispersed settlements. By this tim e, the modern coastline was more or less established, and there was a period of oyster bioherm growth and marsh aggradation (Wright et al. 2005). At about this time, environmental conditions became more favorable for estuarine resources such as oysters as it became warmer, sea level rose, and conditions reached A.D. 200 650 An important change affecting people living along the coast of the Suwannee Estuary was an incr ease in sea level at about A.D. 200 300, when large sites such as Crystal River and Garden Patch became the loci of occupation in the region as people dispersed from vulnerable coastal locations in the Lower Suwannee, abandoning previous sites, leaving the m to be inundated by the rising sea (Goodbred et al. 1998;
40 McFadden 2015; Sassaman and Wallis 2015). Large aggregations of populations at these civic ceremonial centers may have threatened their long term sustainability, as both were eventually abandoned ( Sassaman and Wallis 2015). The abandonment of Crystal River and Garden Patch at about A.D. 600 occurs around the same time as a 200 year period of cooling in the Lower Suwannee, which, as discussed above, could negatively impact estuarine resources, especi ally if coupled with heavy, non sustainable, exploitation (Walker 2013). At around the same time, Shell Mound (8LV42) came online as a place of habitation and ritual activity, culminating in the erection of a massive U shaped mound, constructed primarily o f clean oyster shell. This series of events likely had important consequences for oyster populations in the Lower Suwannee. Although sea level rise at A.D. 200 300 may have been one of the causes for abandonment of coastal sites in the Lower Suwannee, it likely had a positive effect on oysters, creating more subtidal conditions under which oysters flourish due to the steady supply of food. Whereas the change in tidal conditions may have initially negatively impacted oysters, populations could have easily r ebounded, especially as rising sea level and warmer temperatures created more subtidal and saline conditions for oysters to flourish. Furthermore, as the sites were largely abandoned, human predation pressure on oyster populations was relieved, allowing th em to proliferate. Unlike those proximate to Garden Patch or Crystal River, the oysters in unpopulated locations in the Lower Suwannee may not have been as vulnerable to a period of cooling, as their populations were less stressed than those which were reg ularly harvested. In fact, the interplay of salinity and temperature on oysters could be very important, as oysters are able to survive in extreme salinity conditions, but their
41 chances of survival are increased when accompanied by lower temperatures (Heil mayer et. al 2008). The evidence of sustained oyster harvesting at Shell Mound begins at about A.D. 500, with the accumulation of a dense shell midden on the distal arm of the relict dune. Oyster harvesting at Shell Mound likely intensified, reaching its peak at about A.D. 550 600, with the build up of the mound into a monument through the emplacement of massive amounts of clean shell. Although there are myriad reasons for resource collapse today, overharvesting is still cited as the primary cause (Nation al Research Council 1992; Jackson et al. 2001). With changing environmental conditions coupled with such intensive oyster harvesting in the Lower Suwannee at sites such as Shell Mound as well as other contemporaneous sites in the vicinity, it is likely tha t the effects of environmental instability and overharvesting were offset by maricultural practices in order to sustain oyster reefs, not only for subsistence, but also as building material for monumental architecture. This hypothesis is supported with arc heological and ethnographic evidence that mariculture during this time was practiced by various groups around the world. Archaeological Evidence of Mariculture Examples of Ancient Mariculture World Wide Ethnographic accounts of mariculture exemplify their importance for coastal communities. For example, ethnographic research in Canada concerning traditional ecological knowledge passed down among the First Nations peoples conclusively demonstrates that mariculture and management techniques have always been a part of how coastal peoples interacted with their environment (Brown and Brown 2009). Also, experts in marine aquaculture, fisheries biology, fisheries management, and ocean and
42 coastal management, among others, agree that mariculture in some form is ne cessary to sustain regular exploitation of marine resources (National Resource Council 1992:29). The management of shellfish for coastal people is also of particular importance as they occupy a peculiar space in terms of subsistence: they share many charac teristics of a plant in that they are sessile, easy to harvest, and usually grow in large patches, but, unlike plants, they offer the protein benefits of an animal (Whitaker 2008:1115). Therefore, if sedentary coastal communities relied on harvesting large amounts of shellfish as a major source of protein, then mariculture may have been used to sustain those important resources. Recent archaeological investigation on the Northwest Coast of North America has shown the use of mariculture in the form of clam beginning at least 2,000 years ago (Grier 2014; Lepofsky and Caldwell 2013; Lepofsky et al. 2015). Clam gardens are constructed of rock walls in the lowest intertidal zone and tidal flats which have been cleared of boulders. These archaeological features exemplify many aspects of management and culturing; clam gardens enhance the marine resource by creating niches in which clams were highly productive and readily as often clams only of certain sizes or ages were collected, perhaps in an attempt to allow clam beds to repopulate and reach harvestable size to avoid resource collapse (Lepofsky and Caldwell 2013:6). Management techniques such a s these are concrete marine management system resulted in long term sustained and sometimes enhanced
43 Sim ilarly, morphometric analysis of archaeological mussel shells from middens on the West Coast show that past local inhabitants were manipulating shellfish (Whitaker 2008). plucked beds were then left fallow for about two years while other beds were being exploited so to not deplet e the resource. This culturing technique allowed coastal peoples to invest in long term sustained yields of mussels through the sacrifice of maximum immediate returns (Whitaker 2008:1121). Like clams and mussels, oysters have also been manipulated through time using mariculture. The Romans of 2,000 years ago are often attributed with the fist definitive cultivation of oysters, as observed from hanging techniques depicted on vases. Based on the artwork, the oysters were cultured using hanging ropes of heart y material attached to a framework of sticks or poles in the water, a practice still used in the same location today (Gunther 1897:364). Using archaeological oyster shell to interpret potential signs of culturing in the past, Rakov and Brodianski (2007, 20 10) have analyzed oyster shells from middens of several northern Pacific coastal Neolithic and Early Iron Age sites in Russia, as well as ancient and modern oyster farms and argue that the existence of incipient oyster cultivation on the coast of Peter the Great Bay can that the core of aquaculture is in pest control and the sorting of mature individuals by size and age. They argue that both of these aspects are evidenced in the Boisman and Yankovsky shell middens, and also demonstrate that certain morphological features of
44 on bottom cultivation distinguish the oysters in the middens from natural oyster populations; these morphological features include smooth shells lacking in radial ribbing caused by sunlight, unscalloped edges, and small attachment scars (Rakov and Brodianski 2007:40 41). Archaeological evidence also reveals the cultivation of marine fishes. In Hawaii, for example, integrated farming approaches were used by native people where extensive agricultural endeavors were accompanied by equally as extensive maricultural activities. For example, large numbers of fish were raised in salt and freshwater ponds. Once the ponds were created, they were managed and contro lled by the elite (Costa Pierce 1987). The decisive evidence of mariculture has broader interpretive worth than simply understanding subsistence practices. For example, Grier (2014) discusses how the cultivation of clams and the creation of clam gardens on the Northwest Coast can be used as a means to address issues of complexity among non agricultural communities. Purposeful manipulation of abundant, predictable, and aggregated resources and the nization of Northwest Coast societies; specifically, the successful cultivation of important marine resources to create surplus is one of the conditions deemed necessary for the emergence of Northwest Coast cultural complexity (Grier 2014:211). The author argues that through the manipulation of the environment and resources, food production and diversity were sustained and social dynamics were transformed, particularly in regard to resource (clam garden) ownership (Grier 2014:212).
45 Methods in the Archaeolog ical Literature Shellfish management has been investigated by researchers on the Northwest Coast (Cannon and Burchell 2009; Grier 2014; Lepofsky et al. 2015; Whittaker 2008) and Eurasia (Rakov and Brodianski 2007; Rakov and Brodianksi 2010), and methods ha ve devised for interpreting management in the archaeological record that are comparable to present day management techniques (Table 2 2). Similar methods will also be used to determine oyster mariculture at Shell Mound (see Chapter 4). Some methods of disc erning mariculture include measurement of shells to determine size/age selection (Cannon and Burchell 2009; Whittaker 2008) and selective harvest location (Whittaker 2008), and examination of parasitism and attachment scars to determine pest control and of f bottom cultivation (Rakov and Brodianski 2007, 2010). Table 2 1 Methods for determining mariculture in the archaeological record. TYPE OF CULTURING ARTICLE AUTHOR(S), YEAR SHELLFISH ARCHAEOLOGICAL METHODS SIMILAR PRESENT DAY TECHNIQUES Gardens Lepofs ky et al. 2015 Clams Evidence of beach clearing and ancient clam gardens still in place Specific areas leased to growers, habitat construction Size/Age Selection Cannon and Burchell 2009 Clams Consistency in size/ages of shellfish; low frequency of juven ile shellfish in a context Legal size restrictions (3 inches) Selective Harvest Location Whittaker 2008 Mussels Consistency in size/ages of shellfish; low frequency of juvenile shellfish in a context Closure of beds/reefs to commercial harves ting; legal size restrictions (3 inches) Rakov and Brodianski 2007, 2010 Pacific Oysters Low frequency of biofouling or parasitic attack Grow houses Off bottom Cultivation Rakov and Brodianski 2007, 2010 Pacific Oysters morphological feat ures include smooth shells lacking in radial ribbing caused by sunlight, unscalloped edges, and small attachment scars Off bottom cultivation
46 Discussion how much humans impa ct the world around us. Today we are seeing this world wide and on a huge scale, with effects such as global warming, but people have been significantly impacting the environment at various scales since the beginning of human history. Scholars such as John Erlandson and Torben Rick (2008), along with several of their colleagues, are investigating the archaeological residues of human impacts, specifically on marine ecosystems, in deep history. They, along with other scholars in the field, have seen how drama tically people can alter the environment by acts such as overharvesting marine resources and disrupting the food web. avey et al. 2011). In the shore reefs have collapsed, and in peoples are left with a fraction of what were once prolifi c bars and reefs. Many coastal livelihood have had to find ways to manipulate the resource to sustain their livelihood by implementing maricultural practices. It is necessa ry to try to rebound oyster populations, not only as a food resource, but because they are a keystone species that provide many essential ecological functions such as habitat for numerous other organisms, shoreline protection, and increasing water quality through filter feeding. The primary factors affecting oyster populations are temperature, salinity, substrate, and location in the water column, which are variable within and between or oysters are
47 subtidal environments with salinities between 14 28 ppt, temperatures close to 15 C, in areas with muddy sand bottoms. Although conditions today can be close to ideal, such as in Apalachicola Bay, during times of environmental change and he avy exploitation, humans must intervene in natural conditions using maricultural practices such as relaying, culling, shelling, size/age selection, selective harvest location, and off bottom techniques to sustain the resource. In pre Columbian history, coa stal people may have struggled with the same challenges of oyster depletion due to environmental and anthropogenic influences. Intensive and sustained oyster harvesting, which may have been done in quantities even greater than that of today, is evidenced b y huge oyster mounds and numerous middens (Sassaman et al. 2011). Furthermore, based on limited paleoclimatic data, it appears that the ecological history of the Lower Suwannee is one of significant change, which would have impacted marine resources to a l arge degree (Sassaman and Wallis 2015). Environmental conditions in the some of the earliest years of occupation were possibly ideal for estuarine production. Archaeological proxies for environmental conditions indicate that there were high salinities and warm temperatures which allowed for high species diversity and conditions under which oysters naturally flourish (McFadden and Palmiotto 2013). If these were in fact the early ecological conditions, oysters would have been large and deeply cupped with rapi d growth and reproduction, especially when coupled with the lack of previous significant human predation pressure. Following this period, from ca 3200 2500 BP, environmental conditions became far from optimal for oysters. Significant changes in temperature salinity, and sea level may
48 have killed off the previously flourishing natural oyster populations. If oysters did rebound, it would be in small populations of small, round oysters. At 2500 BP, conditions in the Lower Suwannee changed again, although this time to the benefit of oysters, with warmer temperatures, higher salinity, and relative stability (Wright et al. 2005). At around 200 300 BP, a pulse in sea level rise drove coastal communities from vulnerable shorelines, possibly aggregating at more prot ected civic ceremonial centers such as Garden Patch or Crystal River (Goodbred et al. 1998; McFadden 2015; Sassaman and Wallis 2015). A period of cooling at about A.D. 600 roughly coincided with abandonment of these centers, as people returned to previousl y occupied sites on the coast, creating centers of ritualized activity and terraforming such as at Shell Mound. Upon return to previously abandoned sites, subtidal oyster reefs, which had remained more or less unexploited for the last few centuries, in the Lower Suwannee were likely prolific as salinity was likely high and temperatures were cool. Although conditions were relatively good for oyster populations upon reoccupation of Shell Mound, regular harvest, especially when punctuated by periods of intens ification, would likely quickly negatively impact natural oyster reefs in the immediate vicinity. Maricultural techniques may have been used to offset loss or to intensify production. It is archaeologically and ethnographically evident in many other coasta l habitation sites around the world that maricultural practices were used to sustain marine resources for present and future use, especially during times of heavy exploitation (Brown and Brown 2009; Lepofsky et al. 2015). Archaeological evidence of early s hellfish maricultural techniques include habitat creation, size selection, and selective harvesting on the Northwest Coast (Lepofsky and Caldwell 2013; Lepofsky et
49 al. 2015; Whittaker 2008), off bottom hanging techniques in Ancient Rome (Gunthier 1897), an d size selection, transplanting, and culling in the Neolithic Far East (Rakov and Brodianski 2007). Not only does the archaeological evidence provide insight into subsistence practices, but also into larger issues of complexity and social organization (Cos ta Pierce 1987; Grier 2014; Lepofsky and Caldwell 2015). Shellfish in the Lower Suwannee have been significant in one way or another since human occupation at least 4,500 years ago; gastropod and bivalves were consumed in large quantities and their shells were used as building material, tools, ornamentation, and part of funeral practices as is indicated by the archaeological record coastal archaeological sites whether in ex tensive deep middens, small pits, or mounded into monumental architecture. The consumption and use of shell through time in the Lower Suwannee has fluctuated not only with changing environmental conditions, but also with changing cultural traditions, which is the subject of the following chapter.
50 CHAPTER 3 ARCHAEOLOGY AND CULTURE HISTORY AT SHELL MOUND Introduction Through the aegis of the Lower Suwannee Archaeological Survey (LSAS), members of the Laboratory of Southeastern Archaeology (LSA) and associa tes at the Florida Museum of Natural History (FLMNH) have been refining the culture history (1949) and Jerald Milanich (1994). Based on radiocarbon assays from multiple sit es in the Lower Suwannee, it has become evident that previous typologies are not necessarily as time sensitive as they may have thought, nor can they be extrapolated across large areas or regions because many of the processes that led to material cultural changes were time transgressive. Seventy nine new radiocarbon assays for sites tested by the LSAS spanning 3,900 years, from 2600 B.C. to A.D. 1300, enable us to transition from talking about the culture history of the area in broad periods, to discussing it on a century scale (Sassaman et al. 2015:175). After providing an overview of archaeological investigation in the Lower D. 200 700), incorporating general cultural traditions and transitions, specific events at the site itself, as well as major environmental changes. The chapter will conclude with a detailed discussion of excavations at Shell Mound and the associated burial complex, Palmetto Mound on Hog Island.
51 Overview of the Archaeology of the Lower Suwannee The 47 km coastline of the Lower Suwannee study area has at least 111 known archaeological sites. Research by members of the Laboratory of Southeastern Archaeology (LSA) in the Lower Suwannee region began in 2009 with the initiation of the Lower Suwannee Archaeological Survey (LSAS). The LSAS is a long term partnership between the LSA and the U.S. Fish and Wildlife Service to inventory and assess archaeological resou rces in its Lower Suwannee and Cedar Keys National Wildlife Refuges, as well as private and state inholdings within the study area (Sassaman et al. 2011). Through the LSAS, members of the LSA and FLMNH are committed to a sustained program of research, reco nnaissance, and rescue, efforts that are designed to preserve and document important cultural heritage. Furthermore, through this research contributions are being made to the understanding of environmental change and sea level fluctuation in the area, issu es that are as important to coastal communities today as they likely were to coastal dwellers of the past. Since 2009, members of the LSA and its affiliates have conducted various levels of archaeological investigation at 25 sites in the Lower Suwannee ar ea. For the purposes of this thesis, details of the history of archaeological investigation will be somewhat limited to Shell Mound as well as the neighboring mortuary complex, Palmetto Mound (8LV2) on Hog Island, as the proximity of these two sites sugges ts they were likely related to each other in terms of ceremonialism or ritual practices, although the specifics of their cultural relationship is yet to be defined. Before the initiation of the LSAS, the Lower Suwannee as part of the greater Gulf Coastal r egion has been of interest to antiquarians, amateur archaeologists, as well as professional archaeologists due to the long history of occupation and accessibility of
52 Gu lf Coast have been described and reported (e.g. Kohler 1975; Kohler and Johnson 1986; Milanich 1994; Moore 1902; Willey 1949) the archaeology of the Lower Suwannee was comparatively under researched and underreported. Results of the LSAS to date show that the archaeological potential of the area is substantial (Sassaman et al. 2016). It is clear that the study area was most pervasively settled during the Early and Middle Woodland Periods (ca. 500 B.C. to A.D. 750), with no evidence for sites predating 5,00 0 years ago due to transgressive seas. Archaeological findings and interpretations relevant to Shell Mound are summarized below; it is important to note that the inferences made are the best interpretation based on the data available, and are continually b eing refined and reinterpreted by members of the LSAS based on the addition of new data from the study area. Culture History The record of practically unbroken occupation of the Lower Suwannee starts at about A.D. 200, the same time that major cultural an d environmental shifts ensued region wide. Specifically, a pulse in sea level documented in nearby Waccasassa Bay from A.D. 200 300 (Goodbred et al. 1998) which coincided with the start of the cultural tradition of terraforming, or manipulation of the land scape involving fixed infrastructure such as massive shell mounds and ridges. Two nearby coastal sites, Garden Patch and Crystal River, were established as major civic ceremonial centers, and A.D. 200 also marks the beginning of the occupation at Shell Mou nd which emerged as another civic ceremonial center a few centuries later (A.D. 550). Milanich (1994 ) and Milanich et al. (1984) point to the interior sites of McKeithen and Lake Jackson as civic ceremonial centers exemplifying high levels of social
53 compl exity unique to the interior. Not only do excavations at village and mound complexes on the coast, such as Garden Patch, Crystal River, and Shell Mound, show similar complexity, but there is an inherent sampling bias against coastal sites due to the fact t hat they have often been leveled for urban development (i.e. Cedar Key), and other sites may be submerged due to transgressive seas along a low gradient coastline. Woodland s ites are lacking; excavation has shown that the multiple mound complexes, or civic ceremonial centers, on the coast were home to year round occupants as well as guests from the larger region, elaborate mortuary practices, feasting events, and ritual infras tructure (Sassaman and Wallis 2015). For example, Garden Patch was a coastal multi mound center with accompanying circular village plaza where people aggregated for ritual and ceremonial purposes (Wallis et al. 2015). Not only does this site exemplify inte shows further signs of complexity as defined by Milanich (1994 ). Specifically, Garden Patch is a large village site comprised of a U shaped midden and cleared plaza, as well as six mounds, including multiple burial mounds and a platform mound (Wallis et al. 2015). Furthermore, not only is Garden Patch roughly contemporaneous with complex interior sites such as McKeithen, the history of the site in terms of the site plan, popu lation aggregation, monumental construction, and mortuary ritual is comparable to that of McKeithen (Wallis et al. 2015:514). In short, both coastal and interior sites were the home of large civic ceremonial centers and were joined together through regiona l interaction.
54 The establishment of Shell Mound as a civic ceremonial center at about A.D. 550 coincides with the decrease of mound building activity at both Garden Patch and Crystal River. Shell Mound was first occupied in A.D. 200 and functioned as a pla ce of low elevation encampments. During the next century, people shifted their settlements to the remnant dune ridge, possibly to ameliorate the effects of the rising sea. The use of the landscape for small scale settlements transitioned into the construct ion of a U shaped ridge, 180 x 170 m in plan and 7 m tall. The mounded shell accumulated over a 300 year period (A.D. 400 700), with shell accumulating rapidly between A.D. 500 600 (Sassaman et al. 2016). Increased demand on resources due to a ritualized e conomy at Shell Mound resulted in intensification of the maritime economy. Evidence for intensification not only comes from the rapid accumulation of oyster shell, but also massive pits (some 2 m wide and 2 m deep), increased harvesting of big fish and mul let, and large cooking and serving vessels (Sassaman et al. 2015). Shell Mound, Garden Patch, and Crystal River were all abandoned between 600 and 750. Around this time, Hughes Mound was filled with extralocal persons and vessels that have indications of b eing related to civic ceremonial centers such as Kolomoki and Block Sterns (Sassaman and Wallis 2015). Also, after Shell Mound was abandoned as a place of habitation, people apparently returned to the area, continuing to add shell, likely from extant midde ns, to the structure at the same time they were depositing pottery and, presumably, burials at Palmetto Mound. Connections between the coast and interior, reflected in burial ritual, continued to intensify in the proceeding centuries (Sassaman and Wallis 2 015).
55 Archaeology at Shell Mound (8LV42) Shell Mound is the tallest anthropogenic feature in the Lower Suwannee study area and the largest, most formalized, and best preserved arcuate shell ridge in the center of nearly two dozen known archaeological site s in the Shell Mound Tract. The mound was constructed by emplacing massive amounts of mostly oyster shell on the distal end of the relict arm of a parabolic dune between about A.D. 500 and 650 (Sassaman et al. 2013:67; Sassaman et al. 2015:5). The structur e is more or less intact, aside from instances of looting, shell mining, road construction, and erosion. The first systematic below ground testing of Shell Mound was done on the summit of the ridge by Dolan in 1959 and reported on by Bullen and Dolan in 1 960. No further testing has been reported until the 2012 excavation of the site by staff and volunteers of the LSAS (Sassaman et al. 2013). Since initial excavation by the LSA, fourteen test units have been excavated in six areas of the site, including exc avation on the outside perimeter of the ridge, the central open area, the apex of the ridge, and the interior of the slope (Figure 3 2). Based on the results of widespread testing at the site, four phases of site use over five centuries (A.D. 200 700) have been identified (Sassaman et al. 2015:5). Early Excavation Early excavation of Shell Mound by Dolan (Bullen and Dolen 1960) and, recently, the LSA, shows deeply stratified deposits spanning five centuries of site use (Sassaman et al. 2013; Sassaman et al 2015). The inaugural systematic testing by Dolan was a single 10 x 10 foot (~3 x 3 meter) test unit placed near the apex of the mound. Based on the recovered pottery sherds, initial testing suggested that the upper two meters of the mound likely formed n o earlier than 2,000 years ago. Within the well stratified sequence
56 of organic soil matrix that was unearthed was abundant oyster shell, vertebrate fauna, and shell tools. The maximum depth of the test unit did not exceed ~3 meters. Recent Excavation Shel l Mound was excavated by the LSA with the goals of obtaining radio carbon dates and exposing the stratified sequence of construction and habitation at the mound (Sassaman et al. 2013). The overarching goals of excavation at the site include understanding v ariations in occupational sequence and site use across the mound area, and have since been expanded by the research questions of graduate students in the LSA. In 2012 a team of archaeologists from the LSA excavated two 1 x 2 m test units (TU1 and TU2). Te sting also encompassed bucket auguring in the interior of the mound as well as the excavation of one 1 x 1 m unit (TU 3) in the interior opening. Results of this initial investigation indicated that the mound accumulated over at least two centuries (~1500 1300 cal. B.P. or ~cal. A.D. 450 650), and that the arcuate shape of the mound is original, not the result of shell mining, as previously thought. The LSA continued excavation at Shell Mound in 2013 with three 2 x 2 m te st units, Test Unit 4 (TU4) and Test Unit 5 (TU5), in the western portion of the interior of the mound, and Test Unit 6 (TU6) on the outside perimeter (Figure 3 1). TU4 and TU5 were placed in an effort to locate evidence of residential areas, as it is assumed that houses were arranged in a s emi circular fashion surrounding the plaza. In 2014, the first archaeological field school was held at Shell Mound, excavating three test units, Test Unit 7 (TU7), Test Unit 8 (TU8) and Test Unit 9 (TU9). To further the goal of defining site use, these uni ts were placed in three separate areas of the
57 mound: the interior slope (TU7), the apex of the mound (TU8), and the outside perimeter (TU9). Figure 3 1 LiDAR topographic map of Shell Mound showing excavation units (from Sassaman et al. 2015). In 2015 the LSA held the second archaeological field school at Shell Mound, opening five units at various parts of the site, focusing on the interior slope. Three units, Test Units 10, 13, and 14 (TU10, TU13, and TU14), were located adjacent to TU7, Test
58 Unit 12 (TU12) was located in the central area of the interior slope in the west, and Test Unit 11 (TU11) was located on the interior slope of the southern arm of the ridge. Each of these three areas revealed very different composition, with large shell free pit s to the north (TU10, TU13, and TU14), dense oyster midden and pit features to the west (TU12), and deeply stratified deposits (~2 m) of bedded clean whole oyster shell to the south (TU11). Excavation on the Outside Perimeter of the Ridge Four test units have been excavated on the outside perimeter of the ridge: TU1, TU2, TU6, and TU9. Sherds of the Pasco, Swift Creek, and Deptford traditions were recovered in TUs 1 and 9. These same two units also showed evidence of a large scale burning of the area, whi ch occurred a few centuries before the emplacement of the dense shell deposits comprising the mound (Sassaman et al. 2015:26). Charred hickory nut shell from this burned stratum in TU9 was estimated by AMS assays to date from ca. A.D. 180 340. TU6, on the western outer perimeter, expressed a lower artifact density and lacked the charcoal rich stratum seen in the other test units on the outside perimeter, likely due to its marginal proximity to the shell ridge (Sassaman et al. 2015:21). Based on the dates a nd pottery sequences from the outside perimeter of the mound, it seems as though this was the location of the initial phase of Woodland era settlement, ca. A.D. 200 400. Excavation on the Apex of the Ridge Excavation on the apex of the ridge has revealed t hat the basal strata of TU8 is likely indicative of the second major phase of occupation at Shell Mound, ca. A.D. 400 550. The third phase of occupation also expresses itself in TU8, with the emplacement of mass amounts of shell between A.D. 550 and 650. T U8 was situated near the highest
59 point of the mound in order to access some of the deepest stratigraphy at the site. The unit was situated in one of several circular depressions, about 10 m in diameter, which was hypothesized to be an accumulation of shell around a domestic structure. The unit was placed on the southeast portion of the depression, to examine the flatter area where it met a steep incline (Figure 3 2 ). The stratigraphy of TU8 consists of three he top down: a 1.6 m thick mantle of bedded oyster shell dating to the sixth century A.D., organically enriched sands with moderate to sparse shell dating to the fifth century A.D., and submidden sand into which pit features were dug and backfilled (Sassam an et. al 2015). TU8 was the deepest of the units dug by the LSA, revealing that the elevation of the northern arm of the shell ridge is both anthropogenic, with the addition of shell, and geological, with the underlying sands being part of the relict dune on which Shell Mound is constructed. TU8 will be discussed further in the next chapter, as it is the location from which samples of oysters used in this study were obtained.
60 Figure 3 2 Topographic map showing the location of TU8 (Sassaman et al. 2015:38). Excavation on the Interior of the Slope TU7 was the first unit to be excavated on the interior of the slope, and was sited to locate evidence of domestic architecture related to the second phase of occupation at Shell Mound. What was unearthed was an amalgam of massive intersecting pit features with little to no shell, postholes, and exotic cultural materials. TU10, 13, and 14 were excavated adjacent to TU7 to expand upon the earlier findings. Although some of the pits were similar to those seen elsewhere on the site, several of them, in terms of
61 size and content, were unique among pits from all other excavations at Shell Mound. These large pits suggest that Shell Mound, at least at times, went beyond the realm of domestic activities, likely to i nclude large scale social gatherings involving the consumption of large quantities of food (Sassaman et al. 2015:65). Furthermore, as the pits intersected one another, this location was returned to on more than one occasion for pit digging activities (Sass aman et al. 2015:66). TU11 was excavated on the interior slope of southern arm of the ridge to compare activities at this part of the site to those to the north (TU7, 10, 13, and 14). Revealed was a deeply stratified sequence of clean, whole oyster shell placed on top of earlier highly organic sands (midden), a pattern similar to that of TU8 at the apex of the mound. It is likely that this unit also reveals the activity of the third phase of occupation at Shell Mound which included the building up of the s outhern arm of the relict dune. Flooding from heavy rains in 2015 precluded excavation below the top of the subshell midden. The final test unit excavated on the interior was TU12 located on the western portion of the interior of the mound, about equidista nt from the units to the north and south. This unit was comprised of a dense organic shell midden emplaced over large refuse pits. Excavation of the Central Open Area mound, with evidence of the fourth phase of occupation at the site which lasted about 50 years. TU4 and TU5 were placed in a checkered board fashion to expose possible continuous architectural features in the unit floors and profiles. Most of the exposed pit
62 fea tures were ambiguous, with the exception of a single posthole, but indicate that it is likely that stronger architectural evidence exists in the vicinity (Sassaman et al. 2015). Conclusion Based on the recent work of the LSAS, four phases of occupation at Shell Mound have been described by Sassaman et al. (2015:100): (1) an early phase spanning the second and third centuries A.D. During this time most activity at the site was from the occupation of the outside perimeter of a relict dune; (2) an intermediat e phase spanning the fifth and early sixth centuries A.D. During this time there was intensive occupation primarily on top of the dune; (3) a mid sixth to mid seventh century A.D. phase. During this phase the bulk of oyster shell was emplaced on the dune t o the south, creating the U shape of the mound; and (4) a final century spanning the mid seventh to mid eighth centuries A.D. During this phase, occupation was concentrated on the interior opening of the ridge with continual deposition of shell on the ridg e. Palmetto Mound on Hog Island (8LV2) Palmetto Mound (8LV2) is a heavily damaged mortuary facility on Hog Island, just to the west of Shell Mound. The site has been excavated by both amateur and professional archaeologists, as well as compromised by loot ers. The burial complex on Hog Island was started as early as 300 B.C., preceding Shell Mound as a place of importance. This 55 x 25 m sand and shell mound is located directly 500 m west of Shell Mound, and is likely a related site, especially given their proximity (Sassaman et al. 2015). The 2 m tall mound contained hundreds of burials, ceramic vessels, many of which were intact, and other artifacts, many of which have been donated to the Florida Museum of Natural History (FLMNH).
63 Figure 3 3 LiDAR top ographic map of Palmetto Mound and Shell Mound (Sassaman et al. 2015). Some of the earliest known work on the island was in the 1880s by Decatur Pittman, an amateur archaeologist. Montague Tallant, another amateur archaeologist, also dug at the site in th e 1930s, followed by archaeologist Dr. John Goggin in 1952 and 1962 (Willey 1949). This early work contributed to the large collection of artifacts now housed at the FLMNH and South Florida Museum, including over 5,000 sherds from at least 600 vessels from the mound (Dono p 2015).
64 Recent work at Palmetto Mound was conducted by the LSA as part of graduate research as well as the 2014 archaeological field school. Since 2014, members of the LSA along with field school students have mapped the topography of the mound, systemati cally tested the site with shovel test pits, and excavated a 2 x 2 m test unit on the northwestern edge of the mound. The test unit was placed in a former looter hole, in an effort to find intact deposits and stratigraphy below the looter disturbance. The and material culture including shell beads (Donop 2015). Discussion Continued archaeological projects in the Lower Suwannee are redefining the culture history of th Coastal peoples. Shell Mound is adding to data relating to coastal civic ceremonial centers by contributing to the narrative of regional interaction and ceremonialism. Proximity and orient ation of Shell Mound and Palmetto Mound indicate that they were likely related sites of gathering and ritualized activity involving terraforming, burial of the dead, and large scale consumption of food and intensification of shellfish harvesting. This infe rence is based on the massive pits found on the northern arm of the ridge which include nonlocal materials and large quantities of mammal, fish, and bird remains; the proximity of the monumental architecture at Shell Mound and the mortuary facility on Hog Island; and the large volume of oyster shell at Shell Mound and purposeful placement of oyster shell at both sites. Oyster shell is present, in varying quantities, in every test unit excavated at Shell Mound, and was used as a capping material at Palmetto Mound. It is evident from the deep, dense shell deposits excavated from the apex and southern arm of Shell Mound that oyster was harvested and
65 consumed in large quantities, perhaps surpassing that of daily consumption, and purposefully placed in such a way shape which was created in only a couple of centuries. This intensification of resource use may have led to the need for sustainable shellfish harvesting strategies in the form of mariculture due to increased daily subsistence de mand as well as the cultural traditions of ritualized consumption, gathering, and terraforming. Methods for determining evidence for chapter.
66 CHAPTER 4 SAMPLING, HYPOTHESE S AND METHODS Although some researchers have stated that it should be an accepted fact that people of the ancient past were using some form of mariculture (e.g., Caldwell and Lepofsky 2013; National Research Council 1992; Rakov and Brodianski 2010), these claims must be empirically tested. The objective of this thesis is to determine if it is possible to infer the use of mariculture during the Woodland period in the Lower Suwannee by analysis of archaeological oyster shell from Shell Mound (8LV42). The typ es of oyster maricultural practices in consideration are shelling, relaying, size/age selection, selective location of harvest, culling, and off bottom techniques. Researchers investigating marine management have developed methods for determining size/age selection (Cannon and Burchell 2009; Lepofsky and Caldwell 2013), selective harvest location (Whittaker 2008), and off bottom techniques (Rakov and Brodianski 2007). I have adapted and incorporated the methods of these researchers while also developing met hods for interpreting all six of these practices, primarily based on morphological attributes of archaeological shell. These six maricultural practices were chosen for two primary reasons: first, because of their likelihood of being practiced on the Gulf C oast 2,000 years ago; and second, because there are comparable archaeologically identified practices of shellfish management from other cultures in the past. Shell Mound was chosen as the site used to determine these practices because of the rate at which shell was accumulated (within a century or two) and the probable ritual importance of the site where shellfish, specifically oysters, were harvested for more than daily subsistence activities.
67 Proxies for these maricultural practices are drawn from multipl e morphological attributes apparent on whole, left oyster valves from relatively temporally continuous bulk samples spanning the Woodland period occupation of Shell Mound (A.D. 200 700). The aims of my research are twofold: first, to use and create methods to determine what culturing practices would look like on oyster shells recovered from archaeological contexts; and second, to apply those methods to determine if maricultural practices were used by inhabitants of Shell Mound. Patterned variation in sampl es from a stratified midden is inherently diachronic and thus have the potential to reveal changes in oysters attending maricultural practices. There are three periods that are of particular interest: early occupation (~A.D. 200 550), mound building or int ensification (A.D. 550 650), and abandonment as residential center (A.D. 650 700). Similarities and differences in patterns between these periods at Shell Mound provide meaningful diachronic data related to the processes of shellfish harvesting and managem ent. It must be acknowledged that there are limitations to the interpretations, most notably discerning environmental processes affecting shell morphology versus human intervention affecting shell morphology (e.g., Claassen 1998; Erlandson and Rick 2008; Swadling 1976). Measurements of attributes used to infer mariculture are meaningful when there is patterning observed in the context of broader social processes and site formation. For example, it would follow that during early occupation (A.D. 200 550) ma riculture would not have been a necessary practice as populations would be low and the primary reason for oyster harvesting would be subsistence. In contrast, maricultural practices may have been necessary during times of intensification (A.D. 550 650)
68 ass 2002), involving population increase, the construction of monumental architecture, and possible coalescent and/or feasting events. Finally, evidence of mariculture would dissi pate once the site was abandoned as a place of residence (A.D. 650 700); although oyster shell was still mobilized at this time for mound construction, people would not be at the site to tend to local reefs. Furthermore, based on radiocarbon dating, it app ears that the shell used in the final depositional events were taken from nearby middens and re deposited on the underlying strata of unconsolidated shell. Archaeological evidence of culturing and management processes during the same time period and social context elsewhere on the Northwest Coast provide comparable material to inferring mariculture at Shell Mound (i.e. Lepofsky and Caldwell 2013; Lepofsky et al. 2015; Whittaker 2008). Premises This research is predicated on four premises: Premise 1: Oyster shells used in the construction of Shell Mound were shell are assumed to be a function of local environmental circumstances as well as human procurement and interven tion within a limited area of resource niches that would be impacted by regular and intensified procurement strategies. Premise 2: Oyster shell was harvested by the inhabitants of Shell Mound. It is assumed that the people living at Shell Mound were the on es who procured oysters from resource niches within the vicinity of the site, and oysters were not brought to the mound and deposited from elsewhere in the region.
69 Premise 3: Shell Mound was a place of ritual importance and symbolic significance likely r elated to the burial complex at Palmetto Mound. It is assumed that resource intensification was a function of an increased demand on oyster resources due to ritual activity likely involving feasting or ritual consumption, mound building activity, and incre ased sedentism. Premise 4: Shell Mound was transformed from a place of quotidian habitation to a civic ceremonial center with a purposefully constructed ridge of shell forming a U shape. It is assumed that the shell was harvested with the intention, beyond consumption, of being placed within a pre determined monumental structure. Hypotheses and Implications Hypotheses and test implications are listed below for each of the procurement and maricultrual strategies that may have been used by the residents of S hell Mound. Hypothesis 1 (H1): The location of collection (i.e., subtidal versus intertidal) will covary with the scale and intensity of oyster procurement. Hypothesis 2 (H2): There will be no change in the location of harvest in accordance with changes i n the scale and intensity of oyster procurement. Implications: If the scale and intensity of oyster harvesting increased for daily subsistence as well as feasting events and mound building activities, then subtidal oysters would be preferred due to their larger size and generally better quality. If H1 is true, then the majority of oysters in samples associated with Shell Mound as a civic ceremonial center would be subtidal, whereas those associated with daily subsistence activities would be intertidal due to the ease of access of intertidal oysters. If H2 is true, then there would be no difference in the percentage of subtidal or intertidal oysters between occupational phases.
70 Hypothesis 3 (H3): The amount of right oyster valves used for cultch (shell retu rned to the water to enhance oyster reefs) will covary with the scale and intensity of oyster procurement. Hypothesis 4 (H4): There will be no change in the ratio of left to right oyster shells in accordance with changes in the scale and intensity of proc urement. Implications: If right valves were used as cultch for the maricultural practice of shelling, then there would be a disproportionate number of left and right valves, with less right valves in a sample associated with an increase in the scale and i ntensity of oyster procurement. In the first phase of occupation, where mariculture was not likely to be practiced, there would be an equal number of left and right valves, as no shell would be returned to the water to enhance oyster reefs. If H3 is true, then the ratio of left to right valves would be high. If H4 is true, then there would be an equal ratio of left to right valves. Hypothesis 5 (H5): The amount of oysters that were transplanted from suboptimum conditions to optimal conditions through relay ing will covary with the scale and intensity of oyster procurement. Hypothesis 6 (H6): There will be no indication that oysters were transplanted in accordance with the scale and intensity of oyster procurement. Implications: If oysters were being transp lanted from one resource patch to another environmentally distinct resource patch in order to increase the growth rate or quality of the oysters, then there would likely be indications of multiple environmental sources on the shell. If H5 is true, then the re would be a large number of shells with
71 evidence of relaying. If H6 is true, then there would be no evidence of this practice, or the same amount of evidence of this practice between occupational phases. Hypothesis 7 (H7): Evidence of oyster culling wil l covary with the scale and intensity of oyster procurement. Hypothesis 8 (H8): There will be no indication that oysters were culled in accordance with the scale and intensity of oyster procurement. Implications: If the maricultural practice of culling w as being used, and live oysters were being returned to the water to maintain oyster populations, there will be evidence of culling on the oyster shells during phases of occupation associated with large scale intensive harvesting. If H7 is true, then the am ount of oysters being culled will be high during the mound building phase of occupation. If H8 is true, then there will be no evidence of culling, or the evidence of culling would be equal between the phases of occupation. Hypothesis 9 (H9): The size rang e of oysters will covary with the scale and intensity of oyster procurement. Hypothesis 10 (H10): There will be no change in the size range of oysters in accordance with the scale and intensity of oyster procurement. Implications: If oysters were being s elected based on size/age in order to maintain and stabilize oyster populations then evidence of that would be apparent in the height of oysters from a sample. If H9 is correct, then during the phases of occupation in which oysters were harvested quickly a nd intensively, the range of the size of oysters would be very small. If H10 is correct, then the range of the oyster sizes would not change between different occupational phases.
72 Hypothesis 11 (H11): The amount of oysters grown using off bottom technique s will covary with the scale and intensity of oyster procurement. Hypothesis 12 (H12): There will be no evidence of off bottom techniques being used in accordance with the scale and intensity of oyster procurement. Implications: Oysters grown using off b ottom techniques will show evidence of this in the form of their attachment scar. If H11 is true, then there will be a large number of oysters in samples from the intensive phases of occupation that show evidence of being grown using off bottom techniques. If H12 is true, then there will be no variation of oysters that show evidence of being grown off the bottom at all, or between occupational phases. Sampling A total of 3,252 left oyster valves were analyzed from a continuous 30 x 30 cm column sample take n from TU8, near the apex of Shell Mound. Twenty bulk samples were collected encompassing the entire depth of the unit, 2.1 m below surface. Based on the stratigraphy, TU8 was separated into three macrostratagraphic units, or 2015:40 47): the first, uppermost macrounit (Strata I III) is comprised of unconsolidated bedded oyster shell; the second macrounit (Strata IV V) consists of organically enriched sands with moderate to sparse shell; the third, deepest macrounit (Strata VI VII) is primarily submidden sand, into which pit features were dug and subsequently backfilled. The first two macrounits (Strata I V) are anthropogenic, whereas the last microunit is natural dune sand, with the exception of the pit features. The upper 1.4 m of unconsolidated shell was excavated in column bulk samples taken from the east profile, and the samples from the ~80 cm of subshell midden were taken from the west profile. Bulk samples were numbered 1 20 from the top of the unit
73 at surface, increasin g in numerical sequence to the bottom of the unit. Samples were taken in 10 cm increments. Bulk samples 1 13 were taken from the upper macrounit (Strata I III) of unconsolidated shell, and bulk samples 14 20 were taken from the second macrounit (Strata IV V ) of sand and sparse shell. All matrix from these bulk samples were returned to the LSA in Gainesville, Florida, and processed with a Dausman Flote Tech flotation machine and then fractionated for secondary analysis. All oyster shell was separated from t he rest of the material in the bulk samples and then sorted into whole left valves, whole right valves, and fragments. Whole right valves were counted and weighed and the fragments were only weighed. Whole left shells were also counted and weighed, and the n set aside for further analysis, described below. Figure 4 1. Drawing and photograph of TU8 east profile (Sassaman et al. 2015)
74 Figure 4 2. Drawing and photograph of TU8 east profile (Sassaman et al. 2015: 41).
75 Methods and Techniques The methods us ed for inferring mariculture and harvesting strategies from the archaeological record at Shell Mound have been drawn from archaeological literature on ancient mariculture/management techniques, the biological and ecological literature on oyster biology and ecology, as well as first hand experiences in the field with oystermen and oyster biologists in Apalachicola Bay and the Suwannee Sound (see Chapter 2). Described below are the methods and associated hypotheses. Harvesting Location In a study done in Mar yland, Brett Kent (1988) classifies oysters based on harvesting location. In order to determine this, Kent (1988) used the height to length ratio (HLR). The height to length ratio is determined by dividing the height, or the measurement of the dorsal end t o the ventral end, and the length, or the measurement from the posterior end to the anterior end. Shell height, length, and HLR were used to determine the environment of har vest for the shells from Shell Mound. Maryland and the greater Chesapeake area, in Florida and the southeast, the best way to classify the environment from which oysters were harvested would be to differentiate between intertidal oysters and subtidal oysters (Lawrence 1988). Oysters harvested from the intertidal will show evidence of extreme clustering with large left valve attachment scars indicative of attachment to other oysters, hav e thinner shells, and be elongate (Lawrence 1988:267). Oysters harvested from subtidal conditions are more likely to grow singly, therefore not always having attachment scars or having small attachment scars, have more ovate or subovate shell outlines, hav e thicker shells, and
76 increased valve cupping due to less crowding (Lawrence 1988:267). Subtidal oysters are also often larger than intertidal oysters, as they are constantly exposed to food, whereas intertidal oysters cannot feed during low tide. Furtherm ore, as many organisms associated with oysters in a fouling community, including predators and parasites, are not adapted to living in intertidal conditions, the frequency of biofoul living upon or within oyster shells will be higher on oysters harvested f rom subtidal environments. Based on these distinctions, the attributes I recorded from the oyster shells include the height, HLR, presence or absence of attachment scars, presence or absence of biofoul, presence or absence of parasitism, and left valve co ncavity. Shell height and length were both measured to the hundredth of a millimeter (two places after the decimal) using electronic calipers. The height was taken at the longest measurement from the dorsal to the ventral, and length was measured at the w idest point of the shell from the anterior to the posterior. All measurements were recorded and then the height of the shell was divided by the length of the shell to produce the HLR, rounded to the nearest hundredth of a millimeter. Presence or absence of attachment scars, biofoul, and parasitism were recorded for each shell. The type of biofoul and/or parasite was specified for each shell in which they were present. Each shell was also give a left valve concavity value between one and three, one being alm ost flat, two being average, and three being extremely cocave. The left valve concavity value was determined based on the ratio of depth to the height and length of the valve. Inferring whether oysters were harvested from intertidal or subtidal environment s is strongest when these attributes appear in concert, as variations do naturally occur in
77 dynamic estuarine environments (Lawren ce 1988). Summarized in Table 4 1 below are the attributes associated with environmental source areas from which oysters were harvested. Table 4 1. Inferences regarding environment of harvest and associated attributes evident on archaeological shell (adapted from Lawrence 1988:268). Inference Height HLR Presence/Absence of Biofoul Presence/Absence of Parasitism Concavity Sour ce area Intertidal Smaller on average Elongate Biofoul that cannot live in intertidal environments absent Evidence of parasitism from organisms that cannot live in intertidal environments absent Flatter, less c oncave shells value mostly 1 and 2 Source a rea Subtidal Larger on average Ovate or subovate Presence of biofoul that can live only subtidally Evidence of parasitism from organisms that can live only subtidally Shells that are very concave value mostly 3 with some 2 The differences between intert idal or subtidal environments provide insight into and can be harvested easily and locally at low tide. Conversely, harvesting subtidal oysters requires a larger expen diture of energy to collect the resource from reefs farther out in the water that would likely require a boat to access. In terms of how this differentiation relates to mariculture lies in the quality and productivity of the oysters. Subtidal oysters ofte n produce higher quality meat (plump, regular exposure to food (Korringa 1952). Also, subtidal oysters are often more prolific.
78 Therefore, if oysters were managed in order to produce large quantities of high quality oysters, the oysters in a cultured deposit would likely be primarily subtidal. Shelling Shelling enhances oyster reef productivity by building habitat through returning shell to the water for oyster lar vae to settle on. In order to discern this archaeologically, it would be best to test extant reefs for evidence of this practice. To discern this practice simply by looking at the shells in a sample, the ratio of left (cupped) to right (flat) valves could be used as a possible measure. In terms of ecology, oyster larvae settle most readily on smooth flat surfaces of their own species (Crisp 1967) and, since right valves are flatter than left valves, they would be ideal cultch. If shelling was being practice d, it is likely that there would be more left valves in a sample than right. Also, if oysters were being processed on the water, right valves would be shucked and then immediately returned to the water, as the oyster meat sits in the cupped left valve. Als o, if there are many dead shells, those with heavy parasitic encrustation or with biofoul present on the inside of the shell, it may be indicative that they were collected and deposited in the sample because they were attached to desired oysters that were collected for consumption (Lawrence 1988:267). Relaying The best way to determine relaying in the archaeological record would likely be through the use of isotopic analysis. In order to discern relaying in the absence of isotopic data, morphological evide nce of two distinct environments would be present on the shell if oysters were transferred from one set of environmental conditions to another. For example, if a high fraction of oysters in a sample are elongate with HLRs close to 2 mm, but also have evide nce of biofoul or parasites that are only associated with subtidal
79 taken from the intertidal, which is a suboptimum environment, and transferred to a subtidal env ironment, which provide oysters with more resources for fast growth and reproduction. Another practice of relaying may be to take oysters from one salinity regime to another, based on the desired results, in which high salinity biofoul, such as barnacles, are more likely to be present. The relevant attributes to be used to infer relaying would be height, HLR, presence/absence and type of parasites, presence/absence and type of biofoul, and left valve concavity These attributes are used to indicate salinity regimes as well as intertidal or subtidal conditions, as described above. Inferring relaying involves a multivariate approach, where the relevant attributes vary in a way contradictory than what would be expected if the oysters only grew and lived in one type of environment. Off Bottom Techniques Off bottom techniques include growing oysters on lines or poles to protect them from sedimentation which may be harmful for the growth and reproduction of oysters. Oysters are xenomorphic, meaning that they faith fully replicate the substrate on which they grow (Lawrence 1988). Because of this, if attachment scars are shaped in a way that is indicative of cordage, rope, or twigs and branches, it would indicate that they were not naturally settling on other shells. Aside from protection from sedimentation, growing oysters off the bottom allows for placement in optimum and controlled areas of the estuary. The type or shape of attachment scar would be the attribute used to distinguish this type of culturing technique.
80 Culling Culling is visible on archaeological shell by examining attachment scars, biofoul, and parasitism. If an oyster burr is culled, then the desired oysters would be kept for consumption and the dead shell and smaller oysters and spat are returned to the water to continue growing. If an oyster is returned to the water after being removed from other oysters in a burr, then the oyster would have an attachment scar. Once returned to the water, the attachment scar becomes more vulnerable to parasitic attac k and set tling of biofoul, whereas it would be protected from these organisms when attached to other oyster shells. If an oyster has an attachment scar that has evidence of parasitism or biofoul, then it could be inferred it was returned to the water after being separated from other oysters and substrate (Figure 4 3) Figure 4 3. Photograph of two left valves showing sponge parasitism on attachment scars.
81 Size/Age Selection Size and age selection as culturing strategies largely relate ecologically to the reproduction of oysters. If coastal inhabitants were selectively harvesting for size or age, larger and older oysters were harvested first in order to give smaller juvenile oysters time to mature through reproductive cycles. If large oysters were routi nely collected it could be selectively harvesting based on resource patch (Whittaker 2008). To determine this, both height and age can be determined from archaeolog ical oyster shell. Height can be used as a proxy for age, where the larger oysters are thought to be older and the younger oysters are thought to be smaller, but this is not always the case when increased predation pressure affects the biology and morpholo gy of oysters (Swadling 1976). Discussion Using shell from archaeological contexts as the primary means to infer mariculture requires the close examination of morphological attributes of shell that serve as proxies for various practices or environmental c onditions. An understanding of the natural conditions, including the dynamic nature of estuaries in which oysters grow, the variation of estuaries in a region, and the biological and environmental factors that affect oyster growth and reproduction is essen tial in interpretation. Those natural factors, when combined with the equally dynamic and variable social processes of the coastal populations in question, can provide insight into if, how, and why maricultural practices were in place. In a highly dynamic estuarine environment oyster populations are vulnerable. Although oysters are conditioned to be resilient and prolific, able to withstand a broad
82 range of salinity, temperature, turbidity, and dissolved oxygen, as well as predation pressure, sudden and dr amatic shifts in salinity or temperature for prolonged periods of time make oyster reefs and beds susceptible to collapse and morphological change (Rick and Lockwood 2013). Because of the possibility of shifting morphology due to environmental shifts and/ or storm events, the methods used in determining mariculture also lie in the context of procurement, in this case, ritual consumption, intensification, and/or increased demand for subsiste nce or architectural resources. These methods were devised based on management in the archaeological record, the ecological literature concerning oyster morphology and ecology, as well as experiences in both Apalachicola Bay and the Suwannee Sound with oystermen engagi ng in mariculture. Specifically, the observation of oyster harvesting and culturing practices in Apalachicola Bay influenced the methods devised for determining culling and the technology involved in harvesting subtidal oysters. The multiple variables use d to in these methods were recorded for 3,252 oysters from Shell Mound and statistically analyzed, the results of which are described in the following chapter.
83 CHAPTER 5 RESULTS AND ANALYSIS Reported below are the results of the analysis of all of the wh ole left oysters from the subsistence column from Test Unit 8 near the apex of Shell Mound. Twenty bulk samples were recovered from TU8 in a subsistence column from the east and west profiles. The samples were then floated and all oyster shell was removed for further analysis. In the previous chapter I described the methods for analysis and the relevant variables. The variables used for analysis are height, length, height to length ratio (HLR), attachment scars, sponge parasitism, left valve concavity and biofouling. The number of left to right valves is also compared. Each variable is divided below by subsistence column sample number (1 20) and some are grouped by macro unit for comparison, where there is the Upper Macrounit (Samples 1 13, n = 3,098) and t he Lower Macrounit (Samples 14 20, n = 154). The Upper Macrounit was then further subdivided into two subunits, Upper Macrounit 1 (Samples 1 6, n = 1,261) and Upper Macrounit 2 (Samples 7 13, n = 1,837). The Upper Macrounit was divided into the two subunit s based on a stratigraphic break as well as breaks in patterning of fauna and material culture. Radiocarbon dates at the base of each macrounit show that there may be reverse stratigraphy, where the Upper Macrounit 1(A.D. 405 550) accumulation of shell may have been redeposited on top the Upper Macrounit 2 (A.D. 545 645) from an older midden. Also, in some cases, extremely small sample sizes in the subsistence column samples, specifically Sample 14 ( n = 3), created potentially false anomalies, therefore the comparison of macrounit may be more meaningful. Before addressing the
84 hypotheses defined in Chapter 4, the descriptive statistics are presented by variable, as multiple variables are used in conjunction to test the hypotheses. Height, Length, and Height to Length Ratio For each of the 20 samples, the mean, minimum, maximum, and standard deviation of those three variables is listed and graphed below (Tables 5 1, 5 3, and 5 4, and Figures 5 1 5 3). The range of oyster height is also presented by number of o ysters per macrounit as well as percent of oysters per macrounit wi thin 20 mm size ranges (Table 5 2). The total range of the height of the oysters sampled is 130.95 mm, ranging from a minimum of 14.96 mm (Sample 11) to a maximum of 145.91 mm (Sample 12). The mean height is the smallest in samples 15 20, being less than 50 mm on average (30.54 49.60 mm). In Samples 1 13, the oysters remain in 50 mm range, with an outlier in Sample 14, where the mean height is 64.48 mm. In Samples 7 20 the average remains b etween 55 and 59 mm (55.63 58.36 mm), dropping to between 50 and 56 mm (50.67 55.78 mm) in Samples 1 6. The total range of the length of the oysters sampled is 70.89 mm, ranging from a minimum of 6.23 (Sample 18) to a maximum of 77.12 (Sample 12). Similar to the mean height, the mean length is the smallest in Samples 15 20, not exceeding 30 mm (17.15 28.83 mm). Again, there is an increase in mean length in samples 1 13, ranging from about 30 mm to about 33 mm (30.26 33.23 mm). Like with mean height, the me an length of Sample 14 is anomalous at 36.87 mm. The mean height to length ratio ranges between 1.56 1.81. The mean remains within a range of 9 percent in sample 6 13 (1.71 1.80). In Sampl es 1 5, the mean HLR ranges 18 percent (1.56 1.74), and in Sample 14 20, the HLR ranges 31 percent (1.58 1.89).
85 Table 5 1. Descriptive statistics for oyster height by subsistence column sample. Height (mm) Sample n = Mean SD Minimum Maximum 1 37 50.67 9.61 31.19 77.34 2 184 52.10 12.17 26.67 96.62 3 315 55. 38 13.09 23.18 93.67 4 312 53.97 14.73 20.26 112.19 5 190 55.78 15.73 18.40 95.95 6 223 54.36 14.93 21.82 95.92 7 246 57.26 16.05 15.68 14.77 8 254 57.22 17.95 19.40 110.03 9 187 55.63 18.76 18.55 115.77 10 258 57.08 16.91 24.30 109.96 11 292 57.74 18.77 14.96 104.87 12 276 58.36 20.56 19.82 145.91 13 324 57.68 19.02 20.01 129.95 14 3 64.48 10.20 52.73 70.99 15 14 48.08 16.58 26.52 82.23 16 14 43.20 10.89 27.26 60.56 17 11 30.54 10.97 17.39 49.39 18 49 43.48 16.6 8 19.50 82.64 19 16 49.60 17.71 26.54 84.06 20 47 49.19 15.22 26.08 85.25 Figure 5 1. Mean and range oyster height by subsistence column sample. 0.00 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Height (mm) Subsistence Column Sample Number
86 Table 5 2. Number and percent of oysters in each macrounit by size range. Height By Size Range (mm) Range Upper Macrounit 1 Upper Macrounit 2 Lower Macrounit n = % n = % n = % 0.1 20 2 0.16 9 0.49 4 2.60 20.01 40 188 14.91 337 18.35 54 35.06 40.01 60 676 53.61 765 41.64 70 45.45 60.01 80 342 27.12 542 29.50 21 1 3.64 80.01 100 52 4.12 146 7.95 5 3.25 100.01 120 1 0.08 38 2.07 0 0.00 Total 1261 100 1837 100 154 100 Table 5 3. Descriptive statistics for oyster length by subsistence column sample. Length (mm) Sample n = Mean SD Minimum Maximum 1 37 32.39 5.88 21.43 44.12 2 184 31.83 6.81 16.74 55.87 3 315 32.73 7.92 14.19 58.08 4 312 31.08 8.01 12.71 62.30 5 190 33.23 8.75 11.76 57.14 6 223 30.26 7.64 12.33 52.08 7 246 33.56 9.33 11.16 62.95 8 254 31.63 9.82 10.09 61.65 9 187 31.14 9.58 11.60 57.35 10 258 32.06 8.17 13.10 59.14 11 292 32.75 10.11 10.44 60.69 12 276 32.94 10.72 11.37 77.12 13 324 32.00 10.00 9.66 67.91 14 3 36.87 8.77 27.07 43.97 15 14 28.83 8.21 13.71 42.01 16 14 27.32 7.40 17.16 40 .14 17 11 17.15 5.07 11.30 25.10 18 49 23.00 7.17 6.23 37.02 19 16 26.90 8.52 13.65 47.16 20 47 27.88 7.74 13.55 50.92
87 F igure 5 2. Mean and range oyster length by subsistence column sample. Table 5 4. Descriptive statistics for oyster HL R by subsistence column sample. Height to Length Ratio Sample n = Mean SD Minimum Maximum 1 37 1.56 0.17 1.16 1.87 2 184 1.64 0.20 1.17 2.23 3 315 1.69 0.24 1.20 2.50 4 312 1.74 0.28 1.19 3.21 5 190 1.68 0.29 1.08 3.00 6 223 1.80 0. 30 1.14 2.93 7 246 1.71 0.27 1.23 2.63 8 254 1.81 0.28 1.29 2.81 9 187 1.79 0.30 1.17 2.92 10 258 1.78 0.28 1.14 2.79 11 292 1.76 0.28 1.13 2.76 12 276 1.77 0.31 1.14 3.05 13 324 1.80 0.30 1.11 3.09 14 3 1.75 0.17 1.61 1.95 15 14 1.67 0.28 1.33 2.38 16 14 1.58 0.21 1.25 1.88 17 11 1.78 0.45 1.28 2.88 18 49 1.89 0.39 1.21 3.13 19 16 1.84 0.30 1.04 2.49 20 47 1.76 0.32 1.42 2.35 0.00 5.00 10.00 15.00 20.00 25.00 30.00 35.00 40.00 45.00 50.00 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Length (mm) Subsistence Column Sample Number
88 Figure 5 3. Mean and range oyster HLR by subsistence column sample. Presence/Absence and Type of Attachment Scars The presence/absence of attachment scars is shown for Samples 1 20 in the table and figure below (Table 5 5 and Figure 5 4) and then divided into the Upper Macrouni t and Lower Macro Unit (Table 5 6), as well as Upper Macrounit 1 and Upper Macrounit 2 (Table 5 7). Both the number of shells in each sample with and without attachment scars as well as the percentage of shells in each sample with and without attachment scars are listed. The lowest percentage of oyster shells with at tachment scars is 43 percent (Sample 4), and the highest percentage of oyster shell with attachment scars is 86 percent (Sample 1). With the exception of Samples 3, 4, and 19, the amount of oyster shells with attachment scars in a sample does not fall belo w 50 percent. In the Upper Macrounit, 64 percent of the shells have attachment scars, and in the Lower Macrounit, 60 percent of the shells have attachment scars. In the Upper Macrounit 1, 52 percent of 0.00 0.50 1.00 1.50 2.00 2.50 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Height to Length Ratiio Subsistence Column Sample Number
89 the oyster shells have attachment scars, and in the Up per Macrounit 2, 60 percent of the shells have attachment scars. The type of substrate, shell o 8). The sample with the highest number of oysters that may have been grown on line or cordage is 6 percent (Sample 3), and 11 ou t of 20 samples have no shells that appear to be grown on line or cordage. Because of this extremely low number, no further analysis was done on this attribute, as it seems as though it does not apply to this sample. Table 5 5. Presence or absence of att achment scar by subsistence column sample. Presence/Absence of Attachment Scar n = Present Absent Sample n = % n = % 1 37 32 86 5 14 2 184 112 61 72 39 3 315 155 49 160 51 4 312 133 43 179 57 5 190 100 53 90 47 6 223 128 57 9 5 43 7 246 162 66 84 34 8 254 159 63 95 37 9 187 142 76 45 24 10 258 193 75 65 25 11 292 228 78 64 22 12 276 207 75 69 25 13 324 234 72 90 28 14 3 2 67 1 33 15 14 8 57 6 43 16 14 7 50 7 50 17 11 8 73 3 27 18 49 37 76 12 24 19 16 7 44 9 56 20 47 24 51 23 49
90 Figure 5 4. Percentage of oyster shells with parasitism by subsistence column sample. Table 5 6. Presence or absence of attachment scars Upper and Lower Macrounits. Presence/Absence of Attac hment Scar n = Present Absent n = % n = % Upper Macrounit 3098 1985 64 1113 36 Lower Macrounit 154 61 60 93 40 Table 5 7. Presence or absence of attachment scars by Upper Macrounits 1 and 2 and the Lower Macrounit. Presence/Absence of Attachment Scar n = Present Absent n = % n = % Upper Macrounit 1 1261 660 52 601 48 Upper Macrounit 2 1837 1325 72 512 28 Lower Macrounit 154 61 60 93 40
91 Table 5 8. Type of substrate by subsistence column sample. Type of Substrate Sample n = Shell Line None or Unknown n = % n = % n = % 1 37 32 86 0 0 5 14 2 184 104 57 8 4 72 39 3 315 132 42 18 6 165 52 4 312 131 42 4 1 177 57 5 190 93 49 5 3 92 48 6 223 122 55 5 2 96 43 7 246 155 63 5 2 86 35 8 254 157 62 0 0 97 38 9 187 130 70 9 5 48 26 10 258 195 76 0 0 63 24 11 292 220 75 9 3 63 22 12 276 202 73 4 1 70 25 13 324 233 72 0 0 91 28 14 3 2 67 0 0 7 33 15 14 7 50 0 0 7 50 16 14 7 50 0 0 7 50 17 11 8 73 0 0 3 27 18 49 37 76 0 0 12 24 19 16 7 44 0 0 9 56 20 47 23 49 0 0 24 51 Presence/Absence of Sponge Parasitism The presence/absence of parasitism is shown for Samples 1 20 in the table below (Table 5 9 and Figure 5 5) and then divided into the Upper Macrouni t and Lowe r Macro Unit (Table 5 10), as well as Upper Macrounit 1 and Upper Macrounit 2 (Table 5 11). Both the number of shells in each sample with and without sponge parasitism as well as the percentage of shells in each sample with and without sponge parasitism ar e listed.
92 The lowest percentage of oyster shells with sponge parasitism is 6 percent (Sample 18), and the highest percentage of oyster shell with sponge parasitism is 67 percent (Sample 14). The percent of oysters with sponge parasitism for Sample 14 is a nomalous likely due to the small sample size ( n = 3). The percentage of oyster shells with sponge parasitism in Samples 1 13, the Upper Macrounit, ranges from 39 percent to 61 percent, whereas Samples 14 20, the Lower Macrounit, range from 6 percent to 67 percent of oysters with sponge parasitism per sample. In the Upper Macrounit, 48 percent of the shells have sponge parasitism, and in the Lower Macrounit, 30 percent of the shells have sponge parasitism. In the Upper Macro Unit 1, 46 percent of the oyster shells have sponge parasitism, and in the Upper Macro Unit 2, 49 percent of the shells have sponge parasitism. Figure 5 5. Percent of oyster with sponge parasitism by subsistence column. 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Percentage Subsistence Column Sample Number Percent with Sponge Parasitism
93 Table 5 9. Presence or absence of parasitism by subsistence co lumn sample. Presence/Absence of Parasitism n = Present Absent Sample n = % n = % 1 37 20 54 17 46 2 184 72 39 112 61 3 315 141 45 174 55 4 312 149 48 163 52 5 190 93 49 97 51 6 223 105 47 118 53 7 246 131 56 115 47 8 2 54 116 46 138 54 9 187 95 51 92 49 10 258 141 55 117 45 11 292 142 49 150 51 12 276 125 45 151 55 13 324 149 46 175 54 14 3 2 67 1 33 15 14 2 14 12 86 16 14 5 36 9 64 17 11 2 18 9 82 18 49 3 6 46 94 19 16 7 44 9 56 20 47 25 53 22 47 Table 5 10. Presence or absence of parasitism by Upper and Lower Macrounits. Presence/Absence of Parasitism n = Present Absent n = % n = % Upper Macrounit 3098 1479 48 1619 52 Lower Macrounit 154 46 30 108 70 Table 5 11. Presence or absence of sponge parasitism by Upper Macrounits 1 and 2 and the Lower Macrounit. Presence/Absence of Parasitism n = Present Absent n = % n = % Upper Macrounit 1 1261 580 46 681 54 Upper Macrounit 2 1837 899 49 9 38 51 Lower Macrounit 154 46 30 108 70
94 Presence/Absence of Parasitism on the Attachment Scar The presence/absence of parasitism on attachment scars is shown for Samples 1 20 in the table below (Table 5 12 and Figure 5 6) and then divided into the Uppe r Macrouni t and Lower Macro Unit (Table 5 13), as well as Upper Macrounit 1 and Upper Macrounit 2 (Table 5 14). Both the number of shells in each sample with and without parasitism on the scar as well as the percentage of shells in each sample with and wit hout parasitism on the scar are listed. Furthermore, tables showing percentage of shells with scars with parasitism and with scars without parasitism is disp layed for Samples 1 20 (Table 5 15 and Figure 5 7), the Upper Macroun it and Lower Macrounit (Table 5 16), and Upper Macrounit 1 and Upper Macrou nit 2 (Table 5 17), in both numeric and percentage form. The lowest percentage of oyster shells with parasitism on the scar is 6 percent (Sample 19), and the highest percentage of oyster shell with parasitism o n the scar is 34 percent (Sample 10). Samples 14, 15, 17, and 18 have no shells with parasitism on the scar. The percentage of oyster shells with parasitism on the scar in Samples 1 13, the Upper Macrounit, ranges from 13 percent to 34 percent, whereas Sam ples 14 20, the Lower Macrounit, range from 0 percent to 11 percent of oysters with parasitism on the scar per sample. In the Upper Macrounit, 20 percent of the shells have parasitism on the scar, and in the Lower Macrounit, 5 percent of the shells have pa rasitism on the scar. In the Upper Macro Unit 1, 15 percent of the oyster shells have parasitism on the scar, and in the Upper Macro Unit 2, 24 percent of the shells have parasitism on the scar. Of the shells that have attachment scars, the lowest percent age of oyster shells with parasitism on the scar is 14 percent (Sample 19), and the highest percentage of
95 oyster shell with parasitism on the scar is 46 percent (Sample 10). Samples 14, 15, 17, and 18 have no shells with scars with parasitism on the scar. The percentage of oyster shells with sponge parasitism in Samples 1 13, the Upper Macrounit, ranges from 17 percent to 46 percent, whereas Samples 14 20, the Lower Macrounit, range from 0 percent to 21 percent of oysters with scars with parasitism on the s car per sample. In the Upper Macrounit, 32 percent of the shells that have scars have parasitism on the scar, and in the Lower Macrounit, 8 percent of the shells that have scars have parasitism on the scar. In the Upper Macro Unit 1, 29 percent of the oyst er shells that have scars have parasitism on the scar, and in the Upper Macro Unit 2, 34 percent of the shells that have scars have parasitism on the scar. Figure 5 6. Percent of oyster with sponge parasitism on attachm ent scar by subsistence column. 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Percent Sample Number Percent with Parasitism on the Scar
96 Table 5 12. Presence or absence of sponge parasitism on the attachment scar by subsistence column sample. Presence/Absence of Parasitism on Scar n = Present Absent Sample n = % n = % 1 37 8 22 29 78 2 184 19 10 165 90 3 315 44 14 27 1 86 4 312 40 13 272 87 5 190 34 18 156 82 6 223 48 22 175 78 7 246 63 26 183 74 8 254 60 24 194 76 9 187 54 29 133 71 10 258 88 34 170 66 11 292 73 25 219 75 12 276 49 18 227 82 13 324 53 16 271 84 14 3 0 0 3 100 15 14 0 0 14 100 16 14 1 7 13 93 17 11 0 0 11 100 18 49 0 0 49 100 19 16 1 6 15 94 20 47 5 11 42 89 Table 5 13. Presence or absence of sponge parasitism on attachment scars by Upper and Lower Macrounits. Presence/Absence of Parasitism on Scar n = Present Absent n = % n = % Upper Macrounit 3098 633 20 2465 80 Lower Macrounit 154 7 5 147 95 Table 5 14. Presence or absence of sponge parasitism on attachment scars by Upper Macrounits 1 and 2 and the Lower Macrounit. Presence /Absence of Parasitism on Scar n = Present Absent n = % n = % Upper Macrounit 1 1261 193 15 1068 85 Upper Macrounit 2 1837 449 24 1388 76 Lower Macrounit 154 7 5 147 95
97 Table 5 15. Presence or absence of sponge parasitism on the att achment scar on shells with attachment scars by subsistence column sample. Shells With Scars With Parasitism n = Present Absent Sample n = % n = % 1 32 8 25 24 75 2 112 19 17 93 83 3 155 44 28 111 72 4 133 40 30 93 70 5 100 34 34 66 66 6 128 48 38 80 63 7 162 63 39 99 61 8 159 60 38 99 62 9 142 54 38 88 62 10 193 88 46 105 54 11 228 73 32 155 68 12 207 49 24 158 76 13 234 53 23 181 77 14 2 0 0 2 100 15 8 0 0 8 100 16 7 1 14 6 86 17 8 0 0 8 100 18 37 0 0 37 100 19 7 1 14 6 86 20 24 5 21 19 79 Figure 5 7. Percent of shells with attachment scars that have sponge parasitism on the attachment scar by subsistence column sample. 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Percent Sample Number Percent With Scars and Parasitism on the Scars
98 Table 5 16. Presence or absence of sponge parasi tism on attachment scars on shells with attachment scars by Upper and Lower Macrounits. Shells With Scars With Parasitism n = Present Absent n = % n = % Upper Macrounit 1985 633 32 1352 68 Lower Macrounit 93 7 8 86 92 Table 5 17. Presence or absence of sponge parasitism on attachment scars on shells with attachment scars by Upper Macrounits 1 and 2 and the Lower Macrounit. Shells With Scars With Parasitism n = Present Absent n = % n = % Upper Macrounit 1 660 1 93 29 467 71 Upper Macrounit 2 1325 449 34 876 66 Lower Macrounit 154 7 5 147 95 Left versus Right Shells The amount of left versus right oyster shells is shown for Samples 1 20 in the table and graph below (Table 5 18 and Figure 5 8) and then divided into the Upper Macrouni t and Lower Macro Unit (Table 5 19), as well as Upper Macrounit 1 and Upper Macrounit 2 (Table 5 20). Both the number of left and right shells in each sample as well as the relative and absolute frequencies of left and r ight shells in each sample are listed. Generally, left oysters predominate the samples in the Upper Macrounit, especially in Macrounit 2, the sample is comprised of 65 percent left oyster shells. The ratio of left to right valves is also displayed below by subsistenc e column sample number (Table 5 21 and Figure 5 9).
99 Table 5 18. Number and percentage of left and right oyster valves by subsistence column sample. Right Versus Left Oyster Shells n = Right Left Sample n = % n = % 1 87 50 57 37 43 2 378 194 51 184 49 3 616 301 49 315 51 4 672 360 54 312 46 5 446 256 57 190 43 6 478 255 53 223 47 7 436 190 44 246 56 8 437 183 42 254 58 9 321 134 42 187 58 10 398 140 35 258 65 11 420 128 30 292 70 12 363 87 24 27 6 76 13 430 106 25 324 75 14 6 3 50 3 50 15 36 22 61 14 39 16 37 23 62 14 38 17 22 11 50 11 50 18 117 68 58 49 42 19 49 33 67 16 33 20 114 67 59 47 41 Figure 5 8. Percentage of left and right oyster valves by subsistence colum n sample. 0 10 20 30 40 50 60 70 80 90 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Percentage Subsistence Column Sample Number Right versus Left Oyster Shells Right Left
100 Table 5 19. Percent of right and left valves by Upper and Lower macrounit. Right Versus Left Oyster Shells n = Right Left n = % n = % Upper Macrounit 5482 2384 43 3098 57 Lower Macrounit 381 227 60 154 40 Table 5 2 0. Percent of right and left valves by Upper Macrounits 1 and 2 and Lower Macrounit. Right Versus Left Oyster Shells n = Right Left n = % n = % Upper Macrounit 1 2677 1416 53 1261 47 Upper Macrounit 2 2805 968 35 1837 65 Lower Macrou nit 381 227 60 154 40 Left Valve Concavity The left valve concavity ranking, 1 3 with 1 being the least concave and 3 being the most c oncave is displayed below by subsistence column sample number (Table 5 21 and Figure 5 9) and by Upp er and Low er Macrounit (Table 5 22) and Upper Macrounit 1 and Upper Macrounit 2 (Table 5 23). Both the number of shells in each sample as well as the percentage of shells in each sample with the left valve concavity value are listed. The shells in samples 14 20 are the most concave with 70 percent of the shells having a left valve concavity value of 3, 25 percent having a left valve concavity value of 2, and only 5 percent having a left valve concavity value of 1. The next most concave shells are from the Upper Macr ounit 2, with 44 percent of the shells having a left valve concavity value of 3, 46 percent having a left valve concavity value of 2, a nd 10 percent having a left valve concavity value of 1. The Upper Macrounit 2 has the least concave shells on average, wi th 30 percent having a left valve concavity value
101 of 3, 55 percent having a left valve concavity value of 2, and 15 percent having a left valve concavity value of 1. Table 5 21. Number and percentage of shells with each left valve concavity value by subs istence column sample. Left Valve Concavity Sample n = Value of 1 Value of 2 Value of 3 n = % n = % n = % 1 37 5 14 24 68 8 22 2 184 33 18 106 58 45 24 3 315 33 17 191 61 71 23 4 312 50 16 172 56 88 28 5 190 21 11 93 49 76 40 6 223 35 16 103 46 85 38 7 246 20 8 94 38 132 54 8 254 16 6 124 49 114 45 9 187 19 10 72 39 96 51 10 258 27 10 113 44 118 46 11 292 28 10 151 52 113 36 12 276 42 15 122 44 112 41 13 324 32 10 173 53 119 37 14 3 0 1 0 0 3 100 15 1 4 0 1 4 29 10 71 16 14 0 1 4 29 10 71 17 11 3 27 2 18 6 55 18 49 4 8 11 22 34 69 19 16 0 0 4 25 12 75 20 47 1 2 13 28 33 70
102 Figure 5 9. Percent of oyster shells with each left valve concavity value by subsistence column sample. Table 5 22. Number and percenta ge of shells with each left valve concavity value by Upper and Lower macrounit. Left Valve Concavity n = Value of 1 Value of 2 Value of 3 n = % n = % n = % Upper Macrounit 3098 381 12 1540 50 1177 38 Lower Macrounit 1 54 8 5 38 25 108 70 Table 5 23. Number and percenta ge of shells with each left valve concavity value by Upper Macrounits 1 and 2 and the Lower Macrounit. Left Valve Concavity n = Value of 1 Value of 2 Value of 3 n = % n = % n = % U pper Macrounit 1 1261 197 16 691 55 373 30 Upper Macrounit 2 1837 184 10 849 46 804 44 Lower Macrounit 154 8 5 38 25 108 70 Biofouling The presence or absence of barnacle biofouling is displayed below by subsistenc e column sample number (Ta ble 5 24) and by Upp er and Lower Macrounit 0 20 40 60 80 100 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Percentage Subsistence Column Sample Left Valve Concavity Cuppyness Value 1 Cuppyness Value 2 Cuppyness Value 3
103 (Table 5 25) and Upper Macrounit 1 and Upper Macrounit 2 (Table 5 26). Both the number of shells in each sample as well as the percentage of shells in each sample are listed for the presence and absence of biofoul ing. The highest percentage of shells that show evidence of biofouling is in the Upper Macrounit (6 percent), particularly Upper Macrounit 2 (7 percent), and almost no shells (1 percent) have evidence of biofouling in the Lower Macrounit. The highest perce ntage of shells with evidence of biofouling from the subsistence column samples is Sample 9 (11 percent), and the lowest is 0 percent in Samples 1 and 14 18. Table 5 24. Number and percentage of shells with biofouling present and absent by subsistence co lumn sample. Presence/Absence of Biofouling n = Present Absent Sample n = % n = % 1 37 0 0 37 100 2 184 5 3 179 97 3 315 6 2 309 98 4 312 15 5 297 95 5 190 13 7 177 93 6 223 14 6 209 94 7 246 12 5 234 95 8 254 24 9 230 91 9 187 21 11 166 89 10 258 20 8 238 92 11 292 21 7 271 93 12 276 15 5 261 95 13 324 22 7 302 93 14 3 0 0 3 100 15 14 0 0 14 100 16 14 0 0 14 100 17 11 0 0 11 100 18 49 0 0 49 100 19 16 1 6 15 94 20 47 1 2 46 98
104 Tab le 5 25. Number and percentage of shells with biofouling present and absent by Upper and Lower Macrounit. Presence/Absence of Biofouling n = Present Absent n = % n = % Upper Macrounit 3098 188 6 2910 94 Lower Macrounit 154 2 1 152 99 Table 5 26. Number and percentage of shells with biofouling present and absent by Upper Macrounit 1 and 2, and Lower Macrounit. Presence/Absence of Biofouling n = Present Absent n = % n = % Upper Macrounit 1 1261 53 4 1 208 96 Upper Macrounit 2 1837 135 7 1702 93 Lower Macrounit 154 2 1 152 99 Hypotheses 1 and 2: Location of Harvest In order to test Hypotheses 1 (H1), the location of collection (i.e., subtidal versus intertidal) will covary with the sca le and intensity of oyster procurement, and Hypothesis 2 (H2), there will be no change in the location of harvest in accordance with changes in the scale and intensity of oyster procurement, the relevant variables to consider are height, length, HLR, attac hment scars, biofouling, parasitism, and left valve concavity The mean height of the oyster shells is 54.03 mm in the Upper Macrounit 1 and 56.46 mm in the Upper Macrounit 2. The shells in Upper Macrounit 2 are, on average, 2.43 mm smaller in height tha n the shells in the Upper Macrounit 1. This difference was tested with the t Test function of Excel (Version: 14.0.6129.5000) assuming unequal variances and unequal sample sizes (results of all t Tests are provided in the Appendix ). Results indicate the diff erences in mean height between the Upper
105 Macrounit 1 and Upper Macrounit 2 are statistically significant at less than 0.01 probability (t = 4.14; df = 3,059). The oysters from the Upper Macrounit 2 are on average 10.73 mm larger in height than those in t he Lower Macrounit, with a difference in means that is statistically significant at less than 0.01 probability (t = 8.24; df = 192). The oyster shells from Upper Macrounit 1 are on average 8.3 mm larger in height than those in the Lower Macrounit, with a d ifference of means that is also statistically significant at less than 0.01 probability (t = 6.43; df = 186). The mean length of the Upper Macrounit 1 is 31.68 mm and the mean length of the Upper Macrounit 2 is 31.86 mm, where the Upper Macrounit 2 is 0.1 8 mm longer than the Upper Macrounit 1. Using the same t Test that was used to determine the significance of the difference of means for height, results indicate the differences in mean length between the Upper Macrounit 1 and Upper Macrounit 2 in not stat istically significant (p = 0.29; t = 0.57; df = 3,016). The oysters in the Upper Macrounit 2 are on average 6.27 mm longer than in the Lower Macrounit. The difference in mean length between the Upper Macrounit 2 and the Lower Macrounit is statistically si gnificant at less than 0.01 probability (t = 9.25; df = 194). The oysters in the Upper Macrounit 1 are on average 6.08 mm longer than the Lower Macrounit, and the difference in mean length is statistically significant at less than 0.01 probability (t = 9; df = 192). Both the mean height and length of the oysters in the Upper Macrounits, especially Upper Macrounit 2, are larger than in the Lower Macrounit. Further, based on the results of t tests performed on the measurements of height and length, it appea rs that the oysters in the Upper Macrounits and Lower Macrounit were not drawn from the same population.
106 In terms of shell shape, the mean HLR of the Upper Macrounit 1 is 1.72, the Upper Macrounit 2 is 1.78, and the Lower Macrounit is 1.80, so as the inte nsity of harvest increases the shells become less elongate. Further evidence that the oysters harvested during times of large scale intensive harvesting are subtidal include the increase in percentage of shells with sponge parasitism from 30 percent in the Lower Macrounit to close to 50 percent in the Upper Macrounits (46 percent in Upper Macrounit 1 and 49 percent in Upper Macrounit 2), and the increase in percentage of shells with evidence of barnacle biofouling from 1 percent in the lower Macrounit to 7 percent in Upper Macrounit 2 and 4 percent in Upper Macrounit 1.The final variable, left valve concavity is opposite from what was expected, where subtidal oysters are more concave than intertidal oysters. Seventy percent of the oysters in the Lower Macro unit have a left valve concavity value of 3, whereas in Upper Macrounit 2, 44 percent of oysters have a left valve concavity value of 3, and in Upper Macrounit 1, only 30 percent of the oysters have a left valve concavity value of 3. Based on the results, H1 is proven true and H2 is proven false; as the scale and intensity of oyster procurement increases, the oysters in the samples and macrounits shift from intertidal oysters in the Lower Macrounit to more subtidal oysters in the Upper Macrounits. This obs ervation is supported by a shift from the Lower Macrounit to the Upper Macrounit 2, where there are generally larger oysters, shell shape changes from elongate shells to rounder shells (HLR), and the increase in the percentage of shells with biofoul and sp onge parasitism. Hypotheses 3 and 4: Shelling To test Hypothesis 3 (H3), that the amount of right oyster valves used for cultch (shell returned to the water to enhance oyster reefs) will covary with the scale and
107 intensity of oyster procurement, and Hypot hesis 4, that there will be no change in the ratio of left to right oyster shells in accordance with changes in the scale and intensity of procurement, the amount of right and left valves per sample and macrounit were counted and compared. The ratio of le ft to right valves is particularly high in the transition from the Lower Macrounit to Upper Macrounit 2 (Samples 10 13), when the scale and intensity of harvest increases. The ratio in the Lower Macrounit stays around 50:50, whereas, at some points, the ra tio of left to right valves is 3:1 in Upper Macrounit 2, when the scale and intensity of oyster procurement increases. Based on these results, H3 is proven true, there number of right vlaves used for cultch does covary with the scale and intensity of oyste r procurement, and H4 is proven false. Hypotheses 5 and 6: Relaying To test Hypothesis 5 (H5), that the amount of oysters that were transplanted from suboptimum conditions to optimal conditions through relaying will covary with the scale and intensity of oyster procurement, and Hypothesis 6 (H6), that there will be no indication that oysters were transplanted in accordance with the scale and intensity of oyster procurement, the variables tested were height, HLR, presence or absence of sponge parasitism, pr esence or absence of biofoul, and left valve concavity Many of these variables were discussed when testing H1 and H2 above, and it appears that there is little to no evidence of transplanting when the scale and intensity of oyster harvesting increases, th erefore proving H5 false and H6 true. Hypotheses 7 and 8: Culling Hypothesis 7 (H7), evidence of oyster culling will covary with the scale and intensity of oyster procurement, and Hypothesis 8 (H8), there will be no indication that
108 oysters were culled in accordance with the scale and intensity of oyster procurement, were tested by comparing the percent of attachment scars with evidence of parasitism on them. Oysters from Upper Macrounit 2 have the most evidence of culling, with 24 percent of oysters havin g parasitism on the attachment scars, an increase from five percent in the Lower Macrounit. The percent of shells that show evidence of culling again decreases to 15 percent in Upper Macrounit 1. Based on the results, H7 is proven true as evidence of culli ng increases with the increase in the scale and intensity of oyster procurement, and H8 is proven false. Hypotheses 9 and 10: Size Selection Hypotheses 9 (H9), that the size range of oysters will covary with the scale and intensity of procurement, and Hypo thesis 10 (H10), that there will be no change in the size range of oysters in accordance with the scale and intensity of oyster procurement, were tested based on the height of the oysters in each sample and macrounit. Based on the t Tests performed between macrounits, it appears as though, based on height, the oysters from the three macrounits are likely drawn from d i fferent populations (Appendix ). Based on the size ranges of harvested oysters in each of the macrounits, the height for the majority of the oysters is from 20 mm to 80 mm, with the height range of 40 mm to 60 mm having the highest percentage for each population. It does not appear that there is any major change in the size range of oysters being harvested based on an increase of intensity of h arvest, but the consistency in height range throughout the macrounits may indicate a consciousness of size selection throughout time, regardless of the scale and intensity of harvest. Based on these results H9 is proven false, and H10 is proven true.
109 Hypo theses 11 and 12: Off Bottom Growing In order to test Hypothesis 11 (H11), that the amount of oysters grown using off bottom techniques will covary with the scale and intensity of oyster procurement, and Hypothesis 12 (H12), that there will be no evidence of off bottom techniques being used in accordance with the scale and intensity of oyster procurement, the type of attachment scar was compared. Because of the very small number of shells that appear to have an attachment scar representing substrate other t han shell ( n = 67 of 3,252), H12 is proven to be true and H11 is proven to be false. Palmetto Mound In order to test the assumption that Shell Mound and Palmetto Mound were related to each other, oyster shell from Test Unit 1 (TU1) at Palmetto Mound was an alyzed. Furthermore, data were collected from three other sites spanning from north to south in the study area: Bird Island (8DI52), Cat Island (8DI29), and North Key (6LV6 5). In the table below (Table 5 27), descriptive statistics for Palmetto Mound are l isted for the height, length, and height to length ratio. The mean height is 53.54 mm, with a minimum of 13.93 mm and a maximum of 111.59, the mean length is 30 mm, and the mean HLR is 1.80, with a minimum of 1.17 and a maximum of 3.67. Table 5 27. Descri ptive statistics for oyster height, length, and HLR for Palmetto Mound. 8LV2 Height, Length, HLR (mm) Mean SD Minimum Maximum Height 53.54 16.16 13.93 111.59 Length 30.00 8.46 6.56 69.94 HLR 1.80 0.29 1.17 3.67 When compared to the rest of th e sites, what is particularly striking is the similarity of the percentage of shells with parasitism in Upper Macrounit 1 and 2 and
110 Palmetto Mound. The percent of shells with parasitism for those three samples are close to fifty percent, whereas for all ot her midden sites, including the Lower Macrounit in TU8 at Shell Mound are c lose to thirty percent (Table 5 28). Other than that, the means of the height, length, and HLR are all similar, with the exception of Cat Island which has larger shells on average t hat are more elongate. Table 5 28. Comparison of mean height, length, HLR, and percentage of parasitism between five sites in the Lower Suwannee research area. Height, Length, HLR, and Parasitism for Five Sites in the Lower Suwannee Site n = Mean Height Mean Length Mean HLR With Parasitsim n = % 8DI29 373 62.80 33.07 1.91 114 31 8DI52 1275 52.55 30.47 1.74 364 29 8LV65 1081 52.98 30.04 1.78 328 30 8LV2 519 53.54 30.00 1.80 256 49 8LV42 Lower Macrounit 154 45.73 25.60 1.80 46 30 8LV42 Upper Macrounit 1 1261 54.03 31.68 1.72 580 46 8LV42 Upper Macrounit 2 1837 56.46 31.86 1.78 899 49 Discussion Based on the results presented above, it appears that the location of collection (i.e., subtidal versus intertidal), the amount of right oyster valves used for cultch, and the evidence of culling covary with the scale and intensity of oyster harvesting at Shell Mound, and there is little to no evidence that oysters were transplanted or grown using off bottom techniques in accordance with the scale and intensity of oyster harvesting at Shell Mound. Evidence of size selection remains relatively consistent across samples, the com parison of oyster size and the amount of oysters with evidence of sponge parasitism indicates that oysters were possibly taken from Shell Mound, particularly the
111 Upper Macrounit 1, and used to cap burials at Palmetto Mound. These results will be discussed in more detail below. The data have been divided three different ways for the purpose of this analysis: by sample number which corresponds to the subsistence column which was excavated by 10 cm arbitrary levels, by upper and lower macrounits that separate the submound midden from the mounded shell, and then the Upper Macrounit was further divided into Upper Macrounit 1 and Upper Macrounit 2, following a break in stratigraphy of the mounded shell. Dividing the data into macrounits for analysis is helpful to separate different phases of occupation and construction at Shell Mound; this is particularly true of Samples 14 20 in the submound midden, or the Lower Macrounit, where the sample sizes are so small and some data are anomalous almost certainly due to sma ll sample size (i.e., Sample 14 where n = 3). For the purposes of the following discussion, I discuss the data in their relationship to Macrounits to make the comparisons more meaningful. As the macrounits are divided based on stratigraphy, I start with th e oldest samples or the Lower Macrounit of submound midden. I then discuss the Upper Macrounit 2, and finally the Upper Macrounit 1. I then compare the results of the analysis of each of the macrounits to the data obtained from oyster shells taken from cap ping of burials from Palmetto Mound. Lower Macrounit (Samples 14 20) The Lower Macrounit has far fewer shells ( n = 154) in comparison to the Upper Macrounit 1 ( n = 1,261) and the Upper Macrounit 2 ( n = 1,837). In general, these oysters are small (mean he ight = 45.73 mm; mean length = 31.87 mm) and elongate (mean HLR = 1.80 mm), lacking in parasitism (70 percent without sponge parasitism), the majority of which have attachment scars (60 percent). These attributes are all strong
112 indicators that these oyster s came from intertidal conditions. The high left valve concavity value of these shells is surprising (70 percent with a left valve concavity value of 3), given that all other variables indicate that the oysters were harvested from intertidal conditions. Wh at could account for this is more selection in harvesting, as the amount that were harvested and deposited into the submound midden were far fewer than what was accumulated on top during the intensive phase of mound building. Evidence for any form of mari culture is lacking from the Lower Macrounit; the percentage of right and left valves does not indicate shelling (60 percent right valves and 40 percent left valves), there seems to be no signs of relaying, as all attributes are generally indicative of inte rtidal conditions, only five percent of the shells have parasitism on the attachment scars, so the oysters do not appear to be culled and returned to the water. There is also no convincing patterning between samples in the Lower Macrounit where attributes are widely varied. Upper Macrounit 2 (Samples 7 13) On average the shells in Upper Macrounit 2 are larger (mean height = 56.45 mm; mean length = 31.87 mm) and rounder (mean HLR = 1.78) in the Upper Macrounit 2 than in the Lower Macrounit or Upper Macrouni t 1, and almost half of the oysters have evidence of sponge parasitism (49 percent). The combination of these attributes indicates that the oysters in Upper Macrounit 2 were likely primarily from subtidal conditions. Indications of mariculture are stronges t in this macrounit. On average, 65 percent of the total shell count are left valves, meaning that on average, there are 15 percent more left valves in a sample than right valves. The patterning of more left than right valves is particularly strong in Samp les 10 13 which where the percentage of left valves
113 ranges from 65 to 76. These high percentages are indicative of shelling practices, where the right valves are not present in the samples as they were returned to the water for spat recruitment. Evidence o f culling is present in the consistent percentage of shells with evidence of parasitism on the attachment scars. The most striking evidence of this is in Samples 7 11 where the percentage of oysters with parasitism on their attachment scars ranges from 32 to 46 percent. What makes these data convincing of maricultural practices is not just the numbers or percentages, but that there is a strong patterning of consistency in almost every variable with no extreme ranges or outliers between samples. Upper Macro unit 1 (Samples 1 6) Like in Upper Macrounit 2, the oysters from Upper Macrounit 1 appear to be harvested primarily from subtidal areas. This is evident due to the larger size (mean height = 54.03 mm; mean length = 31.68 mm), low mean HLR (1.72 mm), and hi gh percentage of parasitism (46 percent). In this macrounit, evidence of maricultural practices declines through time from Sample 6 to Sample 2, where Sample 1 is sometimes an outlier which may be the result of disturbance due to its proximity to the prese nt day ground surface. Through time the mean height decreases from about 55 mm to about 50 mm, mean length increases from about 30 mm to 32 mm, and the mean HLR decreases from 1.80 mm to 1.56 mm, indicating that the harvested oysters are becoming smaller a nd rounder. Evidence of culling also drops off from 38 percent to 17 percent of shells with sponge parasitism on the attachment scar, and finally, the ratio of right to left valves remains close to 50 percent in all samples, with an average of 47 percent r ight valves and 53 percent left valves.
114 Palmetto Mound The oyster shells recovered from Palmetto Mound are assumed to be purposefully placed as capping material over individual burials, rather than the concentrated effect of consumption on the island. In order to test if the shells at Palmetto Mound were transferred from Shell Mound, the samples were compared. When comparing height, the difference between the mean of the Upper Macrounit and Palmetto Mound is statistically significant at less than 0.01 prob ability (t = 2.45; df = 711). Further dividing the Upper Macrounit into Upper Macrounit 1 and Upper Macrounit 2, the difference in mean height between Palmetto Mound and the Upper Macrounit 1 is not statistically significant (p = 0.28; t = 0.59; df = 846), whereas the difference in mean height between Palmetto Mound and the Upper Macrounit 2 is statistically significant at less than 0.01 probability (t = 3.46; df = 915). The results of the t Tests are reported in the Appendix Of particular interest is the h igh percentage of oysters with parasitism within the samples of the Upper Macrounits 1 and 2 and Palmetto mound, all being close to 50 percent. This is convincing of the transferal of oyster shell from Shell Mound to Palmetto Mound because at three other s ites, Bird Island, Cat Island, and North Key, as well as the submound midden or Lower Macrounit at Shell Mound, the percentage of shell with parasitism from middens is close to 30 percent. Based on this data, it appears as though shell may have been taken from the Upper Macrounits at Shell Mound, particularly Upper Macrounit 2, and placed over burials at Palmetto Mound, a hypothesis that needs further testing with additional data, as the differences may be attributed to microhabitats.
115 Based on the results of this analysis I would argue that there is a distinct pattern of no evidence of mariculture in the submound midden, compelling evidence of mariculture in the initial phase of mound building, and a steady decline in evidence of mariculture in the last pha se of mound building, perhaps when the site was abandoned and returned to periodically leaving no one to manage the oysters or because the shell was procured from an earlier midden and then placed ontop of the Upper Macrounit 2, creating reverse stratigrap hy. The oysters from the Lower Macrounit appear to be the refuse from meals in which intertidal oysters were used to supplement meals and there was no real need for any sort of maricultural practices. Based on the data from the Upper Macrounit 2, I would argue that during the initial phase of mound building at Shell Mound, right shells were returned to the water to shell reefs and beds, a practice which was sustained, but less so as time went by. The initial practice of shelling may have been to prepare th e existing reefs for heavy exploitation during mound building events. The practice may have fallen out of favor due to the environment of the area, where shelling may not prove effective because the reefs are not protected like they are in Apalachicola Bay and Tampa Bay, therefore storm events may have washed the loose shell from the reefs back to the shore, defeating the purpose. With the decline in shelling came a rise of culling where smaller oysters were broken away from burrs and returned to the water to continue growing and living through reproductive cycles. During this phase of intensive mound building mostly subtidal oysters were harvested, and size and age selection appear to be a usual consideration, where mostly juvenile and adult oysters were ha rvested and very few spat made it out of the water and into the mound. During
116 the second phase of mound construction, evidence of mariculture dissipates gradually, but the oysters remain large and subtidal. The association of Shell Mound and Palmetto Moun d is argued with evidence suggesting that oyster shell may have been taken from the mound at Shell Mound and then placed over burials at Palmetto Mound. Both height and the percentage of shell with parasitism are strikingly similar between the samples take n from the two sites. The combination of the data analysis, environmental considerations, and culture history of the Lower Suwannee provide compelling evidence of mariculture during the intensive phase of mound building at Shell Mound. The concluding rema rks in the following final chapter provide a summary, concluding discussion, and suggestions future directions for this work.
117 CHAPTER 6 SUMMARY AND CONCLUSION Oyster shell from bulk samples from Test Unit 8 near the apex of Shell Mound on f Coast were analyzed for evidence of maricultural practices during the fifth and sixth centuries A.D. Although shell mounds, ridges, and middens are ubiquitous in the Lower Suwannee study area, Shell Mound was chosen as the site used to test the hypothesi s that humans were engaged in oyster maricultural practices due to the size of the mound, the speed at which the mound was constructed (~200 years), the cultural and environmental factors at play during the occupation of the site, and the potential symboli c importance of the site in relation to the nearby burial complex, Palmetto Mound. Test Unit 8 is representative of three main phases of occupation or activity, with a submound midden, a strata of unconsolidated, mounded shell, and a strata of mostly oyste r shell which may have been deposited from nearby middens in the final phase of mound construction. A total of 3,252 left oyster valves were analyzed for morphological attributes including height, length, height to length ratio, presence of sponge parasiti sm, presence of attachment scars, presence of parasitism on attachment scars, left valve concavity and ratio of left to right valves. Each of these attributes is indicative of environmental circumstances and/or human influence on local populations. In or der to devise methods of analysis, a detailed understanding of oyster biology and ecology as well as the local environment is important and was discussed in Chapter 2. Based on paleoenvironmental data, it appears that the environmental conditions on Florid regression creating changes in water temperature and salinity, which are extremely
118 influential in oyster growth and reproduction. These shifts in environmental conditions or harves ting niches are visible on the morphological aspects of oyster shells. As well as environmental variability, the changing cultural traditions, which are discussed in Chapter 3, impacted oyster harvesting practices, especially in regard to the intensity of harvesting. Shell Mound is one of three well documented civic ceremonial in place starting at about 200 A.D. The civic ceremonial centers are argued to be places of year round occupation, aggregation events and feasting, and increased populations, all of which would increase the amount of oysters being harvested from local populations which may have had a negative impact on oyster reefs, potentially leading to overharvesti ng or collapse. I argue that, in order to offset environmental variability and cultural practices resulting in increased harvesting, the inhabitants of Shell Mound employed maricultural practices. Based on ethnographic, ethnohistoric, and archaeological d ata from around the world, people have been engaged in maricultural practices since at least 2,000 years ago, although it has never been investigated at shell bearing sites on the East or Gulf coasts of North America. Because of the lack of archaeological studies investigating oyster mariculture, methods were devised based on oyster biology, ecology, previous studies of shellfish management practices, and personal experiences with present day oyster mariculturalists and oyster biologists in Cedar Key and Ap alachicola Bay. The methods used were discussed in Chapter 4. In order to infer oyster mariculture, morphological attributes which are indicative of environmental conditions and human intervention were analyzed. The recorded
119 attributes provide information regarding the salinity of the water from which the oysters were harvested, if the oysters were harvested from intertidal or subtidal areas, the size size based on t he left valve concavity of the shells. Furthermore, indications of five different maricultural practices were assessed including shelling, relaying, size or age selection, culling, and off bottom growing. Based on the results described in Chapter 5, it app ears that the inhabitants of Shell Mound were practicing shelling and culling during the initial phase of mound building which involved sustained intensive harvesting of oysters. Also, during this time there was a shift from intertidal oysters from the sub mound midden to subtidal oysters used in mound construction. Patterning for these practices dissipates in the final phase of mound building, where oyster shell from nearby middens may have been emplaced on the mounded shell after the site was abandoned as a place of residential occupation. The results also indicate an association of Shell Mound to the burial complex, Palmetto Mound, where shell, especially from the final phase of mound construction at Shell Mound, is very similar to that excavated from Palm etto Mound in terms of size and percentage of shell with sponge parasitism. Of particular interest is the high percentage of shells with parasitism from the mounded shell at Shell Mound and the shells from Palmetto Mound when compared to the relatively low percentage of shell with sponge parasitism from midden sites distributed across the study area. The similarities of the shell recovered from Shell Mound and Palmetto Mound may be indicative of using the mounded shell at Shell Mound as capping material dur ing burial practices at Palmetto Mound, further supporting the assumption of the relationship of the two sites.
120 Since this analysis has revealed compelling evidence of maricultural practices in place at Shell Mound, it is important to replicate the patter ns in order to test the validity and secure the interpretation. The next step for investigating mariculture at Shell Mound would be to conduct the same analysis, using the same methods, on oyster shell recovered from another subsistence column from another test unit on the mound and compare the results. Then, the methods should be applied to various other sites in the study area, in particular Garden Patch and Crystal River, in order to test if mariculture was happening at other civic ceremonial centers in the region Another important step in oyster research at Shell Mound, and in the Lower Suwannee in general, would be to obtain seasonality data from the shells. I also argue that a detailed analysis such as this one, where all aspects of the oyster shells are recorded and compared, could be useful in addressing questions other than those related to mariculture, including questions of feasting, mound building activities, environmental change, and shell symbolism. Not only does this research provide insight into the practices and lifeways of the inhabitants of Shell Mound during the fifth and sixth century A.D., but it also addresses important challenges that coastal communities are facing today, with oyster populations dwindling to a tenth of what they once were world wide. Mariculture has been taking off in places such as the Chesapeake Bay as well as along the Gulf Coast in order to revitalize this economically and environmentally important resource. Using the truly deep time perspective that archaeology ca n offer, as well as the insights into past practices used to sustain oyster populations such as those discussed in this thesis,
121 contributions could be made to broader research aims of environmental reconstruction and restoration.
122 APPENDIX T TESTS Table A 1. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit 1 Upper Macrounit 2 Mean 54.03374306 56.44986391 Variance 198.2686922 337.9054387 Observations 1261 1837 Hypothesized Mean Difference 0 df 3059 t Stat 4.1364687 P(T<= t) one tail 1.81094E 05 t Critical one tail 1.645351905 P(T<=t) two tail 3.62189E 05 t Critical two tail 1.960739792 Table A 2. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit 2 Lower Macrounit Mean 56.44986391 45.72623377 Variance 337.9054387 232.6619818 Observations 1837 154 Hypothesized Mean Difference 0 df 192 t Stat 8.237411584 P(T<=t) one tail 1.33934E 14 t Critical one tail 1.652828589 P(T<=t) two tail 2.67868E 14 t Critical two tail 1.9723 96491
123 Table A 3. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit 1 Lower Macrounit Mean 54.03374306 45.72623377 Variance 198.2686922 232.6619818 Observations 1261 154 Hypothesized Mean Difference 0 df 18 6 t Stat 6.432351451 P(T<=t) one tail 5.16826E 10 t Critical one tail 1.653087138 P(T<=t) two tail 1.03365E 09 t Critical two tail 1.972800114 Table A 4. t Test: Two Sample Assuming Unequal Variances Length Upper Macrounit 1 Upper Ma crounit 2 Mean 31.68486915 31.86527382 Variance 61.9610077 95.07529818 Observations 1261 1837 Hypothesized Mean Difference 0 df 3016 t Stat 0.567961737 P(T<=t) one tail 0.285051614 t Critical one tail 1.645359012 P(T<=t) two tail 0.570103228 t Critical two tail 1.960750857
124 Table A 5. t Test: Two Sample Assuming Unequal Variances Length Upper Macrounit 2 Lower Macrounit Mean 31.86527382 25.59707792 Variance 95.07529818 62.81719467 Observations 1837 154 Hypothesized Mean Difference 0 df 194 t Stat 9.245372026 P(T<=t) one tail 2.13805E 17 t Critical one tail 1.652745977 P(T<=t) two tail 4.27611E 17 t Critical two tail 1.972267533 Table A 6. t Test: Two Sample Assuming Unequal Variances Length Uppe r Macrounit 1 Lower Macrounit Mean 31.68486915 25.59707792 Variance 61.9610077 62.81719467 Observations 1261 154 Hypothesized Mean Difference 0 df 192 t Stat 9.004975138 P(T<=t) one tail 1.06866E 16 t Critical one tail 1.652828589 P(T<=t) two tail 2.13732E 16 t Critical two tail 1.972396491 Table A 7. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit Palmetto Mound Mean 55.46641382 53.54533719 Variance 282.3951877 270.9013558 Observations 3098 519 Hypothesized Mean Difference 0 df 711 t Stat 2.453416334 P(T<=t) one tail 0.007194805 t Critical one tail 1.646999574 P(T<=t) two tail 0.014389611 t Critical two tail 1.963306103
125 Table A 8. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit 1 Palmetto Mound Mean 54.03374306 53.54533719 Variance 198.2686922 270.9013558 Observations 1261 519 Hypothesized Mean Difference 0 df 846 t Stat 0.592628155 P(T<=t) one tail 0.2767942 t Critical one tail 1.646656758 P(T<= t) two tail 0.553588399 t Critical two tail 1.962772035 Table A 9. t Test: Two Sample Assuming Unequal Variances Height Upper Macrounit 2 Palmetto Mound Mean 56.44986391 53.54533719 Variance 337.9054387 270.9013558 Observations 1837 519 Hypothesized Mean Difference 0 df 915 t Stat 3.457005444 P(T<=t) one tail 0.000285682 t Critical one tail 1.646520646 P(T<=t) two tail 0.000571363 t Critical two tail 1.962560005
126 LIST OF REFERENCES Anderson, David G., and Kenneth E. Sassam an 2012 Recent Developments in Southeastern Archaeology: From Colonization to Complexity Society for American Archaeology Press, Washington, D.C. Andrus, C. Fred T., and Victor D. Thompson 2012 Determining the Habitats of Mollusk Collection at th e Sapelo Island Shell Ring Complex, Georgia, USA Using O xygen Isotope Sclerochronology. Journal of Archaeological Science 39:215 228. Beck, Michael W., Robert D. Brumbaugh, Laura Airoldi, Alvar Carranza, Loren D. Coen, Christi ne Crawford, Omar Defeo, Graham J. Edgar, Boze Hancock, Matthew C. Kay, Hunter S. Lenihan, Mark W. Luckenbach, Caitlyn L. Toropova, Guofan Zhang, and Ximing Guo 2011 Oyster Reefs at Risk and Re commendations for Conservation, Restoration, and Manage ment. Bioscience 61 (2): 107 114. Bergquist, Derek, Jason Hale, Patrick Baker, and Shirley Baker. 2006 Development of Ecosystem Indicators for the Suwannee River Estuary: Oyster Reef Habitat Quality along a Salinity Gradient. Estuaries and C oasts 29(3):353 360. Bird, Douglas W., Rebecca Bliege Bird, and Jennifer L. Richardson. 2004 Meriam Ethnoarchaeology: Shellfishing and Shellmiddens. Memoirs of the Queensland Museum, Culture 3(1):183. Bledsoe, Erin L., and Edward J. Phlips 2000 Relationships Between Phytoplankton Standing Crop and Physical, Chemical, and Biological Gradients in the Suwannee River and Plume Region, USA. Estuaries 23 (4):458 473. Borremans, Nina Thanz n.d. Unpublished records of archaeological excavat ions at North Key and Seahorse Key, Florida, ca. 1989. Records on file, Florida Museum of Natural History, University of Florida, Gainesville. Brose, David S., and George W. Percy 1974 Weeden Island Settlement Subsistence and C eremon ialism: A Reappraisal in Systemic Terms. Paper presented at the 37 th Annual Meeti ng of the Society for American Archaeology, Washington D.C. Brown, Frank, and Y. Kathy Brown 2009 Staying the Course, Staying Alive. Coastal Fir st Nations Fundam ental Truths: Biodiversity, Stewardship and Sustainability. Biodiversity British Columbia, Victoria, British Columbia.
127 Bullen, Ripley P., and Edward M. Dolen 1960 Shell Mound, Levy County, Florida. The Florida Anthropologist 13:17 23. Butle r, P.A. 1949 Gametogenesis in the Oyster Under Conditions of Depressed Salinity. Biology Bulletin 96:263 269. Cannon, Aubrey and Meghan Burchell 2009 Clam Growth Stage Profiles as a Measure of Harvest Intensity and Resource Management on th e Central Coast of British Columbia. Journal of Archaeological Science 36:1050 1060 Castagna, Michael, Mary C. Gibbons, and Kenneth Kurkowski 1996 Culture: Application. In Eastern Oyster edited by Victor S. Kennedy and Roger I.E. Newell, pp. 675 690. Maryland Sea Grand College, College Park, Maryland. Chanton, Jeffrey and F. Graham Lewis 2002 Examination of Coupling Between Primar y and Secondary Production in a River Dominated Estuary: Apalachicola Bay, Florida, U.S.A. Limnolo gy and Oceanography 47(3):683 697. Claassen, Cheryl. 1998 Shells. Cambridge: Cambridge University Press. 2008 Shell Symbolism in Pre Columbian North America. Early Human Impact on Megamollsuscs :37 43. Costa Pierce, Barry A. 1987 Aquacult ure in Ancient Hawaii. BioScience 37(5):320 331. Crisp, D.J. 1967 Chemical Factors Inducing Settlement in Crassostrea virginica Journal of Animal Ecology 36(2):329 335. Crook Jr, Morgan R. 1992 Oyster Sources and their Prehistoric Use on the Geo rgia Coast. Journal of Archaeological Science 19 (5):483 496. Custer, Jay F., and Keith R. Doms. 1990 Analysis of Microgrowth Patterns of the American Oyster ( Crassostrea virginica ) in the Middle Atlantic Region of Eastern North America: Archaeological Applications. Journal of Archaeological Science 17(2):151 160.
128 Dame, R. F. 1972 The Ecological Energies of Growth, Respiration and Assimilation in the Intertidal American Oyster Crassostrea virginica Marine Biology 17(3):243 250. 2009 Shifting Through Time: Oysters and Shell Rings in Past and Present Southeastern Estuaries. Journal of Shellfish Research 28(3):425 430. 2012 Ecology of Marine Bivalves: An Ecosystem Approach CRC Press, Boca Raton. Dietler, Michael, and Brian Hayden 2001 Digesting the Feast: Good to Eat, Good to Drink, Good to Think: An Introduction. In Feasts: Archaeological and Ethnographic Persp ectives on Food, Politics, and Power, edited by Michael Dietler and Brian Hayde n, p p. 1 20. The University of Alabama Press, Tuscaloosa. Dietler, Michael 2001 Theorizing the Feast: Rituals of Consumption, Co mmensal Politics, and Power in African Contexts. In Feasts: Archaeological and Ethn ographic Perspectives on Fo od, Politics, and Power, edited by Michael Dietler and Brian Hayden, pp. 65 114. The University of Alabama Press, Tuscaloosa. Donop, Mark C. 2015 Palmetto Mound (8LV2). In Lower Suwannee Archaeological Survey 2013 2014 Shell Mound and Cedar Key Tracts. Technical Report 23, Laboratory of Southeast ern Archaeology, Department of Anthropology, University of Florida. Doucet, Julie Ann 2012 Oysters and Catfish: Resource Exploitation at Rollins Shell Ring, Ft. George Island, Flor ida. Louisiana State University, Baton Rouge. Dragovich, Alexander and John A. Kelly 1964 Ecological Observations of Macro Invert ebrates in Tampa Bay, Florida 1961 1962. Bulletin of Marine Science of the Gulf and Caribbean 14(1):74 102. Eble, Albert F., and Robert Scro 1996 General Anatomy. In The Eastern Oyster: Crassostrea virginica eds Victor S. Kennedy, Roger I.E. Newell, and Albert F. Eble. Pp. 19 73. Erlandson, Jon M. 1988 The Role of Shellfish in Prehistoric Economies: A Protein Perspective. A merican Antiquity 53(1) :102 109.
129 Erlandson, J.M., T.C. Rick, and R.L. Vellanoweth 2004 Human Impacts on Ancient Environments: Northern Chan nel Islands. In Voyages of Discovery: The Archaeology of Islands, edited by S.M. Fitzpatrick, pp. 51 83. Praeger, New York. Erlandson, J.M., T.C. Rick, J.A. Estes, M.H. Graham, T.J. Braje, and R. L. Vellanoweth 2005 Sea Otters, Shellfish, and Hum ans: A 10,0 00 Year Record from San Miguel Island, California. In Proceedings of the Sixth California Islands Symposium edited by D. Garcelon and C. Schwemm, pp. 9 21. National Park Service Technical Publication, Arcata, California. Erl andson, Jon M., and Torben C. Rick 2008 Archaeology, Marine Ecology, and Human Impa cts on Marine Environments. In Human Impacts on Ancient Marine Ecosystems: A Global Perspective edited by Torben C. Rick and Jon M. Erlandson, pp. 1 19. U niversity of California Press, Berkeley. Goodbred, S.L. and A.C. Hine. 1995 Coastal Storm Deposition: Salt Marsh Response to a Severe Extratropical Storm. Geology 23:679 682. Goodbred, Steven L., Albert C. Hine, and Er ic E. Wright 19 98 Sea Level Change and Storm Surge Deposition in a Late Holocene Florida S alt Marsh. Journal of Sedimentary Research 68:240 252. Goodell, H.G., and D.S. Gorsline 1961 The Hydrography of Apalachicola and Florida Bays, Florida. No. GFDI CONT RIB 1. Florida State University, Tallahassee. Grier, Colin 2014 Landscape Construction, Ownership and Soci al Change in the Southern Gulf Islands of British Columbia. Canadian Journal of Archaeology 38:211 249. Gu nther, R.T. 1 897 The Oyster Culture of the Ancient Romans C ambridge: Cambridge University Press. 1957 Oysters. Geological Society of America Memoirs 67(1):1129 1134. Harding, Juliana M., Howard J. Spero, Roger Mann, Gregory S. Herbert, and Jennifer L. Sliko. 2010 Reconstructing Early 17th Century Estuarine Drought Conditions from Jamestown Oysters. Proceedings of the National Academy of Sciences 107 (23): 10549 10554.
130 Heilmayer, Olaf, Julian Digialleonardo, Lianfen Quan, Guritno Roesijadi 2008 Stress Tolerance of a Subtropical Crassostrea virginica Population to the Combined Effects of Temperature and Salinity. Estuarine, Coastal, and Shelf Science 79:197 185. Hine, Albert C., Daniel F. Belknap, Joan G. Hutton, Eric B. Osking, and Mark W. Evans 1988 Recent Geological History and Modern Sedimentary Processes along an Incipient, Low Energy, Epicontinental Sea Coastline: Northwest Florida. Journal of Sedimentary Research 58:567 579. Hopkins, S.H. Jackson, Jeremy B.C., Michael X. Kirby, Wolfgang H. Berger, Karen A. Bjornal, Louis W. Botsford, Brocue J. Bourque, Roger H. Bradbury, Richard Cooke, Jon Erlandson, James A. Estes, Terence P. Hughes, Susan Kidwell, Carina B. Lange, Hunter S. Lenihan, John M. Pando lfi, Charles H. Peterson, Robert S. Stenec, Mia J. Tegner, Robert R. Warner 2001 Historical Overfishing and the Recent Collapse of Coastal Ecosystems. Science 27(293):629 637. Kennedy, Victor. 1996 The Ecological Role of the Eastern Oyster, Cras sostrea Virginica with Remarks on Disease. Journal of Shellfish Research 15 (1):177 183. Kent, Bretton. 1988 Making Dead Oysters Talk: Techniques for Analyzing Oysters from Archaeological Sites Annapolis: Maryland Historical Tr u st, Historic St. Mary's City, Jefferson Patterson Park and Museum. Kidder, Tristram R. 2006 Climate Change and the Archaic to Woodland Tran sition (3000 2600 cal B.P.) in the Mississippi River Basin. American Antiquity 71:195 231. Kilgen, Ronald H., and Ronald J. Dugas. 1989 The Ecology of Oyster Reefs of the Northern Gulf of Mexico: An Open File Report US Department of the Interior, Fish and Wildlife Service, Research and Development, National Wetlands Research Center. Kir by, Michael Xavier. 2004 Fishing Down the Coast: Historical Expansion and Collapse of Oyster Fisheries Along Continental Margins. Proceedi ngs of the National Academy of Sciences of the United States of America 101(35):13096 13099
131 Korringa, Pieter 1952 Recent Advances in Oyster Biology. The Quarterly Review of Biology 27(3):266 308. 1976 Farming the Cupped Oysters of the Genus Cr assostrea: A Multidisciplinary Treaties. Elevier Scientific Publishing, New Yor k. Lawrence, David R. 1988 Oysters as Geoarchaeologic Objects. Geoarchaeology 3(4):267 274. Lepofsky, Dana and Megan Caldwell 2012 Indigenous Marine Resource Management o n the Northwest Coast of North America. Ecological Processes 2:1 12. Lepo fsky, Dana, Nicole F. Smith, Nathan Cardinal, John Harper, Mary Morris, Gitla (Elroy White), Randy Bouchard, Dorothy I.D. Kennedy, Anne K. Salomon, Michelle Puckett, and Kirsten Rowell 2015 Ancient Shellfish Mariculture on the Northwest Coast of North Ame rica. American Antiquity 80(2):236 359. Lightfoot, K.G., R. M. Cerrato, and H.V.E. Wallace 1993 Prehistoric Shellfish Harvesting Strategies: Implic ations from the Growth of Soft Shell Clams ( Mya arenaria ). Antiquity 37:358 369. Livingston, R. J., F.G. Lewis, G. C. Woodsum, X. F. Niu, B. Galperin, W. Huang, J. D. Cristensen, M. E. Monaco, T. A. Battista, C. J. Klein, R. L. Howell IV, and G. L. Ray 2000 Modelling Oyster Population Response to Variation in Freshwater Input. Estuarine, Coast al and Shelf Science 50: 655 672. Mannino, Marcello A., and Kenneth D. Thomas 2002 Depletion of a Resource? The Impact of Prehistoric Human Foraging on Intertidal Mollusc Communities and Its Significance for Human Settlement, Mobility and Di spersal. World Archaeology 33(3):452 474. Mattson, Robert A. 2002 A Resource Based Framework for Establishing Freshwater Inflow Requirements for the Suwannee River Estuary. Estuaries 25 (6): 1333 1342. McFadden, Paulette S. 2015 Late Holocene Coastal Evolution and Human O ccupation on the Northern Gulf Coast of Florida, Horseshoe Cove, Dixie County, Florida. Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville.
132 McFadden, Paulette S., and Andrea Palmi otto 2013 Archaeological Investigations at Ehrbar (8LV282), Levy County, Florida Technical Report 17. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. McKee, Alexander 1969 Farming th e Sea New York: Thomas Y. Crowell Company. McNulty, J. Kneeland, William N. Lindall Jr., and James E. Sykes. 1972 Cooperative Gulf of Mexico Estuarine Inventor y and Study, Florida: Phase I, Area Description. Milanich, Jerald T. 1994 Archaeo logy of Precolumbian Florida Gainesville: University Press of Florida. 2002 Weeden Island Cultures. The Woodland Southeast. University of Alabama Press, Tuscaloosa: 352 372. Milanich, Jerald T., Ann S. Cor d ell, Vernon J. Knight, Jr., Timot hy A. Kohler, and Brenda J. Sigler Lavelle 1984 McKeithan Weeden Island: The Culture of Northern Florida, A.D. 200 900. New York: Academic Press. Nash, Colin E. 2011 The History of Aquaculture Wiley Blackwell, Ames, Iowa. National Research Co uncil (U.S.) Committee on Assessment of Technology and Opportunities for Marine Aquaculture in the United States 1992 Marine Aquaculture: Opportunities for Growth. Washington D.C.: National Academy Press. Patillo, M.E., T.E. Czapla, D.M. Nelson, an d M.E. Monaco 1997 Distribution and Abundence of Fishes and Invertebrates in Gulf of Mexico Estuaries. Volume II. Species Life History Summaries. ELMP Rep. 11. Pearse, A.S. and Gordon Gunter 1957 Salinity. Geological Society of America Memoirs 1 :129 158. Rakov, V.A. and D.L. Brodianski 2007 Ancient Oyster Farming in the Boisman Culture of the Primorye Neolithic. Archaeology, Ethnology, and Anthropology of Eurasia 31(1):39 43. Rakov, V.A. and D.L. Brodianski 2010 Oyster Cultivation and A rchaeology as Producing Activities. Archaeology Ethnology & Anthropology of Eurasia 38(1):26 31.
133 Randall, Asa R., Micah P. Mones, and Kenneth E. Sassaman 2010 Visit Shell City: Another Coastside Attraction. Paper presented at the 67 th Annual Meeting of the Southeastern Archaeological Conference, Lexington, Kentucky. Rick, Torben C., and Jon M. Erlandson, eds 2008 Human Impacts on Ancient Marine Ecosystems: A Global Perspective Berkley: University of California Press. Rick, To rben C., and Rowan Lockwood. 2013 Integrating Paleobiology, Archeology, and History to Inform Biological Conservation. Conservation Biology 27(1):45 54. Russo, Michael. 1991 A Method for the Measurement of Season and Duration of Oyster Collection : Two Case Studies from the Prehistoric South East US Coast. Journal of Archaeological Science 18(2):205 221. Sampson, Christina Perry 2015 Oyster Demographics and the Creation of Coast al Monuments at Roberts Island Mound Complex, Florida. S outheastern Archaeology 34(1):84 94. Sassaman, Kenneth E., Paulette S. McFadden, and Micah P. Mons 2011 Lower Suwannee Archaeological Survey 2009 2010: Investigations at Cat Island, Bird Island, and Richards Island. Technical Report 10. Laborato ry of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Sassaman, Kenneth E. 2012 Futurologists Look Back, Archaeologies: Journ al of the World Archaeological Congress 250 2268. Sassaman, Kenn eth E., Andrea Palmiotto, Ginessa Mahar, Micha P. Mon s, and Paulette S. McFadden 2013 Archaeological Investigations at Shell Mound (8LV42), Levy County, Florida : 2012 Testing Technical Report 16, Laboratory of Southeastern Archaeology, Dep artment of Anthropology, University of Florida. Sassaman, Kenneth E. and Neill J. Wallis 2015 Crisis of Opportunity: How Rapid Sea Level Rise at A.D. 200 300 Relates to the Structural Realignment of Communities of the Northern Gulf Coast of Florida and Beyond. Paper presented at the 47 th Annual Meeting of the Canadian Archaeology Association, Newfoundland.
134 Sassaman, Kenneth E., Ginessa J. Mahar, Mark C. Donop, Jessica A. Jenkins, Anthony Boucher, Christina I. Oliveira, Joshua M. Go odwin 2015 Lower Suwannee Archaeological Survey 2013 2014 Shell Mound and Cedar Key Tracts. Technical Report 23, Laboratory of Southeast ern Archaeology, Department of Anthropology, University of Florida. Sassaman, Kenneth E., Neill J. Wallis Paulette S. McFadden, Ginessa J. Mahar, Jessica A. Jenkins, Mark C. Donop, Michah P. Mon s, Andrea Palmiotto, Anthony Boucher, Joshua M. Goodwin, and Cristina I. Oliveira 2016 Keeping Pace with Rising Sea: The First Six Years of the Lower Suwannee Archaeological Survey, Gulf Coastal Florida. Journal of Island and Coastal Archaeology DOI: 10.1080/15564894.2016.1163758. Saunders, Rebecca W., and Michael Russo 2011 Coastal Shell Middens in Florida: A View from the Archaic. Quaternary Int ernational 239:38 50. Seavey, J.R., W.E. Pine III, P. Frederick, L. Strumer, and M. Berrigan 2011 Ecosphere 2(10):1 14. Schmidt, Daniel and Dexter Haven 2004 A Pilot Study on the Origins of Oysters at James Fort. The Journal of Jamestown Rediscovery Center 2:1 6. Schmidt, Nancy and Mark E. Luther 2002 ENSO Impacts on Salinity in Tampa Bay, Florida. Estuaries 24(5):976 984. Shumway, Saundra E. 1996 Natural En vironmental Factors. In The Eastern Oyster: Crassostrea virginica eds Victor S. Kennedy, Roger I.E. Newell, and Albert F. Eble. Pp. 467 513. Speilmann, Katherine A. 2002 Feasting, Craft Specialization, and the Ritual Mod e of Production in Small Scale Societies. American Anthropologist 104(1):195 207. Stapor, F.W., T.D. Mathews, and F.E. Lindofors Kearns 1991 Barrier Island Progradation and Holocene Sea Level History in Southwest Florida. Journal of Coastal Research 7:815 838. Supan, John 2002 Extensive Culture of Crassostrea virginica in the Gulf of Mexico Region. Southern Regional Aquaculture Center.
135 Swalding, P. 1976 Changes Induced by Human Exploitation in Prehistoric Shellfish Populations. Mankind 10(3):15 6 162. Thomas, David H., and Matthew C. Sanger (editors) 2011 Trend, Tradition, and Turmoil: What Happened in the Southeastern Archaic? Anthropological Papers 93. American Museum of Natural History, New York. Thompson, Victor D., and John E Worth 2011 Dwellers by the Sea: Native American Adaptations along the Southern Coasts of Eastern North America. Journal of Archaeological Research 19:51 101. Twiss, K.C. 2008 Transformations in an Early Agricultural Society: Feasting in the Southern Levantine Pre Pottery Neolithic. Journal of Anthropological Archaeology 27:418 442. Walker, Karen J., Frank W. Stapor, Jr., and William H. Marquardt 1995 Archaeological Evidence for a 1750 1450 BP Higher Than Present Sea Level Journal of Coastal Research Special Issue 17:205 218. Wallis, Neill J. 2011 The Swift Creek Gift: Vessel Exchange on the Atlantic Coast. Tuscaloosa: University of Alabama Press. Wallis, Neill J., and Meggan E. Blessing 2015 Big Feasts and Small Scale Foragers: Pit Features as Feast Events in the American Southeast. Journal of Anthropological Archaeology 39:1 18. Wallis, Neill J., Paulette S. McFadden, Hayley M. Singleton 2015 Radiocarbon Dating the Pace of Monument Constr uction and Village Aggregation at Garden Patch: A Ceremonial Center on the Florida Gulf Coast. Journal of Archaeological Science: Reports 2:507 516. Wells, Harry W. 1961 The Fauna of Oyster Beds, with Special Referen ce to the Salinity Factor. Ecological Mongraphs 31(3):239 266. Whitaker, Adrian R. 2008 Incipient Aquaculture in Prehistoric Californ ia? Long Term Productivity and Sustainability vs. Immediate Returns for the Harvest of Marine Inverteb rates. Journal of Archaeological Science 35:1114 1123.
136 Willey, Gordon R. 1949 Archaeology of the Florida Gulf Coast 113 Washington D.C.: Smithsonian Institution Press. Wright, Eric E., Albert C. Hine, Steven L. Goodbred, Jr., and Stanley D. Locker 2005 The Effect of Sea Level and Climate Change on the Development of a Mi xed Siliciclasti Carbonate, Deltaic Coastline: Suwannee River, Florida, U.S.A. Journal of Sedimentary Research 75:621 635.
137 BIOGRAPHICAL SKETCH Jessica Jenkins attende d the College of William and Mary, graduating with her B.A. in anthropology in 2012 As an undergraduate, Jessica worked for three years with Dr. Martin Gallivan at aboriginal sites on the York River, sparking her interest in subsistence practices, specifi cally shellfish gathering and use. After graduating from William and Mary, Jessica worked with Dr. Frederick Smith at historic archaeological sites in Barbados and was employed with several cultural resource management firms, working in Louisiana and Virginia. Jessica began her graduate ca reer at the University of Florida in 2013 with Dr. Kenneth Sassaman, continuing her research on co ing Coast.