THE ARCHAEOLOGY OF NORTHEAST FLORIDA SPRINGS By JASON MICHAEL GERALD A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2015
4 ACKNOWLEDGMENTS First and foremost, I must thank my committee members. Susan Gillespie has had a profound influence on m y g raduate (and undergraduate) car eer. Back in the spring of 2004, after I completed her undergraduate archaeological theory course, she asked me not if I was considering graduate school, but where I would be applying. Her backing was critical to my acceptance in the program at the University of Tennessee, and to my return to Gainesville for a Ph.D. I am truly grateful for her steadfast support a nd for advice and critique that is always insightful. Mark Brenner and Willie Harris have been encouraging and helpful committee member knowledge of Florida paleolimnology h as been illuminating , and his critical perspective much appreciated. I am thankful for his generosity with advice, equipment, and laboratory space . Willie taught me to read the soil in ways I had not thought possible . I have thoroughly enjoyed our conversations about pedology and archaeology , and found his interest encouraging a nd his ability to simplify complex concepts im mensely helpful. Ken Sassaman taught the first anthropology class I took as a freshman at the University of Florida. His enthusiasm for the subject was palpable and inspired me to explore further courses in the discipline and to eventually switch my major from mathematics. Ken also directed my first field experience with the St. Johns Archaeological Field School and was an avid supporter of my venture to Tennessee and my eventual return to the University of Flor ida. Ken shepherded me through t he Ph.D. process and was always available to offer advice, encouragement, resources, and research opportunities. I am truly fortunate to have had him as an advisor and friend, and could not have completed this work without h is support.
5 M y fello w graduate students bolstered my spirits and gave much inspiration along the way. Meggan Blessing and Asa Randall were TAs at the St. Johns Archaeological Field S chool where I first learned how to hold a trowel. In many ways they have b analyze archaeological data in innovative ways is inspiring, as i s his meticulous attention to detail . I have enjoyed many conversations with Asa about the St. Johns River Archaic . H is influence runs through out this dissertation. intolerance of sloppy scholarship have provided a compass for my own research. Zack Gilmore has been a frequent collaborator and reliable friend whose consistency and work ethic I try to emula te. Others who have helped ease the trials and tribulations of grad school include Chris Altes, Josh Goodwin, Paulette McFadden, Ginessa Mahar, and Isaac Shearn, among many others . A number of field school students and lab volunteers contr ibuted to this dissertation by sorting, cataloging, and analyzing materials. Anthony Boucher did the heavy lifting with the Silver Springs lithic assemblage and went well beyond what was required of him . Catherin e Au st likewise sorted much of the Salt Spri ngs assemblage. I am also indebted to the stu dents and TAs o f the S t . Johns Archaeological Field School, who generated much of the data pertaining to Silver Glen Springs. T he field crews who conducted the surveys of Otter, Silver, and Weeki Wachee spring s were professional and hard working . I appreciate the efforts of Mark Winburn, Randy Crones, Kris Hall, Micah Mones, Ginessa Mahar, Zack Gilmore, Shaun West, Brad Lanning , and the contributions each made to this research .
6 Triel Lindstrom, Archaeologist with the Florida Department of Environmental Protection, Division of Recreation and Parks, Bureau of Natural and Cultural Resources facilitated the fieldwork at Silver and Weeki Wachee springs, while Dave Dickens , Chief of the Bureau of Administrative and Oper ations at the Suwannee River Water Management District enabled the research at Otter Springs. I am indebted to both for their support. My thanks to the Juniper Hunt Club of Louisville, Kentucky for granting access to their property adjacent to Silver Glen Springs and for supporting our research there. I also extend my thanks to Jon Endonino of Eastern Kentucky University, who performed the lithic provenance determinations presented in Chapter 5 and graciously shared data from other sites. Likewise, I thank John Jaeger of the Department of Geological Sciences at the University of Florida for loaning the vibracoring unit used at Silver Glen Springs. Funding for this research was provided by the Hyatt and Cici Brown Endowment for Florida Archaeology and a John W. Griffin Student Grant from the Florida Archaeological Council. The final stages of writing this dissertation were facilitated by a University of Florida Graduate School Dissertation Award. I am deeply indebted to my family. My parent s , Michael and Cat herine never doubted that I could finish. Meanwhile, my brothers, Nicholas and Patrick, have tolerated extended archaeological discussions with genuine interest and have provided many opportunities for distraction from academia. My sister, Erica, continues to be a font of inspiration and resolve. More recently, m y two sons have brought immeasurable joy to my life, and helped me to put things in perspective when the stress became overwhelming. Finally , my wife, Amanda, has been a true partner in what has been a somewhat lengthier
7 graduate career than originally anticipated. She has endured the best and worst of me with equal measures of patience and grace. I am in awe of the creative capacity of her lo ve as an artist, mother, and wife. Her unending support has been a source of comfort and provided the space for me to complete this work. I am forever grateful to her.
8 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ .......... 11 LIST OF FIGURE S ................................ ................................ ................................ ........ 13 ABSTRACT ................................ ................................ ................................ ................... 17 CHAPTER 1 INTRODUCTION: SMOKE ON THE WATER ................................ ......................... 19 ................................ ................................ ................ 20 Springs and the Archaeology of Florida ................................ ................................ .. 24 Outline and Organization ................................ ................................ ........................ 30 2 ................................ ................ 36 Springs and Aquifers ................................ ................................ ............................... 38 Geological History of the Florida Peninsula ................................ ............................ 42 Florida Basement ................................ ................................ ............................. 43 Carbonate Platform ................................ ................................ .......................... 44 Siliciclastic sediments ................................ ................................ ....................... 46 Karst Terrain in Florida ................................ ................................ ........................... 48 The Evolution of Karst Landscapes ................................ ................................ .. 49 Karst Hydrology ................................ ................................ ................................ 53 Recharge ................................ ................................ ................................ ... 53 Flow ................................ ................................ ................................ ........... 54 Discharge ................................ ................................ ................................ ... 57 Quaternary Environments of Florida and the St. Johns River Valley ...................... 60 Late Pleistocene conditions in Florida ................................ .............................. 62 Holocene Conditions ................................ ................................ ........................ 66 St. Johns River Valley ................................ ................................ ............................. 71 Springs of Florida and the St. Johns River ................................ .............................. 76 3 SPRING ORIGINS ................................ ................................ ................................ .. 89 The Mount Taylor Period in Northeast Florida ................................ ........................ 9 4 The Early Mount Tayl or Phase (7400 5700 cal B.P.) ................................ ...... 98 Thornhill Lake Phase (5700 4600 cal B.P.) ................................ ................... 102 Discussion ................................ ................................ ................................ ...... 106 Events and Non E vents ................................ ................................ ........................ 107 Spring Chronology I: Modeling ................................ ................................ .............. 113
9 Spring Chronology II: Archaeology and Hydrology ................................ ............... 118 Salt Springs ................................ ................................ ................................ .... 119 Recent Archaeological Investigations ................................ ...................... 120 Chronology and Depositional History ................................ ....................... 123 Hydrological History ................................ ................................ ................. 125 Silver Glen Springs ................................ ................................ ......................... 127 Archaeological Overview ................................ ................................ ......... 128 Shell Deposition at Locus A ................................ ................................ ..... 129 Hydrological History ................................ ................................ ................. 131 Discussion ................................ ................................ ................................ ............ 134 4 TRANSFORMATIVE HISTORY AT SILVER GLEN SPRINGS ............................. 153 Sacred Site, Sacred Spring ................................ ................................ ................... 155 Situating the Silver Glen Springs Complex ................................ ........................... 159 Late Pleistocene and Holocene Antecedents ................................ ....................... 163 Thornhill Lake Phase ................................ ................................ ............................ 166 Orange Period ................................ ................................ ................................ ...... 172 St. Johns Peri od ................................ ................................ ................................ ... 179 Discussion: Transformative Histories at Silver Glen ................................ ............. 184 Was Silver Glen Springs Sacred? ................................ ................................ ......... 192 5 SPRINGS ON THE MOVE ................................ ................................ ................... 201 Silver Springs ................................ ................................ ................................ ........ 204 Archaeological Sites at Silver Springs ................................ ............................ 205 Least Cost Paths ................................ ................................ ............................ 213 Lithic Provenance Determination ................................ ................................ .... 219 Debitage Analysis ................................ ................................ ........................... 222 The Silver Springs Gateway? ................................ ................................ ......... 223 Beyond the St. Johns ................................ ................................ ............................ 226 Summary and Discussion ................................ ................................ ..................... 229 6 SPRINGS ETERNAL ................................ ................................ ............................ 250 The Past in Springs Conservation Narratives ................................ ....................... 254 Entanglement ................................ ................................ ................................ ........ 258 Humans Depend on Springs ................................ ................................ ................. 261 Springs Depend on Other Things ................................ ................................ ......... 265 Springs Depend on Humans ................................ ................................ ................. 267 Springs Entangled ................................ ................................ ................................ 274 Conclusion s and Future Directions ................................ ................................ ....... 281 APPENDIX A STRATIGRAPHIC DESCRIPTIONS ................................ ................................ ..... 293 B SILVER SPRINGS LITHIC ARTIFACT ATTRIBUTES ................................ .......... 299
10 LIST OF REFERENCES ................................ ................................ ............................. 349 BIOGRAPH ICAL SKETCH ................................ ................................ .......................... 391
11 LIST OF TABLES Table page 2 1 Spring magnitude and equivalent discharge rate ................................ ................ 88 5 1 Inventory of pre Columbian cultural materials recovered from STPs excavated at Silver Springs (8MR93 and 8MR 1082) ................................ ........ 232 5 2 Frequency of lithic artifacts at selected archaeological sites adjacent to springs in the St. Johns River valley ................................ ................................ . 232 5 3 Provenance of lithic debitage from Silver Springs (8MR93) ............................. 233 5 4 Provenance of lithic debitage from Silver Glen Springs (8MR123) ................... 233 5 5 Provenance of lithic debitage from Lake Monroe Outlet Midden (8VO53) ........ 234 5 6 Provenance of lithic debitage from the T hornhill Lake Complex (8VO58 60) ... 234 5 7 Provenance of lithic tools from Silver Springs (8MR93) ................................ .... 235 5 8 Provenance of lithic tools from Silver Glen Springs (8MR123) ......................... 235 5 9 Provenance of lithic too ls from Lake Monroe Outlet Midden (8VO53) .............. 236 5 10 Provenance of lithic tools from the Thornhill Lake Complex (8VO58 60) ......... 236 5 11 Summary results of debitage analysis from Silver Springs (8MR93) ................ 237 5 12 Summary results of debitage analysis from Silver Glen Springs (8MR123) ..... 237 5 13 Summary results of debitage analysis from the Lake Monroe Outlet Midden (8VO53) ................................ ................................ ................................ ............ 238 5 14 Summary results of debitage analysis from the Thornhill Lake Complex (8VO58 60) ................................ ................................ ................................ ...... 238 5 15 Inventory of Cultural Materials Recovered from STPs excavated at Otter Springs (8GI12) ................................ ................................ ................................ 239 A 1 Stratigraphic units documented in 2009 test excavations and cores at Salt Springs (8MR2322) ................................ ................................ .......................... 294 A 2 Description of vibracore BC01 at Silver Glen Springs (8LA1 W) ...................... 295 A 3 Description of vibracore BC02 at Silver Glen Springs (8LA1 W) ...................... 295 A 4 Description of vibracore BC03 at Silver Glen Springs (8LA1 W) ...................... 296
12 A 5 Description of vibracore EF01 at Silver Glen Springs (8LA1 W) ....................... 296 A 6 Description of vibracore EF02 at Silver Glen Springs (8LA1 W) ....................... 297 A 7 Description of vib racore EF03 at Silver Glen Springs (8LA1 W) ....................... 298 B 1 Lithic debitage provenance and attribute data ................................ .................. 300 B 2 Key to headings and abbreviations for lithic debitage attribute data. ................ 342 B 3 Lithic tool provenance and attribute data. ................................ ......................... 343
13 LIST OF FIGURES Figure page 1 1 Aerial photograph of Silver Glen Springs. Photo by Overexposed Aerial Photography ................................ ................................ ................................ ....... 35 2 1 The home of Mae Rose Williams descends into the crater, May 9, 1981. Photo by Barbara Vitaliano, Orlando Sentinel ................................ .................... 81 2 2 Conceptual diagram illustrating the hydrology of the Floridan Aquifer System. Image courtesy of the St. Johns River Water Management District .................... 82 2 3 Generalized extent and degree of confinement of the Floridan Aquifer System ................................ ................................ ................................ ............... 83 2 4 Generalized extent of areas recharging the Floridan Aquifer System in Florida ................................ ................................ ................................ ................ 84 2 5 Physiographic provinces in the vicinity of the St. Johns River valley, northeast Florida. Divisions follow those of Cooke (1939) ................................ .. 85 2 6 Major river basins and surface water features of the St. Johns and Ocklawaha River basins ................................ ................................ ..................... 86 2 7 Distribution of karst districts (following Brooks 1981) and springs recorded in the 2011 inventory of the Florida Department of Environmental Protection. Note: four submarine springs are not shown ................................ ...................... 87 3 1 Kissengen Spring, Polk County, Florida in 1894. Photo courtesy of the State Archives of Florida, Florida Memory , https://floridamemory.com/items/show/117843 ................................ ................ 139 3 2 Kissengen Spring ca. 1947, three years before it went dry. Photo courtesy of the Florida Geological Survey Photo Archive ................................ ................... 140 3 3 Kissengen Spring, April 2006. Photo courtesy of the U.S. Geological Survey .. 141 3 4 Regional culture historical timescale (after Anderson and Sassaman [2012:Table 1 1]) and local traditions in the St. Johns River valley .................. 142 3 5 Distribution of Mount Taylor era sites in the St. Johns River valley, highlighting locations mentioned in the text. Also shown are two pond burial sites, Windover and Gauthier ................................ ................................ ........... 143 3 6 Results of the GIS model, showing areas of potential spring flow under conditions of l ower than present potentiometric surface of the Florida Aquifer . 144
14 3 7 Location of Lake George and surrounding springs ................................ ........... 145 3 8 Salt Springs quadrangle (above) and position o f 2009 NPS and UF excavations relative to the pool of Salt Springs (below) ................................ ... 146 3 9 Topographic map of subaqueous portion of the S alt Springs site (8MR2322), showing locations of University of Florida excavation trench and cores and National Park Service excavation block. Elevation relative to arbitrary datum . 147 3 10 Grid west profile of trench at 8MR2322. A) Composite photograph. B) Profile drawing ................................ ................................ ................................ ............. 1 48 3 11 Deposits observed in percussion cores at Salt Springs, Core 1 (left) and Core 2 (right), in relation to trench stratigraphy. Elevation relative to arbitrary datum ................................ ................................ ................................ ............... 149 3 12 Map of the Silver Glen Complex, highlighting locations with Mount Taylor era deposits. Site footprints courtesy of Asa Randall (2014b) ................................ 150 3 13 Deposits observed in vibracores at Silver Glen Springs. Elevation relative to NAVD 1988 ................................ ................................ ................................ ....... 151 3 14 Aerial photograph of Silver Glen Springs and Run (above), showing location of vibracores. Cross section of Silver Glen Run (below), showing vibracore transect EF in relation to surrounding landfo rms ................................ .............. 152 4 1 View of the northwest face of Mount Shasta. Photo courtesy of the U.S. Geological Survey (http://3dparks.wr.usgs.gov/shasta/html2/shasta125.htm) .. 197 4 2 Location of the archaeological sites comprising the Silver Glen Co mplex, on .. 198 4 3 Aerial photograph of Silver Glen Spr ings, highlighting the archaeological deposits surrounding the spring and run ................................ .......................... 199 4 4 Current topography (bottom) and reconstructe d, pre mining topography (top) of the Silver Glen Complex. Reconstruction after Randall (2014b) ................... 200 5 1 Location of spring vents at Silver Springs. Image courtesy of the St. Johns River Water Management District (http://floridaswater.com/springs/images/silver_ventlocation_aerial_lrg.jpg) .... 240 5 2 Area surveyed by the LSA and previously recorded archaeological sites, Silver Springs, Florida ................................ ................................ ...................... 241 5 3 Results of the archaeological survey conducted at Silver Springs, Florida ....... 242
15 5 4 Results of the least cost analysis f rom lithic quarry clusters to Silver Springs (8MR93) ................................ ................................ ................................ ........... 243 5 5 Results of the least cost analysis from lithic quarry clusters to Silver Glen Springs (8MR123) ................................ ................................ ............................ 244 5 6 Results of the least cost analysis from lithic quarry clusters to the Lake Monroe Outlet Mid den (8VO53) and the Thornhil Lake Complex (8VO58 60) 245 5 7 Comparison of lithic source use for Silver Springs (8MR93), Silver G len Springs (8MR123), Lake Monroe Outlet Midden (8VO53), and the Thornhill Lake Complex (8VO58 60) ................................ ................................ .............. 246 5 8 Aggregated least cost paths, limited to those quarry clusters comprising greater than 5.0% of the lithic assemblage for a given site .............................. 247 5 9 Subsection location of the LSA Otter Springs survey ................................ .......................... 248 5 10 Shovel test pits excavate d at Otter Springs. All STPS were positive ................ 249 6 1 Railroad and steamboats at Silver Springs in 1886. Photograph by George Barker, courtesy of the Library of Congress Prints and Photographs Division, Washington, D.C. ................................ ................................ ............................. 285 6 2 Interior of the health spa at White Springs, Hamilton County, Florida, ca. 1912. Courtesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/149898 ................................ ................ 286 6 3 The early twentieth century resort at DeLeon Spring. Courtesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/158585 ................................ ................ 287 6 4 Weeki Wachee Spring in 1952. Courtesy of the State Archives of Florida, Florida Memory, http://floridamemory.com/items/show/149889 ........................ 288 6 5 Alexander Springs, circa 1947. Courtesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/125603 ...................... 289 6 6 Representative shovel test pit profiles from Otter Springs. Clockwise from top left: STPs 1, 7, 14, 16. Note: photos are not to scale and were taken at an oblique a ngle ................................ ................................ ................................ .... 290 6 7 Interpolated depth of modern infilling or subsurface disturbance in the vicinity of Weeki Wachee Springs. Modified 291
16 6 8 County median per capita income and the distribution of springs in Florida. Springs data from the Florida Department of Environmental Protection (2012). Four offshore springs are not shown ................................ .................... 292
17 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE ARCHAEOLOGY OF NORTHEAST FLORIDA SPRINGS By Jason Michael Gerald December 2015 Chair: Kenneth E. Sassaman Maj or: Anthropology This dissertation investigate s the archaeological significance of springs in the St. Johns River valley of northeast Florida . A rchaeologists have long f ocus ed ecological capacities and have failed to recognize the importance o f these places to ancient Floridians , beyond subsistence concerns . Meanwhile, contemporary conservation narratives, rarely informed by archaeological knowledge, rely on a simplistic notion of eternal, pristine springs that likewise downplays their past significance. Within the St. Johns River valley , springs have been central to explanations of the appearance of shell mounds during the Mount Taylor period (7400 4600 cal B.P.) . It has been argued that the onset of spring flow was a necessary precondition for t he productive hydric habitats that were exploited by Mount Taylor people. I interrogate this model and develop an alternative approach that foregrounds springs as places of social interac tion with deep historical import . T he geologic context of Florid provides a framework for understanding the ir abundance and distribution , and the factors affecting spring flow . G eog raphic Information System (GIS) based hydrological model ling and cores extracted from near -
18 shore spring deposits indicate that t hey began flowing several millennia before shell mounds first appear in the region . A rchaeological testing of spring side sites likewise demonstrates that the earliest activities at spring s did not involve the deposition of shell. Rather, lithic provenance and technological studies, and least cost modelling of the movement of lithic raw materials from western Florida into the St. Johns River valley demonstrate that springs facilitated regional interaction and exchange. The archaeological record of northeast Florida springs suggests an alternative explanation for their significance. Springs were variously used as habitation sites, repositories for the dead, and the site of regional gatherings. Throughout th is varied history of use, s prings have been t he site of social gatherings th at connected far flung peoples , much as they are today . People were drawn to springs for both their aesthetic qualities and the material legacies left by past visitors. This has impl ications for contemporary conservation, and suggests that both personal experiences and an appreciation of the historical significance of springs are key to mobilizing public sentiment and political will.
19 CHAPTER 1 INTRODUCTION: SMOKE ON THE WATER I sat on the dock and watched a steady parade of boats putter past: motorboats, houseboats, pontoons, the occasional party barge. All were converging on a spot just called locally, is a large artesian spring tha t draws some 65 million gallons of water a day out of the Floridan Aquifer. The spring water spouts up from two main vents to form a circular pool filled with azure water that is 73 degrees year round. The water is a bracing respite from searing summer hea t, and a warm water refuge for manatees and other aquatic life in the winter. Silver Glen is a popular spot for visitors arriving by car and boat alike. It is located on the eastern edge of the Ocala National Forest, and is just a short jaunt by water from Lake George and the St. Johns River. A U . S . Forest Service recreation area surrounds the spring pool, where for a modest fee visitors can have access to a swimming area and manicured lawn replete with picnic tables, charcoal grills, canoe rentals, and col d Coca Cola. On a summer holiday weekend, the flotilla of boats stretches from the edge of the spring pool, down the 1 km long spring run, and into Lake George (Figure 1 1). According to the local land manager, non holiday weekends are often even busier be cause there are fewer law enforcement officers around. This particular Thursday afternoon was July 2, and Independence Day revelers were already pouring in. All, or virtually all, would profess to l ove the place and, perhaps more so, the party. But, they w ould also all be oblivious to, or dismissive of, the impact that such a large gathering of people has on the spring clouds of exhaust, cooking grease spilt overboard, beer cans on the shore, eel grass and other aquatic vegetation
20 trampled, and so on. Perha ps more than any other spring on the river, Silver Glen suffers for its popularity. I was perched on the dock of the Juniper Hunt Club, a private club that from 2007 2013 hosted the University of Florida St. Johns Archaeological Field school. Silver Glen h as been a popular place for longer than most people realize. Indeed, the land surrounding Silver Glen Springs and its run into Lake George housed some of the most imposing p re Columbian structures in the entire state, and it remains one of r archaeological sites. There is evidence of early visitation to the area in excess of 10,000 years ago. By 4,000 years ago Silver Glen Springs was a virtual metropolis with a population that far exceeded that of the surrounding area today. Not only was th ere a resident population, but Silver Glen was the site of massive gatherings of people from far flung places, as it continues to be . These people came from an area spanning hundreds of kilometers and brought with them distinctive goods for use in ritualiz ed feast ing and deposition (Gilmore 2014 ). But this is a hidden history, a history of non events erased or forgotten over the centuries and only now being resuscitated. To wit, a Fore st Service sign marking a 5,000 year old burial mound remarks, simply, th at digging or removal of artifacts is prohibited. Although the site, in the past, was well known to local artifact collectors, few history. prings In 1938, Marjorie Kinnan Rawlings published The Yearling spring, a small collection of sand boils " a few hundred feet from the main pool of Silver Glen:
21 A spring as clear as well water bubbled up from nowhere in the sand. It wa s as though the banks cupped green leafy hands to hold it. There was a whirlpool where the water rose from the earth. Grains of sand boiled in it. Beyond the bank, the parent spring bubbled up at a higher level, cut itself a channel through white limestone and began to run rapidly down hill to make a creek. The creek joined Lake George, Lake George was a part of the St. John's River, the great river flowed northward and into the sea. It excited Jody to watch the beginning of the ocean. There were other begi nnings, true, but this one was his own. He liked to think that no one came here but himself and the wild animals and the thirsty birds (Rawlings 1938:4) . She was not the first to be so str uck remarked on Silver Glen unique in the state. Florida is home to over 1,000 springs, the largest concentration of springs in the world. Indeed, o f the many iconic places on the Florida landscape , springs are perhaps the most beguiling. The best known , like Silver Glen, are enormous pools that dive in to caverns of seemingly limitless depth , drawing scientists, explorers, and throngs of visitors . Others are little more than gurgling puddles, nestle d inconspicuously under verdant canopies of sweetgum, magnolia, and tupelo. The water they bring forth from deep beneath the e arth cool, clear, dancing in the dappled sunlight contrasts starkly with the expansive blackwater rivers, lakes, swamps and wetlan wonder. Pioneering environmentalist Marjor y Stoneman Douglas (1967:24) referred to something authentically Floridian that, unlike beaches and amusement parks, are largely untrammeled by droves of interlopers from the north. In short, springs are significant places in the culture, identity, and heritag e of Floridians.
22 But springs are threatened. In addition to recreational impacts like those suffered at Silver Glen , c hemical pollutants from development and agricultural runoff are increasing, algal blooms proliferate, and spring flows are reduced by groundwater pumped for human consumption (e.g., Florida Department of Community Affairs and Florida Department of Environmental Pr otection 2002; Florida Springs Initiative 2007; Florida Springs Task Force 2000; Pittman 2012 a ). All of this has led to political wrangling over conservation measures a nd to tension between those who seek access to the water, and those striving to protect fragile ecosystems and freshwater resources. However, this concern with the uncertain future of springs, of the unknown changes that will come, belies a view of springs as primordial. This is seen in popular treatments and media accounts. Take, f or example , the Springs Eternal Project ( springseternalproject.org ) , a collaborative effort to promote springs conservation that has featured art exhibitions in both the Florida Museum of Natural History and on publi c transportation vehicles. Ostensibly, projects such as this attempt to motivate public sentiment for conservation by foregrounding the aesthetic appeal of springs and draw ing attention to their fragi lity. However, in so doing , a narrative is constructed that paints springs as unchanging relics of an ancient, pristine Florida landscape that have endured for millennia, only to buckle under the onslaught of modernity. This has the unfortunate side effect of downplaying the significance of springs in the live s of p re Columbian Floridians , beyond unsubstantiated claims that springs were sacred, for Springs have a deep history of entanglement with humanity that has been overlooked in these
23 and intervene to alter it, then it behooves us to explore that history. How did springs figure in the lives of ancient Floridians? What can this knowledge teach us about our current predi cament? Archaeology can provide an important tonic, a perspective that emphasizes both . Springs, perhaps more than any other target of conservation , embody a contradiction. Many spring s have been heavily modified, both in ancient and modern times, by humans. This includes both the obvious terraforming and landscape modification of mound building, channel widening, dredging, and the installation of recreational infrastructure, and the mo re pernicious alterations of water quality and flow But, what people value, what they seek to conserve, are not the cultural features, but the natural the cool, clear wat er, the fragile ecosystems. Conservation efforts recognize the modern alteration of these places, often eliding the thousands of years of transformation that preceded it. They seek to restore springs to a pristine condition. Anthropologists and others have long recognized that the notion of a pristine America prior to European intervention is a myth (e.g., Delcourt and Delcourt 2004; Denevan 1992). However, this realization has not penetrated conservation policy. Anthropologists further recognize that natur e and natural places (as opposed to culture or cultural places) are concepts peculiar to Western thought that artificially separate the civilized human realm from the realm of the exterior, the wild, the other (e.g., Dwyer 1996; Olwig 1993). We cannot simp ly strip away the cultural veneer, the buildings, pollutants, and scoured landscapes to arrive at something natural or pristine. Although
24 springs appear as discrete points on the landscape, they are in fact entangled in complex hydrological, geological, ec onomic, pol itical, historical, and social webs. This is a study of Florida springs archaeology. With over 1,000 springs in the state, obviously I cannot address them all. Rather, I focus on the springs of the St. Johns River valley . I do so for two reasons , first because these are the springs I know best and at which I have first hand experience and, second, because in this region springs have been explicitly used to explain cultural phenomena in the past. But, by focusing on these springs I hope to add not only to our understanding of these specific locales, but also to unearth patterns and processes relevant to springs elsewhere. Springs and the A rchaeology of Florida In contrast to the eternal samene ss of springs, the narrative of pre Columbian Florida an d the St. Johns River valley has largely been one of gradual change (e.g., Milanich 1994; Miller 1998). In brief, a s the environment of Florida shifted from cold, arid state , the people of Florida took advantage of newly available resources, populations grew, and culture slowly became more complex. Eventually, these ancient Floridians shed the simple, nomadic life of hunting and gathering to settle down in villages with hierarchy, domestication, re ligion, monumental architecture, and other trappings of civilization. Archaeological interpretations thus tend to emphasi ze the environment, ecology, subsistence , and adaptation as driving factors during early prehistory . Considerations of politics, power, symbolic life, and the like are reserved for the complex societies of more recent times. In keeping with thi s perspective, springs feature prominently in reconstructions of the Paleoindian (ca. 13,500 11, 5 00 cal B.P.) and Archaic (ca. 11, 5 00 3 5 00 cal B. P.) periods, but fall by the wayside when the emphasis shifts to social, cultural , or
25 ideological explanations for the later Woodland (3500 1200 cal B.P.) and Mississippian periods (1200 500 cal B.P.) . W hen springs are considered it is only for their ecolo gical potential as sources of fresh water, contributors to productive aquatic ecosystems, or attractors of large game (e.g., Dunbar 1991; Miller 1992). The potential social or cultural significance of springs is generally ignored. One illustrative example comes from Little Salt Spring in Sarasota County (Clausen et al. 1979; Purdy 1991:139 158) . The spring itself contains a bounty of ancient material remains and is one of the most significant archaeological sites in Florida. Paleoindian materials at Little Salt Spring were recovered from subaqueous deposits nearly 26 m below the current spri ng surface . The documented inventory includes bone, wood, and antler tools, and the remains of an extinct species of giant land tortoise that was burned after being disp atched with a wooden spear. Radiocarbon assays suggest these remains are in excess of 13,000 years old. In addition, the land surrounding the spring features a large (~ 20,000 m 2 ) Middle Archaic village and subaqueous cemetery , in use from approximately 7,0 00 to 5,000 years ago. The bodies were interred in peat adjacent to the spring after being wrapped with grass and placed on biers of wax myrtle. The cemetery is estimated to contain over 1 , 000 individuals . One individual was buried with a wooden tablet eng raved with a bird effigy ( Purdy 1991:148) . This was clearly an important place in the past, but in a discussion of the site Milanich (1994:80) glosses over these aspects, mus ing s S prings are invoked to explain settlement patterns in the Paleoindian period of Florida, under what has been dubbed the Oasis Model (Dunbar 1991; Dunbar and
26 Waller 1983; Neill 1964). Wh ereas Florida is today characterized by abundant surface wa ter, it was considerably drier during the late Pleistocene and early Holocene. Paleoenvironmental studies indicate that Florida was arid and prairie like with surface water limited to perched ponds and sinkholes (e.g., Watts et al. 1996; Watts and Hansen 1988 ; see Chapter 2 for further discussion ). Recent sea level reconstructions suggest that seas were more than 80 m lower than present when humans first occupied Florida at ca. 13, 5 00 cal B.P. (Balsillie and Donoghue 20 11 ; Otvos 2004). Both sea level and precipitation increased over the course of the early Holocene until near modern conditions were reached at ca. 6 , 000 cal B.P. (Donoghue 2011). Given the arid climatic conditions that prevailed in Florida during the Paleoindian period, it has been argued that deep sinkholes and springs were some of the few locales where fresh water would have been reliably available (Dunbar 1991; Neill 1964). Alt hough highly nomadic, Paleoindian populations may have been te thered to these places, frequently revisiting them in the course of their subsistence pursuits. These watering holes would also have attracted large game, thus affording people ample hunting opportunities. In an updated study, Thulman (2009:271) concluded water sources were the strongest environmental constraint on the occupation patterns rgues that the largest springs are the most likely to have contained water during the late Pleistocene and early Holocene. Received wisdom suggests that this settlement pattern remained relatively unchanged through the early Holocene, although the constraint posed by freshwater availability would have ameliorated gradually over the ensuing millennia, opening up new areas for e xploitation (Milanich 1994:62 63). In the St. Johns River valley , i t was
27 not until the middle Holocene and the inception of the Mount Taylor era (7 ,4 00 4 , 600 cal B.P.) that this pattern changed to one of riverine adaptation and m ore sedentary settlement. M ounds constructed of shell first appear ed at th at time, and Mount Taylor era archaeological sites in the St. Johns River valley are much more numerous than earlier sites, suggesting a spike in population as people (presumably) abandoned the interior upland s and adopted a lifestyle focused on the river. Th e appearance o f shell mounds in the St. Johns is explained by some as simply a response to environmental change that produced new resources for human exploitation. Miller (1992, 1998) has hypothesized that spring flow wa s the key variable in the initiation of riverine adaptation and the consequent appearance of shell sites on the St. Johns River. He argues that under the arid conditions of the late Pleistocene and early Holocene there was insufficient water in the Floridan Aquifer Syste m to support springs. Lacking this input of fresh groundwater estimated to provide nearly one third of the total water flowing in the river the hydrolog ic regime of the St. Johns was likely far different from that of today (Mil ler 1998:67). It may have been a small, rapidly flowing stream, or channeled water may have been discontinuous, seasonally variable, or non existent. Regardless, Miller argues that the productive hydric habitats that today characterize the river valley wou ld not exist absent the input from springs. Johns River in such great numbers coincided with the appearance of habitats for flow is the lynchpin. As sea level and climate approached modern conditions , hydrostatic pressure within the Floridan Aquifer reached a tipping point, resulting in the onset of artesian spring flow. This new input of
28 fresh water, couple d with rising seas, drowned the St. Johns River valley and led directly to the development of ecologically productiv e aquatic biomes. Importantly, a s humans were drawn in greater numbers to the valley habitats, making particular use of the newly abu ndant shellfish. The vast piles of molluscan remains left on the banks of the St. Johns are evidence of this riverine focus (Milanich 1994:87). Thus, the onset of artesian spring flow provided the ecological conditions that underwrote the riverine adaptati on characteristic of the Mount Taylor way of life. Although this model is superficially parsimonious, it reduces the changing cultural practices evi nc ed by the deposition of shellfish to a process of passive adaptation to a climate change event. Further, the springs themselves seem little more than causality between rising sea levels, increased hydrostatic pressure in the Florida n Aquifer System, and spring flow. Given t hat the springs of the region vary in elevation, depth, and underlying geology , initial artesian flow may have been heterogeneous, time transgressive, and punctuated. In addition, variations in precipitation, groundwater mixing, lag in aquifer response, an d aquifer permeability indicate that such generalized models must be tested with local, empirical evidence (Hughes et al. 2009; Miller 1997; Moore et al. 2009; Reese and Richardson 2008). At the broadest level, this study explores the importance and role o f springs in the past . Advocates of the perspective described above would argue that springs provided ideal conditions for subsistence pursuits as they were an im portant source of freshwater and have highly productive ecosystems . This perspective is based on
29 models derived from rational choice theory, cultural ecology, and human behavioral ecology that view human interactions with the environment as transactions wherein humans, as rational actors, attempt to maximize their benefi ts while minimizing their costs. Given enough time, humans become optimally adapted to their environment. Perturbations to the environmental system thus result in reconfigurations of the cultural system; equilibrium is the name of the game. We can see this hypothesis, where in rising seas raised the pressure within the aquifer until a tipping point was reached, initiating spring flow. This in turn had a cascading effect on ancient Floridians, precipitating new adaptations and cultural res ponses . This is evident to us today in the deposition of massive quantities of shell. However, archaeological work at springs in the St. Johns River valley calls this interpretation of the Mount Taylor period into question on several fronts: 1. The earliest s hell deposits in the S t. Johns River valley a re not associated with springs. 2. Some Mount Taylor era deposits at springs are devoid of shell . 3. Spring side practices in the Mount Taylor period were frequently non quotidian and in many cases involved the interm ent of the dead. 4. The onset of spring flow may have occurred much earlier than this model suggests, and was highly variable across the valley. These factors sug gest that the above explanation is insufficient and that an alternative hypothesis is required. This should decouple the onset of spring flow and their ecological productivity from the human uses of these places. In this study, I adopt a perspective informed by phenomenology and practice theory to argue that springs were not simply storehouses of eco logical potential. Rather, they were socially significant places in the landscape, much as they are today. However, I do not seek to
30 deny the importance of the material conditions and biological necessities confronted by people. Rather, I take issue with t he uncritical use of models that emphasize optimization and rational decision making as a human universal. Everywhere people interact with each other and the surrounding world they do so in a distinctive cultural and historical milieu. Thus, in order to ex plore the significance of springs it is necessary to examine the ecological, cultural, and historical threads through which people engaged these places. Outline and Organization The simplistic reference to the past in contemporary springs conservation is d irectly linked to the deficient explanations of springs and their archaeological remains produced by archaeologists. In order to enrich our narratives, in Chapter 2 I review the y for an understanding of why springs are so abundant in Florida and the forces that drive their geographic distribution and hydrology . The chapter begins with a sketch of the geologic history of the Florida Platform and the formation of the Floridan Aquif er System, with emphasis on those events and processes relevant to springs. This is followed by a discussion of the environmental factors affecting spring flow and how these may have been impacted by global and regional climatic changes at the close of the Pleistocene and into the Holocene. Finally, I set the stage for what follows with a discussion of the St. Johns River valley and the springs that feed into it. Chapter 3 tackles head on the question of spring origins, and particularly the intertwined gene sis of spring flow and shell mounding in the St. Johns River valley . I argue that the o nset of spring flow has been treated as an ecological founding event by archaeologists. After discussing the concepts of event and non event as they are used
31 here, I dev elop a critique of this interpretation from three directions. Fi rst, I construct a model of the m iddle St. Johns River valley (where shell mounds and springs are densely concentrated) terrain and springs location s using Geographic Information Systems (GIS) and conduct hypothetical draw downs of pressure within the Floridan Aquifer System to approximate conditions of the early to middle Holocene. This is done neither to predict precisely which springs are most vulnerable to aquifer reductions nor to quantify the magnitude of reduction required to negatively impact springs. Rather, this model is used to critique the notion that the onset of spring flow would have been uniform and rapid across the valley . Following this I discuss the early archaeological and hy drological records of two springs Salt and Silver Glen and the implications of these for nascent shell mounding in the region. To anticipate the results, available evidence indicates that these springs, at least, began flowing far earlier than previously t hought and were visited early and often by Archaic folks who w ere not piling shell along their shores. The onset of spring flow was thus not a significant event that precipitated rapid changes in human lifeways in the St. Johns River valley . I f , in Chapter 3 , I present a major critique of the ecological explanation for springs significance, in Chapters 4 and 5 I begin to sketch an alternative explanation for the use and significance of springs during the Archaic period. In Chapter 4 I deal with Silver Glen Springs directly. This was the site of the largest shell mounds in the St. Johns River valley and greater American Southeast. I consider whether Silver Glen Springs might have been considered sacred in the past. Silver Glen went through several transformations over its history from habitation space to mortuary and back again but throughout was the foca l point of regional gatherings that reached impressive scale and
32 persisted for s everal millennia. From the earliest inhabitant s of the region, springs have drawn people in and been the locales for gatherings big and small. Foregrounding the sociality of springs draws our attention away from their physical parameters, but what is emphasized is that t he two are recursively linked . If springs have been considered sacred, it is a consequence of both their physicality and the histor y of sociality they manifest . Chapter 5 expands the scope of inquiry to other springs both within and outside the St. Johns River valley . It considers the w ays that springs link p eople across the peninsula of Florida. I present the results of archaeological reconnaissance at Silver Springs the largest spring in Florida and western most that feeds the St. Johns River. Of particular interest i s the massive asse mblage of lithic debitage and tools. S tone suitable for the production of flaked tools is notably absent in the St. Johns River valley . Using GIS based spatial analysis, I examine the optimal pathways for the movement of toolstone from source areas in west ern peninsular Florida to four sites in the St. Johns River valley with ample lithic assemblages. I do this specifically to evaluate the hypothesis that Silver Springs and places like it were gateway s or conduit s for the movement of people and objects into the valley . I evaluate the proposed least cost paths with lithic provenance and debitage data. Grappling with these results requires an enlarged viewshed on springs archaeology and underscores the divergent nature of the material assemblage at different s prings. As discussed at the outset, conservation narratives largely gloss over the historical significance E denic, unspoiled Florida. Rather, the ecological, economic, and aesthetic significance
33 o f springs are emphasized. In Chapter 6 I argue that conservation can be fruitfully enhanced by an archaeological sensibility , indeed that the archaeological and historical significance of springs should be intrinsic to their value today. I argue that there is continuity between past and present practice, with springs as attractors that draw people for ritual purposes. In the modern case, these rituals are recreational family gatherings, weekend retreats, holiday festivities that draw people from diverse geo graphies and backgrounds. This continuity of practice makes springs ideal places for demonstrating the relevance of past experience to modern conundrums. This chapter continues the theme s developed in the previous chapter, and weaves threads to the present day. I t begins with discuss ion of some of the contemporary tensions and debates surrounding springs , including barriers to their conservation . Following this, I illustrate the framing of history in media coverage and entanglement theory to examine the myriad ways that springs are caught up with geological, hydrological, social, economic, and political forces . All of this points to a way forward. The goal of conservation cannot be time reversal, or a return to pristine conditions. This would entail the erasure of all the sedimented activities at springs, both ancient and recent. Nor can conservation be an attempt to maintain the status quo, as is impl efforts. Rather, conservation should be aimed towards future states yet to come. Summarizing the preceding chapters, I argue that springs were central to cultural developments in the state, both ancient and modern, but not because o f their ecological potential. The lessons of the past show that springs were significant places, and that social gatherings and the accumulation of historical significance were important factors
34 in the sanctification of springs. This calls for continued us e of springs for recreation (gathering) and greater emphasis on their historical significance to mobilize public sentiment for conservation. I conclude with a discussion of p ossibilities for future research .
35 Figure 1 1. Aerial photograph of Silver Glen Springs. Photo by Overexposed Aerial Photography
36 CHAPTER 2 T he sinkhole was not unexpected. The warning signs had been there for years. Too much water was being sucked out beneath the earth and too many buildings bu ilt atop it. There was particular concern about the area surrounding Winter Park, an affluent suburb of Orlando (Huber 2012; McLeod 1986; Robison 1987). The spring of 1981 was a particularly dry one, with near record drought conditions. On e balmy May eveni ng, Mae Rose Williams stepped outside to the beckoning call of her dog , Muffin, who was tearing around the yard in a frenzy. Unable to calm the dog, or to grasp what had upset her, s he watched, aghast, as a 40 year old sycamore tree disappeared in the corn er of her yard . The next day her home followed the tree into the abyss (Figure 2 1). This was the beginning of the famous Winter Park Sinkhole of 1981. The crater eventually expanded to over 100 m wide, 3 0 m deep, and caused an estimated $4 million in damage (Huber 2012) . Along with Mae R claimed t he municipal swimming pool, a portion of two c ity streets, and a German auto shop, replete with as many as a half dozen Porsches. The dr ama with which this unfolded captured global attention, drawing tourists and gripping viewers on network news broadcasts. of Lake Rose is barely perceptible in the tranquil , tree lined scene today. The same factors that conspired to cleave the earth that day in Winter Park prolonged drought, groundwater withdrawals, development are today sapping the water from features typical of karst
37 landscape s and are ultimately formed by the movement of water across and through the limestone bedrock of Florida. In what follows I discuss the geologic context and history of Fl and the factors that affect their formation, distribution, and operation . I do so to illuminate the physical parameters of artesian abundance while providing a basis for evaluating how var ying environmental conditions can impact s pring flow . The ways that these factors intertwine are relevant both for inferring spring flow dynami cs in the past and predicting them in the future . I begin with a d escription Floridan Aquifer , lifeblood for t Following this , I take a step backwards to sketch the evolution of the Florida Platform, the geologic structure on which the modern Florida Peninsula resides. In particular, I focus on the formation of the vast carbon ate platform that comprises the Floridan Aquifer , and the sands, silts, and clays that entomb it over much of the state. Having outlined the formation of the raw material, the following section turns to the transformation of th e carbonate platform through a discussion of the geomorphology and hydrology of karst terrain. I review the processes by which water sculpts these landscapes and becomes stored in carbonate aquifers. I then examine the dynamics of water within these aquifers: the means by which it ent ers, flows through, and, ultimately, exits th e aquifer through springs and other discharge points and the factors that drive this movement . With this understanding in place, I next examine the changing climatic conditions in Florida over the course of the Late Pleistocene and Holocene, as gl obal climate transitioned from the most recent glacial period and modern regimes became
38 established. Of particular relevance are sea level changes and fluctuations in precipitation, both of which have the potential to augment or retard spring flow. The final section ratchets down from a state wide scale to the St. Johns River basin itself, t he setting for mu ch of what follows . I review the hydrology and geomorphology of the basin, including a discussion of its formationa l history. I then close with a discussion of the geographical distribution of springs in Florida generally and the S t . Johns River valley spe cifically . Springs and Aquifers Over 1,000 springs have been recorded in the state of Florida , t he largest concentration of springs in the world. Broadly, a spring may be defined as a point on the landsca p e where groundwater water stored in rock, soil, or sediment flows , or discharges , onto the surface of the Earth (including the bottom of the oceans; Copeland 2003) . Typically springs flow from aquifers, that is, concentrations of groundwater in sufficient quantity that water can be extracted for human use (Knight 2015: 14 ; Lane 1986:9 ). The amount of water available in an aquifer depends on the porosity and permeability of the rocks, in other words, the amount of available space that can be filled with water and accessibility and interconnected ness of those spaces. This water is typically derived from precipitation that percolates down through overlying materials. Groundwater in aquifers may also come from water that has infiltrated from adjacent aquifers or, in coastal zones, intruding sea water. Aquifers ma y be either unconfined or confined (Lane 1986) . Unconfined aquifers are typically near the surface and are under only normal atmospheric pressure. As a result they are often referred to as surficial or water table aquifers. Springs flowing from these uncon fined aquifers are referred to as seep or water table springs . In the following
39 discussions, water table spring is preferred as seep spring has a specific context in Florida (Copeland 2003). Water table springs occur when water percolating through surficial soils and sediments encounters an impermeable layer. The water moves laterally along this layer until it reaches a point of lowered eleva tion, such as at the base of hills or in a topographic depression, where the land surface drops below the elevation of the aquifer surface (i.e., the water table). Confined , or artesian, aquifers are often more deeply buried, and separated from surficial a quifers by less permeable materials, such as clay, that restrict the flow of water (Lane 1986) . This layer is referred to as a confining or semi confinin g unit, depending on the degree o f impermeability 1 . Recharge of water in confined aquifers occurs in up slope areas where the confining uni t is thin or absent. U pslope recharge and the weight of the water entering the aquifer place it under hydrostatic pressure . This pressure creates the necessary conditions for artesian flow, wherein water will rise to a hi gher elevation than that of the aquifer itself when tapped by a well. The elevation that the water rises to in a tightly cased well, where the water pressure is in equilibrium with atmospheric pressure, is referred to as the potentiometric level . Pressure within a confined aquifer varies over space and time, and is visualized across its extent as a potentiometric surface (Figure 2 2) . Springs that flow from a confined aquifer are referred to as artesian springs. The y occur at places on the landscape where t he confining unit overlying a n aquifer has been 1 Confining and semi confining units are sometimes referred to as aquicludes or aquitards. Confining units can also occur within or above the surficial aquifer, where they create a perched aquifer. These typically have a relatively small areal extent, but can form lakes or ponds whose water level is independent of the surrounding water table.
40 breached and where the pressure within the aquifer is sufficient to push water up onto the surface, that is, where the potentiometric level of the aquifer is higher than the surface elevation (Lane 1986; Scot t et al. 2004) . The vast majority of springs in Florida are artesian springs that discharge from the Floridan Aquifer System, an extensive source of groundwater that underlies all of Florida, much of Georgia and South Carolina, and portions of Alabama and Mississippi , with an aerial extent of over 250,000 km 2 ( Figure 2 3; see Miller 1986, 1997; Williams and Kuniansky 2015) . It is the primary source of freshwater for agricultural irrigation, industrial, mining, commercial, and public supply in Florida. I n 2010, 4,166 million gallons per day were pumped from groundwater aquifers in Florida , supplying potable water for over 17 mil lion people, or 92% :9 ). Approximately 62% of this total was extracted from the Floridan Aqu ifer . The Floridan Aquifer is a thick sequence of highly permeable carbonate rocks that are bounded above and below by confining units. The Floridan Aquifer is at least 10 times more permeable than the se confining units (Hine 2013:113) , and ranges in thic kness from less than 60 m in the panhandle to over 1,100 m thick in the central and southern peninsula (Williams and Kuniansky 2015 :47 ). The upper confining unit consists of mid and late Miocene deposits comprised principally of interbedded sands, silts, and clays (Miller 1986; Williams and Kuniansky 2015:38). In northeast Florida, the upper confining unit consists almost exclusively of Hawthorn group sediments and measures approximately 120 m thick (Brown 1984). The lower confining unit, typically l ow per meability late Paleo gene to early Neogene Eocene rocks , forms the base of the Floridan Aquifer (Scott 2011:22) .
41 The F loridan Aquifer can be divided vertically into an Upper and Lower aquifer, which are separated by a middle confining (or semi confining , de pending on the location ) unit. The U pper Floridan Aquifer is the source of most of the springs in Florida, and is tapped extensively as a source of potable water (Miller 1997). It consists largely of late Eocene to middle Oligocene carbonates measuring up to 200 m thick (Scott et al. 2002:10; Scott 2011:2). The porosity and permeability, elevation, stratigraphic position, and degree of confinement of the Upper Floridan Aquifer vary considerably across the state . The Middle Confining Unit separates the Upper and Lower portions of the Floridan Aquifer. This i s a discontinuous zone of relatively impermeable micritic limestone and dolomitic limestone. The Lower Floridan Aquifer is perhaps the most poorly understood portion, because of its deep burial and the pre sence of saline water (Miller 1986). The Lower Floridan Aquifer varies from a relatively thin zone of carbonate rocks in northwest Florida to a thick, complex sequence of thin permeable zones separated by thick semi confining units in southeastern Florida. It is generally composed of early to middle Eocene carbonates, over 950 m in maximum thickness (Scott 2011:22). Two major aquifers overl ie the Floridan Aquifer 2 : the Intermediate A quifer and the Surficial A quifer (Miller 1986; Reese and Richardson 2008; Scott, et al. 2004). These aquifers are usually separated from the Floridan Aquifer by confining units (e.g., clay or limestone) which prevent, or at least limit, transmission of water. However, in some places they are in communication with the Floridan A quifer to varying degrees. 2 For the s ake of simplicity I forego discussion of local aquifers, such as the Biscayne and sand and gravel aquifers, that, while important to regional water supply, are less relevant to the distribution and mechanics of artesian springs.
42 The surficial aquifer consists of sands, silts, shell, and some limestone and sandstone overlying the Floridan Aquifer . This aquifer is as thick as 120 m i n some portions of the state (Reese and Richardson 2008), but is completel y absent in portions of northwest Florida where the Floridan Aquifer is unconfined at the surface. The Intermediate A quifer extends from the base of the surficial aquifer to the uppermost confining unit of the Floridan Aquifer , primarily in southwest F l ori da . This aquifer similarly consists of sands, silts, clays, and carbonate rocks, and is discontinuous across the state. Ultimately, the presence and distribution of springs is controlled by the physiography and geologic framework of Florida and the product ivity of the Floridan Aquifer . Understanding th e spatial heterogeneity of the Floridan Aquifer and the abundance of Florida springs requires a brief foray into the geologic history of the Florida peninsula, and a discussion of the conditions under which th e Floridan Aquifer was fo rmed. Geologic al History of the Florida Peninsula The Florida Platform is the geologic structure on which the state of Florida resides . This includes not only the emergent portion of the Florida peninsula, but also submerged portions of the continental shelves and slope . Appr oximately 50% of the Florida Platform is submerged under the Atlantic Ocean and Gulf of Mexico. The platform is bounded to the west , in the Gulf of Mexico, by the West Florida escarpment, to the north by t he Georgia Channel System, to the east by the S traits of Florida, and to the south by the passive tectonic margin between the North American and Caribbean plates ( Hine 2013:12 ; Scott 2011 ) . The Florida platform is a product of the seas, and its geologic hi story one of marine processes , as seas continually rose and fell and the platform was alternately submerged and emerged (Schmidt 1997). The Florida Platform
43 is comprised of three basic components basement rocks, carbonate rocks, and s i liciclastic sediments that reflect differing geological processes and histories . The latter two are the most relevant for an understanding of Florida springs, and so I touch only briefly o n the first. Florida Basement The basement rocks of the Florida Platform are igneous and metamorphic rocks that are sutured on to the North American plate . The basement rocks are highest in north central Florida (below mean sea level), dipping to the south, east, and west from 915 m to over 5,000 m below mean sea level (Scott 2011:17). Prior t o approximately 475 million years ago (Ma) (along with the Yucatan and Bahamas) were a constituent of the continent Gondwana, which i ncluded much of the (Hine 2013; Smith and L ord 1997) . The Florida basement rocks were situated between what is now South America and northwestern Africa , as a portion of the African plate . By 250 Ma the supercontinent Pangaea had formed, welding th e Florida basement rocks onto La urentia , the ancestral North American continent . Twenty five million years later, tectonic rifting led to the breakup of Pangaea and the opening of the North Atlantic, Gulf of Mexico, and Caribbean Sea basins . As a result of this rifting, Florida was detached and separ ated from Gondwana and a ttached to Laurentia as exotic terrain . Rifting at that time also resulted in the formation of the South Georgia Rift a rift valley across southern Georgia that isolated the Florida Platform from the rest of North America (Smith and Lord 1997:24) . Following the c reation of the Florida basement, several factors converged to promote the deposition of carbonate sediments and the formation of a
44 thick sequence of carbonate rocks , the second major component of the Florida Platform. Carbon ate Platform The basement rocks of the Florida Platform provided a broad surface for the deposition of carbonate sediments (Hine 2013; Randazzo 1997; Scott 2011) . T h ese sediments would eventually form a thick sequence of carbonate rocks that today house th e Floridan Aquifer The following offers a brief primer on the production and deposition of carbonate sediments before outlining Carbonate sediments are formed from the skeletal materi als of a wide range of life forms , including plants, animal s , and micro organisms (Hine 2013:56) . As a result, carbonate sediments themselves are highly diverse and reflective of the organisms from which they are derived and the environments in which they were deposited . Carbonate sedimentation occurs most readily in clear, shallow marine environments where light penetratio n is sufficient to stimulate photosynthesis. However, these are not uniform environments and different physical and chemical parameters favor different assemblages of organisms. On a n emerging carbonate platform , factors such as elevation, water depth, sal inity, and wave and current energetics form an array of microhabitats that vary across space. Like w ise, the organisms and carbonate sediments produced on a carbonate platform can be varied and horizontally stratified (Hine 2013:57 59) . In addition to this spatial variability in the production of carbonate sediments, t hese sediments can in turn be reworked and redistributed by waves and currents.
45 The end result is a diversity of sedimentary facies that vary both horizontally with micr o habitat diversity and v ertically as sediments accumulate and conditions change through time, for example as a result of sea level fluctuations. Once these sediments are buried they become fused together through a process called cementation, creating carbonate rocks. This cementa tion occurs as water fill s the interstitial spaces in the sedimentary matrix . This water contains dissolved carbonate minerals that can precipitate out and fill these spaces . Th e cementation process can be quite rapid, often occurring over a matter of deca des, or can be ongoing for centuries (Hine 2013:60). The Florida basement rocks provided a broad, elevated surface that was shallowly submerged during sea level high stands . Because of its geographical position at low latitude, the seawater covering the Fl orida basement was relatively warm. These provided ideal conditions for the deposition of a carbonate platform. Carbonate rocks comprising the Florida Platform measure as much as 6 km thick. But, as discussed above, carbonate sediments form in relatively s hallow (<50 m) sea water. A fortuitous combination of two factors allowed for the continued accretion of carbonate sediments that ultimately formed this thick platform : tectonic subsidence and sea level rise . After the breakup of Pangaea, the newl y formed continental margin of North America cooled and began subsiding rapidly , creating space for carbonate sedimentation on the submerged Florida basement. The accumulation of carbonate sedi ments, in turn, propagated subsidence by increasing the lithostatic load on the plate margin . Likewise, sea level rise at th at time creat ed space for continued carbonate sedimentation . This, too, propagat ed tectonic subsidence by increasing the hydrostatic load on the plate (Hine 2013: 62 63) . Thus, the Florida basement sank while the level
46 of the sea rose, cont inually creating space for the production and accumulation of carbonate sediments. Importantly, the space created by subsidence and sea level rise was relatively persistent and kept pace with , but did not outstrip, carbonate de position . If either subsidence or sea level rise had proceeded too quickly, water depth would have increased to such an extent that t he carbonate factory drowned . So, i n sum, the broad elevated surface of the Florida basement rocks provided an ideal substrate for the formation of a carbonate platform. The thickness of this platform was afforded by tectonic subsidence and persistent sea level rise. Sea level oscillations and concomitant variations in micr ohabitat during this overall rise formed laterally and vertically heterogeneous carbonate deposits and unconformities that ultimately resulted in a diversity of carbonate deposits on the Florida Platform. But this suite of ideal conditions could no t persis t indefinitely. T he Florida carbonate platform accreted early and rapidly, peaking at approximately 140 Ma. As the rifted continental margin migrated away from the spreading ocean basin, subsidence slowed exponentially (Hine 2013:63). Sea level rise likewi se attenua ted after ~100 Ma. C arbonate deposition continued, albeit at a reduced rate, through the Eocene and into the Oligocene . Following this, a fun damental shift in sedimentation and deposition al regimes oc c ur r ed on the Florida Platform. This new regim e provided the third major component of the Florida Platform. Siliciclastic sediments The carbonate platform described above ultimately formed the limestone bedrock of Florida. But this bedrock is not visible on the surface. With the exception of the Everg lades and Keys, the entirety of the Florida carbonate platform is now covered by a mantle of siliciclastic sediments composed primarily of qu a rtz sand s, along with some
47 silts and clays . T ypically , this sand is a few meters thick , but in localized areas may be absent, or considerably thicker . This s ediment is derived not from a local source, but from the weathering and transport of non loca l igneous and quartz rich metam orphic and sedimentary bedrock (Hine 2013:137). The closest source of these rocks is the southern Appalachians and Piedmont of Georgia and North and South Carolina. Over the past 250 million years the Appalachians have been subjected to extensive weathering and erosion. This has reduced the height of the mountain range by some 7 , 000 m and prod uced a vast amount of sediment (Hine 2013; Scott 1997, 2011) . These siliciclastic sediments have been transported by streams and rivers to the sea . Along the way they have been reworked and deposited by fluvial processes to build up the coastal plain, cont inental shelf, and continental slope of the lower Southeast . Until approximately 30 Ma, the Florida Platform was isolated from this sediment source by the Georgia Channel System, a remnant of the South Georgia Rift that formed during the Triassic (Scott 20 11 :19 ) . Currents in this channel prevented the progradation of deltaic deposits that might otherwise have increase d turbidity and negatively impacted carbonate formation. However, prolonged sea level low stands after ~30 Ma reduced current velocity in the Georgia Channel System . This was coupled with uplift of the eroding Appalachians , which increased the siliciclastic sediment supply (Scott 2011:25). The combination of increased sediment load and decreased current velocity in the Georgia Channel System res ulted in its gradual infill ing with sediments. These sediments eventually reached the Florida Platform, precipitating the demise of carbonate production in Florida.
48 During sea level high stands , these sands migrated southward by longshore transport along r elict shorelines , visible now as elevated inland ridges oriented parallel to the Atlantic and Gulf coasts (Hine 2013; Schmidt 1997; Scott 2011) . During low stands, nascent rivers and drainage systems carved through these ridges, rework ing sediments and mov ing them from the interior on to the e merged continental shelf. In this way, water transported siliciclastic sediments south, east, and west, covering the carbonate platform and extending onto the continental shel ves of and Gulf coast s . These three components of the Florida Platform basement rocks, carbonate platform, siliciclastic mantle springs. The basement rocks provided the necessary conditions for carbonate deposition a broa d, elevated platform covered by shallow seas and isolate d from deltaic deposition from the North American mainland . These carbonate rocks, in turn, provide the structure that comprise s the Floridan Aquifer and house s the freshwater that feed s ci clastic mantle overlies the Floridan Aquifer , forming confining units, surficial aquifers, and surface topography, and in other ways facilitating or hindering access to it. With this understanding of the str uctures in place, I can now address the processes which sculpted the platform to form karst terrain and the carbonate Floridan Aquifer . Karst Terrain in Florida Carbonate rocks, like those of the Florida Platform, are subject to a variety of weathering pro cesses that form a distinctive topography referred to as karst. Karst terrain is characterized by numerous surface and sub surface solution features such as sinkholes, caves, springs, sink rise str eams, conduits, and fractures that impart a
49 distinctive hyd the Karst region of the former Yugoslavia (now encompassing portions of Serbia and Montenegr o). This locale is the model karst system, featuring significant subsurface drainage through limestone rocks perforated by numerous conduits and caverns. Within the United States, karst terrains can be found in southern Indiana, central Kentucky and Tennes see, New Mexico, and the Appalachian Mountains, but perhaps the most extensive and well known karst region is found in Florida (Lane 1986). The Evolution of Karst Landscapes The primary geomorphic agent in karst terrains is water, particularly through chem ical weathering of carbonate rocks. Chemical weathering of carbonate rock is controlled by 1) the presence of slightly acidic water and 2) a mechanism for transporting that water across and through the rock. As water moves through the atmosphere it absorbs carbon dioxide to form small amounts of carbonic acid (Lane 1986 ). When this water reaches the surface as precipitation, it percolates through the surficial soils/sediments (if present) to reach underlying carbonate rocks. In doing so it mixes with additi onal carbon dioxide derived from soil microbial activity, further enhancing acidity. On reaching carbonate rock , movement of this acidic water is di rected by the porosity and permeability of rocks and the surrounding sedimentary matrix (Lane 1986) and driv en by gravity and gradients in temperature and pressure . Porosity refers to the relative amount of pores or voids within a rock or sedimentary deposit, whereas permeability more specifically references the ability of water to move through those pores. Thus while a high degree of porosity is a necessary precondition for high
50 permeability, it is not sufficient in itself. Numerous factors may limit the permeability of a porous material, such as a lack of interconnection between pore spaces, or the presence of fine materials (e.g., clays or organic matter) within interstitial voids. Further, we can differentiate between primary and secondary porosity. Primary porosity is that which is inherent in the material as a result of deposition. Secondary porosity is form ed as a result of weathering and diagenesis, for example as a result of tectonic fracturing or chemical dissolution . Limestone , like that of the Floridan Aquifer , is typically characterized by both high porosity and high perm eability, attributable to its g ranular structure and the presence of fine fractures and bedding planes. As the water passes through the rock , its wea kly acidic nature c auses some of the carb onate to dissolve, enlarging preexisting fractures, bedding planes, and voids within the rock. As this process progresses, fluid transmission pathways are enlarged and may form interconnected conduits that facilitate the movement of water. Carbonate dissolution is greatest in areas of high groundwater circulation (Hine 2013:124) , so this becomes a sel f sustaining cycle where by dissolution begets greater groundwater transmissivity, which further enhances dissolution and concentrates networks of conduits and caverns . In coastal karst zones, such as along both coasts of Florida, dissolution is enhanced at the contact between salt and fresh water (Hine 2013:118 119) . Fresh water is less dense than salt w ater, and as a result floats above it . If both water bodies are saturated with respect to calcium carbonate, they will be unable to dissolve additional carb onate rock. However, in the the water is brackish and
51 becomes under saturated with respect to calcium carbonate. Thus carbonate rock can be dissolved within the mixing zone and propagated deep within the aquifer . In areas where porosity is co ncentrated and larger cavities eventually form, the dissolution of underlying rock can eventually lead to the collapse of overlying deposits . This can begin deep within the carbonate structure where faults and fractures are concentrated. With each collapse a new void for groundwater circulation and carbonate dissolution is formed. In this way, the collapse propagates upwards through a process called dissolution tectonics (Hine 2013:125). When it reaches the surface, these cause regional subsidence in the fo rm of broad topographic depressions, with underlying fold and sag structures . The dissolution of carbonate rock and karstification of the landscape produces a variety of surface and subsurface features. As dissolution progresses through time these become i ncreasingly prevalent and well developed. A variety of names and definitions are applied to karst landscape features in different regions of the world. The following is borrow ed from the definitions of the Florida Geological Survey (FGS; Copeland 2003). Wh ere carbonate ro ck is exposed at the surface a variety of weathering features are formed, collectively referred to as a karren. These are formed by solution of the rock at the surface. In Florida, these typically consist of pinnacles and depressions in the rock, now buried beneath overlying sediments. More typical in Florida are caves , openings and passages large enough to be entered by humans , and sinkholes , depressions caused by the dissolution of underlying materials (sometimes referred to as cenotes , d olines , or sinks). T hree types of sinkholes are generally recognized : solution, subsidence, and collapse . The mechanism of their
52 formation depend s broadly on the thickness of confining units and sediments overlying the carbonate rock (Kindinger et al. 1999 ; Schiffer 1998 ) . S olution sinkholes form where this overburden is absent and a depression is formed by the removal of rock. Subsidence sinkholes are similar and occur where a thin veneer of sediment covers the rock. This sediment slowly subsides as dissol ution progresses. Collapse, or cover collapse, sinkholes are the most dramatic of the three, common where overburden is thick , but the confining unit is absent or breached. These form as underground cavities or voids and expand in size through dissolution. As the cavity grows larger, the rock forming the roof of the cavity gradually thins. Eventually, it is no longer able to support the weight of overlying sediments, which then collapse into the void. Cover collapse is also responsible for the formation of karst windows, openings that reveal portions of subterranean flow in the aquifer. Another common karst feature, particularly in northern Florida, is a sinking stream or river rise system. This occurs where channeled surface water disappears i nto a karst de pression called a swallow hole (also ponor or swallet). The water flows underground for some distance, where it then re emerges onto the surface, usually at a lower elevation. Although these are generally considered to be springs, they are not technically so , as surface water comprises a significant portion of the resurging water (Copeland 2003) . The features described above are formed by the interaction of water and carbonate rock. Carbonate minerals that have been dissolved in the water are moved in solut ion through the system, eventually exiting when the water discharges into the oceans . In Florida it is estimated that this dissolution occurs at a rate up to 4 cm per
53 1 , 000 years (Hine 2013:115 116). The following section examines the hydrology of karst aq uifers and the forces that drive water through the system. Karst Hydrology The dissolution of carbonate rock described above and the formation of a network of water bearing voids (i.e., secondary porosity) is the mechanism by which the Floridan Aquifer for med and the reason it is so productive. This same process is also responsible for the karst features visible on the surface sinks, springs, sink rise streams, and the like. These features, in turn, are critically important for controlling the movement of w ater into and out of the aquifer. The hydrologic cycle of karst aquifers can be conceptually divided into processes of recharge, flow, and discharge (White 2002) . Although these will be discussed separately below, each process is intricately tied to the ot hers. Recharge Precipitation is the main source of recharge , or water input , to karst aquifers , driven into the system by the force of gravity . This input can be divided into four components (White 2002:87 89). Allogenic recharge occurs when the karst aqui fer captures water , typically through swallets or sinks , from surface streams that drain non karst or non carbonate portions of the landscape. These streams transport water that would not otherwise enter the karst aquifer. More common is d iffuse infiltration , when precipitation falls directly on the karst surface, saturates and percolates through the overlying soil or sediment, and enters the aquifer through fractures and pore spaces . Internal runoff consists of precipitation that falls directly i nto closed basins (sinkholes, lakes, etc.) that are connected to the aquifer and recharge it directly . The sinkhole drains at the base of these features may be open and freely connected to the aquifer, or
54 may have sediment plugs that restrict recharge. Fin ally, some water may be obtained from overlying caprocks or perched water sources (separated from the aquifer by an impermeable layer) if these overflow. Water from these sources may then recharge the karst aquifer through fractures around their edges or t hrough diffuse infiltration. Recharge typically occurs in areas where the confining unit does not appreciably restrict the downward movement of water into the aquifer, that is, where the aquifer is unconfined or thinly confined (Aucott 1988; Miller 1986). Further, r egardless of the mechanism, recharge to a karst aquifer only occurs in areas where the potentiometric surface of the aquifer is below the elevation of the land and/or surficial water table (Belaineh et al. 2012) . Otherwise the positive pressure w ithin the aquifer prevents the recharge of new water. The areas of greatest recharge to the Floridan Aquifer, then, are areas of we ll developed karst that are at relatively high elevation (Figure 2 4) . The surface and groundwater basins that contribute wat er to a given spring comprise the springshed or spring recharge basin (Copeland 2003; Scott et al. 2004; Shoemaker et al. 2004; Williams and Kuniansky 2015). Importantly, these are not static boundaries but fluctuate through time. This is a result of varia tions in pressure gradients and groundwater flow dynamics. Flow The flow of groundwater in karst aquifers is driven by several factors (Lane 1986; Miller 1986; White 2002; Williams and Kuniansky 2015) . First, gravity will drive flow from high elevation recharge areas. If the aquifer is confined this creates hydrostatic pressure that increases with depth and distance from recharge zones. Recharge water is also frequently warmer than existing groundwater in the aquifer. As a result ther e are gradients in both pressure and temperature which drive water flow. Finally, convective
55 forces generated by heat deep within the earth, called Kohout convection, circulate groundwater and can draw seawater into deep coastal aquifer s . This creates a mi xing zone, analogous to that described above , b ut much broader and deeper, that enhances dissolution (Hine 2013:125). Carbonate karst aquifers are generally considered to have three types of porosity or pathways for water transmission: intergranular porosi ty in the matrix, small fractures , and large conduits or caverns (Martin and Dean 2001; White 2002). These three types of porosity are referred to as matrix (or intergranular), fracture, and conduit porosity. Matrix porosity consists of interstitial spaces in fabric of the rock or sedimentary structure. Fractures consist of small mechanical apertures, including such features as joints and bedding planes, which range in size from 50 500 Âµm (White 2002). These may be widened by carbonate dissolution up to abo ut 1 cm. Openings or pathways that have been enlarged by dissolution to a size of greater than 1 cm are referred to as conduits. The distribution and abundance of these pathways in a given portion of an aquifer can have dramatic effects on permeability and flow. In general, and despite the small size of the voids, matrix porosity is thought to provide much of the water storage within the aquifer . Conduits, meanwhile, provide for the majority of flow (Martin and Dean 2001; White 2002). However, many studies of karst aquifers have focused on areas of dense, relatively impermeable rock. In these karst regions flow between the matrix and conduits is relatively restricted. As a result, springs have been conceptualized as a direct output of subsurface flow through conduits (i.e., , with little regard given to the potential input from flow through
56 small fractures or matrix pores (Florea and Vacher 2006; Martin and Dean 2001; Moore et al. 2009; Screaton et al. 2004). Karst can be divided into two main types eogenetic and telogenetic on the basis of age and porosity (Florea and Vacher 2006). Eogenetic karst is young and has not been deeply buried, wh ereas telogenetic karst is much older, having gone through an intermediate mesogenetic stage of d eep burial and subsequent erosion and exposure. These types differ in their physical characteristics, geochemistry, and hydrology. Of particular relevance to karst hydrology are differences in matrix permeability, which decrease roughly with age. As a resu lt of its deep burial and compaction, the matrix permeability of telogenetic karst has been significantly reduced by several orders of magnitude relative to eogenetic karst (Florea and Vacher 2006). T he karst of Florida is relatively young, having formed i n the Eocene and Oligocene (25 to 50 Ma) , and has n ever been deeply buried (Florea and Vacher 2006; Miller 1986; Reese and Richardson 2008). Thus Florida karst i s eogenetic. Although traditional models of karst hydrology minimize the impact of matrix perme ability on groundwater flow , it has much greater effect in eogenetic karst ( Martin and Dean 2001; Moore et al. 2009; Screaton et al. 2004) . The forces of groundwater flow (e.g., gravity, temperature, pressure, Kohout convection) drive water through the aqu ifer to points of discharge on the landscape. At the scale of the entire aquifer , the direction of groundwater flow in the Floridan Aquifer is generally from the interior highlands towards the Atlantic and Gulf coasts (Hughes et al 2009:26). However, at mo re local scale flow is generally directed from recharge zones within springsheds towards points or zones of groundwater discharge.
57 Discharge Springs are the primary discharge point for groundwater in karst aquifers (Scott et al. 2004; White 2002). A variet y of classifications are used for springs, which are subdivided on the basis of size, source of water, landscape position, or discharge mechanism ( Copeland 2003; White 2002:90). The Florida Geological Survey, for example, uses a simple bivariate scheme bas ed on geomorphology that categorizes springs as either terrestrial or marine , with focused or diffuse flow (Copeland 2003). As noted above , s prings that originate from surficial, unconfined aquifers are referred to as water table springs 3 . Those that emerg e at the surface as a result of pressure , like those originating in the Floridan Aquifer, are artesian springs. In Florida these are also referred to as karst springs (Copeland 2003). Most artesian springs opening that concentrates ground water discharge to the Earth's surface, including the bottom of the ocean larger than the pore space of the surrounding rock, often forming a cavern or fissure ( Copeland 2003:16; see also Scott, et al . 2004:10). However, not all groundwater discharge is concentrated at focal points like spring vents. Discharge can also be diffuse, occurring over broader portions of the landscape. Diffuse groundwater discharge (sometime called upward leakage) occurs thr ough the intergranular pore spac es in the aquifer matrix (Scott et al. 2004:10), and thus is more common in the eogenetic karst of Florida than other karst regions (see above) . If this occurs in a relatively restricted area it is referred to as a seep, tha t is, an artesian spring 3 Again, water table springs are also referred to as seep springs, but in the Florida nomenclature seep is reserved as a descriptor of the spring orifice without regard to the water source (Copeland 2003).
58 from the ground The water discharging from springs is typically contained in a water body referred to as a spring pool (Copeland 2003; Scott et al. 2004). Discharge from spring vents can be turbulent, resulting in a notable roiling on the surface of the pool referred to as a spring boil. This water can then be channeled to form a spring run, or may drain direc tly into existing streams, creeks, or other surface water bodies. Springs may also be subaqueous located beneath existing surface water bodies in the beds of rivers, lakes, or beneath the oceans. As with recharge, discharg e from carbonate aquifers is most common where the confining unit is thin or absent an d karst features are prevalent and well developed . Discharge in these areas occurs where the potentiometric surface of the aquifer is higher than the elevation of the land surface and surficial water tab le. Artesian springs in Florida , then, are most common at low elevations in areas of well developed karst (Scott et al. 2004:15). Springs can also be classified on the basis of the amount of water that discharges through them. In Florida, discharge is used to divide springs into 8 magnitude classes (Table 2 1). This encompasses a huge range of variation. The smalles t , eighth magnitude springs discharge a maximum of one pint of water per minute. This is equivalent to approximately 180 gallons per day. In con trast, first magnitude springs discharge a minimum of 64,600,000 gallons of water per day and the largest of these, such as Silver and Wakulla springs, discharge 3 4 times as much (Scott et al. 2004). The intensity of discharge in artesian springs is press ure dependent.
59 This pressure fluctuates both temporally and spatially as a result of several factors that vary within and between individual spring basins , such as precipitation, topography, elevation, soil characteristics, and variations in the physical p roperties of the aquifer (e.g., permeability , thickness of confining unit ; Scott et al. 2004). Current understanding of spring flow dynamics emphasizes precipitation as the main driver of seasonal and annual discharge variation (Knowles et al. 2002; White 2002). However, springs in areas of eogenetic karst, such as Florida, tend to respond differently to precipitation than those on telogenetic karst. They generally have lower amplitude variation in discharge, longer lag time in response to precipitation eve nts, and greater buffering of high frequency/low intensity events, which may not substantially recharge the Floridan Aquifer . Rather, high intensity storms and seasonal, annual, and decadal precipitation cycles appear to exert greater influence on variatio n in spring discharge (Florea and Vacher 2006, 2007). These differences are primarily caused by the greater contribution of matrix permeability to groundwater storage and flow in eogenetic karst aquifers , which buffers variation and increases aquifer inert ia (see above). In addition, deep water upwelling in the Floridan Aquifer can contribute significant amounts of water to spring discharge (Moore et al. 2009). Thus, discharge at springs include s water that entered the aquifer relatively recently and much older waters, recharged as much as 30,000 years ago (Plummer 1993; Toth and Katz 2006). At longer temporal scales , fluctuating sea levels, climate change, and groundwater withdrawals can impact hydrost atic pressure in the Floridan Aquifer and, therefore,
60 spring flow. How these factors changed over the course of the late Pleistocene and Holocene is the subject of the next section. Quaternary Environments of Florida and the St. Johns River Valley From the perspective of spring flow, the most relevant climatic factors over long periods of time are those that can have significant influences on the potentiometric surface of the Floridan Aquifer and the pressure and temperature gradients that drive flow within the aquifer . These are, principally, sea l evel and precipitation. Sea level is generally considered to be the lower limit, or base level , for water movement in both ground and surface water systems (Charlton 2008; Knox 1995 ) . Other things being equal (e.g ., precipitation and evapotranspiration) , a reduction in the base increases the hydraulic gradient. In a surface network, such as a river, this increased gradient generally leads to greater water velocity and down cutting, or incisi o n, of the river channel (Blum and TÃ¶rnqvist 2000) . Likewise, an increased hydraulic gradient in a groundwater system raises water velocity and, in the case of a carbonate aquifer, increases the rate of carbonate dissol u tion. Further, s ea level is relevant because the freshwater of the Floridan Aquifer is in communication with the water bodies surrounding it . Although the Floridan Aquifer ends at the coasts, carbonate rocks of equivalent age extend offshore up to 100 km in the Atlantic and 200 km in the Gulf (Hughes et al. 2009:79 5). Groundwater discharges diffusely onto the continental shelf in many locales and s ubmarine springs are well documented in Florida and elsewhere ( Fleury et al. 2007; Karst Environmental Services 2008; Lane 2001; Swarzenski et al. 2001 ) . Likewise, s ubmarine karst features (i.e., sinkholes) have been found deep within the Straits of Florida (Land and Paull 2000). Recall that freshwater in the Floridan Aquifer floats atop brackish and saline waters
61 deep within the Lower Floridan Aquifer . This water is circulated through Kohout convection, and is also found in isolated pockets within the freshwater of the Upper Floridan Aquifer (Williams and Kuniansky 2015:122). Sea level fluctuations wil l raise and lower the mixing zone between fresh and saline water . A gain with other things being equal, under conditions of lowered sea level, this zone would drop and the saline water in the system would gradually be replaced by freshwater. This would also, however, decrease hydrostatic p ressure in the deep aquifer, lower ing the potentiometric surface and reducing or eliminating spring flow . Conversely, under conditions of sea level rise (like those predicted for the near future), hydrostatic pressure in the aquifer will increase and the elevation of the potentiometric sur face will rise. Spring flow would then likely increase, depending on local circumstances. However, sea level rise would also gradually flush out the freshwater in the system, replacing it with brackish and saline water. It is clear that sea level has a st rong influence on the Floridan Aquifer . However, several factors suggest that the correlation between sea level and spring flow is far from linear or straightforward. Permeability in the aquifer is not uniform, and confining units represent significant imp ediments to flow and groundwater equilibration (Williams and Kuniansky 2015). For example, hydraulic connectivity with the Atlantic Ocean is appreciably restricted in the Upper Floridan Aquifer , relative to the Lower Floridan Aquifer (Bennett 2004 ). R ecent efforts to model the response of the U pper Floridan Aquifer to fluctuating conditions indicate that hydrostatic pressure in the aquifer may require up to 1,0 00 years to stabilize after sea level rise , depending on the rapidity of change and the degree of confinement of the aquifer (Hughes et al. 2009). Temperature
62 and chloride stabilization can require even longer. This indicates that simplistic models correlating sea level rise with artesian spring flow potentially mask other relevant factors while limiti ng the potential t o understand variability among springs. Similarly, long term changes i n precipitation, particularly over a reas where the aquifer receives significant recharge, have the potential to alter the hydraulic gradients and available water within the system. Greater precipitation in recharge areas increases the velocity of groundwater movement, hydrostatic pressure, and spring discharge . However, as with sea level, the relationship between precipitation and spring flow is not linear or direct , particularly in Florida (see above). Both precipitation and sea level have fluctuated since the Last Glacial Maximum in the Late Pleistocene, just prior to the migration o f humans into Florida. Late Pleistocene conditions in Florida General narratives of Late Pleistocene to Holocene climate change in Florida emphasize the gradual inundation of the peninsula as sea level rose and precipitation increased (e.g., Milanich 1994; Miller 1992; Watts and Hansen 1988). This is thought to reflect global and regional scale processes, as temperatures warmed following the Last Glacial Maximum (LGM; approximately 20,000 22,000 y ears ago), and oceanic currents and atmospheric circulation accommodated the influx of glacial meltwater. This reconstruction is consonant with the greater American Southeast, which overall experienced an amelioration of climatic conditions following the L GM (Anderson et al. 1996 ; Anderson and Sassaman 2012:36 46 ) . A number of sea level reconstructions have been put forth for the Gulf of Mexico ( Balsillie and Donoghue 2004 , 2011 ; Donoghue 2011; Otvos 2004 ; Simms et al. 2007;
63 TÃ¶r nqvist et al. 2004; Wright et al. 2005 ) , western Atlantic ( Horton et al. 2009; Toscano and Lundberg 1998; Toscano and M a cintyre 2003 ) and globally (Siddall et al. 2003; Smith et al. 2011) . Although there are divergences and disagreements in the models, there is general agreement on th e pace and direction of sea level change. C ollectively they indicate that sea level was nearly 1 2 0 meters lower than present during the Last Glacial Maximum . This magnitude of reduction in sea level exposed vast expanses of the Florida Platform now inundat ed Coast would have been some 200 km further west than it is today ( Faught and Donoghue 1997 ) . The main source of inference about temperature and precipitation are sediment cores extracted from deep Florida lakes (Grimm et al. 1993; Grimm et al. 2006; Quillen et al. 2013; Watts 1969, 1971, 1975, 1980; Watts et al. 1992 ; Watts and Hansen 1994, 1998; Watts and Stuiver 1980 ). During the LGM , winter temperatures in Florida were, on average, 3 4 Â°C cooler than today (Willard et al. 2007). The Florida peninsula was also considerably drier at the peak of the last glaciation, with less precipitation and redu ced surface water availability. Watts (1971) estimated that the water table in north Florida was some 12 m lower than it is today. Only a held water during this interval and most of the rivers and wetlands were nonexisten t , at least in their current configuration . The combination of sharply reduced pressure gradients from the Gulf and Atlantic and lower recharge rates from precipitation no doubt reduced the potentiometric surface within the Floridan Aquifer . As a result, m any If the caverns and conduits now in place were
64 present at th at time, t hey may have been water bearing sinks or dry caves ( Rupert 1988) . However , the expanse of land exposed by reduced seas likely contained a num ber of artesian springs . Currently documented submarine springs would have been exposed on land, and others no doubt existed. Faure and colleagues (2002) have humans) d uring times of lowered sea level. A further consequence of reduced sea level during the LGM was the replacement of saline water in the Lower Floridan Aquifer by fresh water derived from precipitation (Morrisey et al. 2010). After this apex of glaciation , g lo b al climate warmed and the ice sheets receded. The melting of the glaciers released water into oceans and reduced the load on the continental land masses, resulting in isostatic rebound and an increase of sea level. However, this was not a uniform o r gra dual process. Donoghue (2011 ; also Balsillie and Donoghue 2011 ) argues that Gulf of Mexico sea level rise in the late Pleistocene was punctuated by several per iods of rapid change caused by glacial meltwater influx and/ or global climate change events. The first of these rapid changes in the Gulf of Mexico began at approximately 17,700 cal B.P., when the Gulf rose some 12 m over a span of 750 years. Warming of the Florida peninsula followed shortly after this, beginning in earnest by approximately 17,000 yea rs ago (Willard et al 2007). T he rate of both deglaciation and sea level rise increased markedly after 14,000 cal B.P. , during the BÃ¸lling AllerÃ¸d interval (14,700 12,900 cal B.P.). A second period of rapid change in the Gulf began at 14,300 cal B.P., when s ea level rose 24 m over a span of 500 years (Donoghue 2011:22) . This was likely caused by an input of meltwater
65 fro m the Laurentide ice sheet, delivered to the northern Gulf of Mexico via the Mississippi River drainage. The available records from Lake Tu lane and Lake Annie in southern Florida indicate that Florida was cool and dry during the BÃ¸lling AllerÃ¸d . However, p ollen and ostracod records in Tampa Bay contradict this, indicat ing rather that central Florida reached warm, mois t, near modern conditions at that time (Willard et al. 2007). Rising seas during the terminal P l eistocene had several ramifications for the Floridan Aquifer . Notably, this decreased the hydraulic gradient of the system, reducing the velocity of groundwater flow . Con sequently, salt water intruded into the Lower Floridan Aquifer, trapping freshwater in the Upper Floridan Aquifer (Morrisey et al. 2010) . Springs flowing on the emergent continental shelf were inundated, but the overall potentiometric surface of the aquifer was raised. T his elevated water levels in water bearing sinks on the interior and may h ave initiated spring flow in new places, again depending on local conditions. Following the BÃ¸lling AllerÃ¸d , the Younger Dryas (12,900 11,700 cal B.P.) marked a n interruption of post glacial warming and a rapid return to co oler , but globally variable, conditions (Broe c ker 2006; Broe c ker et al. 2010; Tarasov and Peltier 2005) . The onset of this interval coincide d with another rapid sea level rise in the Gulf of 27 meters over 600 years, from 12,900 12,300 cal B.P. (Donoghue 2011:22). Evidence for the impacts of the Younger Dryas i n Florida is contradictory , but suggests the peninsula may be out of phase with the rest of the northern Atlantic . P ollen records from La ke Tulane (Grimm et al. 2006) in southern Florida indic ate the dominance of pine at that time and point to deeper lake levels and warmer, wetter conditions than the oak -
66 dominated assemblages of the preceding and succeeding intervals . This is corroborated b y the record from Lake Annie. The pollen record from Sheelar Lake (Watts and Stuiver 1980) in northern Florida is likewise dominated by pine from ca. 13,000 11,000 cal B.P. Rather than an increase in year round temperature and precipitation, these records indicate increased summer precipitation and winter temperatures. This is primarily caused warming in the Gulf of Mexico and Atlantic, which in turn is best explained by changes in atmospheric circulation at the time (Donders et al. 2011). In contrast, the Tampa Bay record indicates progressively drier conditions over the course of the Younger Dryas (Willard et al. 2007). Lack of temporal correspondence and resolution in these records may be to blame. Donders and colleagues (2011) suggest that the Younger Dr yas contained both a warm/wet and cool/dry sub phase in Florida. Holocene Conditions The terminus of the Younger Dryas at 11,700 cal B.P. marks a return to post glacial warming and the onset of the Holocene. Sea level rise continued into the Holocene, with pulses beginning at ca. 11,000 and 8700 cal B.P. (Donoghue 2011; Balsillie and Donoghue 2011). Palynological analysis of Early Holocene sediments indicates that the upland forest was dominated by oak and grasses, with scattered pines. Different species of oaks can tolerate a variety of moisture conditions, but the combination of oaks and grasses in dicat es a dry oak scrub with prairie or savanna like openings, similar to that found on xeric hilltops and ridges in Florida today . However, t his interpretatio n is not uncontested. Despite the abund ance of grass pollen at Lake Tulane , low 13 C values of leaf waxes indicate a relative paucity of C4 plants (Huang et al. 2006) . Because the majority of Florida grasses are C4 plants, this contradicts the pollen evidence and suggests instead a dearth of grasses . Further, the
67 grass pollen assemblage has relatively low amounts of herbs, such as Ambrosia , that would indicate an oak grass savanna. An alternative scenario, then, is that the grass pollen is derived from emergent or damp ground grasses on the lake margin and thus is over represented in the core. The grass pollen is then reflective of local conditions and not representative of the regional vegetation. If this supposition holds, the uplands, then, may have contained closed woodlands reflective of moister conditions, and not a savanna /prairie. Regardless, the evidence suggests increasingly wetter conditions as the Holocene progressed. L acustrine sedimentation in shallower lakes began , or was re established after a hiatus , between 10,000 and 9000 cal B.P. (e.g., Donar et al. 2009; Watts 1969) . This is reflective of rising water tables, largely driven by sea level rise rather than increased precipitation. However, early Holocene w ater levels we re lowe r and more seasonal than today. For example, d iatom records from Lake Annie indicate that it was a shallow , seasonal pond by ca. 11,000 cal B.P., with a diverse flora surrounding it (Quillen et al. 2013). Water depth and hydroperiod (i.e., the propo rtion of the year that the lake was inundated) increased slowly over several millennia . Likewise, water became dee per in Little Salt Spring, a 70 m deep sinkhole/spring in Sarasota County, as the water table rose after approximately 11,000 years ago (Zarik ian et al. 2005). By approximately 8000 cal B.P. the rate of sea level rise attenuated as it approached 8 m below present levels (Donoghue 2011; Balsillie and Donoghue 2011) . T he available records indicate a broad transition in Holocene vegetation and (by proxy) temperature and moisture regimes in the Middle Holocene. F orest composition changed
68 from oak dominated to pine dominated , prairie herbs decline d, and wetlands developed. These changes are indicative of increased moisture and temperature. Time transgressive trends, from north to south, are apparent in these record s ( Grimm et al. 2001) , with the transition occurring earlier in northern Florida (ca. 8500 7500 cal B.P. ) than in southern Florida (after ca. 60 00 cal B.P.) . In northern Florida, pine was well established prior to 8000 cal B.P., coincident with an increase in swamp trees and shrubs (Watts and Stuiver 1980). Further south, Lake Annie witnessed a rapid increase in wa ter depth and pine pollen from 6000 5000 cal B.P., some 2 , 000 years later. Donders (2014) synthesized the evidence for increased precipitation in Florida from multiple lake cores across the peninsula. A clear divide is apparent between the north and south Florida records, both in the timing of increased moisture and the apparent underlying causes. These time transgressive trends between northern and southern Florida were a consequence of the differential effects of sea surface temperature and circulation pa tterns in the Atlantic and Gulf (Donders et al. 2011; Enfield et al. 2001; Kelly and Gore 2008). There is little evidence for marked increase in precipitation over the early to middle Holocene in northern Florida. Rather, vegetati ve changes indicative of g reater moisture were likely driven by increasing water table height resulting from sea level rise. In contrast, the southern Florida records show a marked increase in precipitation from 6000 5000 cal B.P. This was likely driven by an intensification of the El Ni Ã± o Southern Oscillation (ENSO) , which drives winter precipitation on Florida and extends the hydroperiod of seasonal wetlands , and, possibly by a shift in the position of the Intert ropical Convergence Zone ( ITCZ, see Donders 2014; Donders et al. 2011 ; Donders et al. 2005; Kelly and Gore 2008).
69 The timing of increased moisture in southern Florida after circa 6000 cal B.P. coincides with a further reduction in the rate of sea level rise as it reached near modern levels. It is thought that this stabiliz ation of sea level enabled the establishment of productive coastal habitats, such as estuaries (Donoghue 2011) . Peat accumulation began in wetlands across Florida after ca. 6 000 cal B.P. (Gaiser et al. 2001; Gleason and Stone 1994; Scherer 19 8 8; Wright et al. 2005) and by 5 , 000 years ago the Everglades reached from Lake Okeechobee to the Florida Keys (Willard and Bernhardt 2011) . By 4 , 000 years ago the seas were within 2 meters of present levels, possibly osc illating above and below modern conditions (Bals illie and Donoghue 2004). Likewise, the intensity and temporality of ENSO events and the position of the ITCZ , reflective of atmospheric and oceanic circulation patterns in the Gulf and Atlantic, approximated their modern configuration (Donders et al. 2005 ). The depth and condition of many Florida lakes was relatively consistent from 4 , 000 years ago to the near present, prior to nineteenth and twentieth century land alterations (Quillen et al. 2013), although there is some indication that water depth increa sed at ca. 2500 cal B.P. (Filley et al. 2001; Watts and Hansen 1994). Although this summary paints a picture of a gradual shift to modern conditions , this was likely not the case. Mayewski and colleagues (2004) have argued that Holocene climate change was far more chaotic and spatially heterogeneous than previously thought. They identify six rapid climate change events in the Holocene, through a synthesis of paleoenvironmental proxies (e.g., ice cores, lake varves). These events had global impacts, although they were not expressed the same in every region
70 and the particular effects in Florida are unclear. Recent evidence suggests that two, in particular, may have been particularly impactful. The 8200 B.P. event was a period of rapid cooling, less severe than the Younger Dryas , which elapsed over two to four centuries ( Alley et al. 1997; Barber et al. 1999 ; Kurek et al. 2004). This was likely caused by the final pulse of meltwater from the Laurentide I ce S heet. A second event at 4 , 200 years ago has been linked to an interval of severe d r ought in the mid latitudes that lasted up to 500 years (Arz et al. 2006; Booth et al. 2005; Magny et al. 2009 ; Rosenswig 2015 ). Again, the specific impacts this had on Florida are unclear, but these are only the largest of sever al intensifications, fluctuations, and reversals of the general trends outlined above. In sum, and at a broad scale, over the past 20,000 years the Florida peninsula has witnessed several dramatic changes as the glaciers retreated and the modern climatic r egime became established. A large swath of the Florida platform was lost to the rising seas. Sea level rise and increased precipitation likewise raised surficial water tables and pressure within the Floridan Aquifer, resulting in greater surface water avai lability and spring flow. However, these changes were not gradual and many reversals cold snaps, prolonged droughts, sea level regressions, and the like are apparent. Further, there wa s spatial variability in the timing, tempo, and magnitude of these fluct uations that is not reflected in continental, hemisphe ric, or global reconstructions. This underscores the need for local reconstructions to contextualize these models. In the follo wing sections I turn from broad scale climatic reconstructions to a discuss ion of the St. Johns River valley and the springs of Florida . The St. Johns River
71 valley is the geographic setting for much of what is discussed in the following chapters. Of particular relevance is the hydrology and geomorphology of the valley and the rel ation of these to the distribution of artesian springs. St. Johns River Valley The St. Johns River is the largest river in Florida and one of the few in the Northern Hemisphere to flow from south to north. It is a broad, shallow, b lackwater system that is tidally influenced with an estuary at its mouth (DeMort 1991; Miller 1998; Smock et al. 2005) . The St. Johns drains approximately 23,000 km 2 , or roughly one fifth the land area of Florida (DeMort 1991:97), and is one of the few large rivers in the southea stern U.S. not impounded by dams. This is primarily because of the low gradient of the river, which drop s only 8 meters in elevation over a course that meanders for some 500 km from headwaters near Vero Beach to the mou th at Jacksonville . As a result of th is low gradient , both water velocity and average discharge are low for a river of this size . The St. Johns also lies at a low elevation, having a maximum floodplain elevation of only 8 m above mean sea level ( amsl ) . As a consequence of its low elevation and gradient, the St. Johns is particularly susceptible to fluctuations in sea level. The physiography of much of peninsular Florida consists of north south trending highlands that roughly parallel the coasts in the northern two thirds of the peninsula (Sc ott 2011:28). Like m any of the larger rivers in Florida, the St. Johns is situated in a low swale or valley between these highland ri dges . These lowland s are generally poorly drained and house isolated wetlands or surface drainage networks (i.e., creeks an d streams). The St. Johns basin lies in the Eastern valley physiographic province ( Figure 2 5 ; following Cooke 1939) . This expansive, low -
72 lying area is bordered to the east by the Atlantic Coastal Ridges and to the west by the Osceola Plain, Deland Ridge, Crescent City Ridge, and Duval Upland. These ridge and swale complexes are the remnants of sea level high stands in the Pleistocene and earlier (Cook e 1945; Miller 1998; White 1970). It is thought that the highlands are relict barrier islands and/or beach ridges, and the lowlands were estuarine or shallow marine environments. The St. Johns River valley was repeatedly inundated during these high stands , to a depth of at least 7.5 m (Alt and Brooks 1965), and formed a system of elongated estuarine bays or lag oons behind barrier islands, connected to the Atlantic Ocean by widely spaced inlets (Belaineh et al. 201 2; DeMort 1991:97 99). When sea l evel later receded, salt water was drained from the lagoon and the valley transitioned to a freshwater drainage. The r iver itself is more reminiscent of a chain of lakes than the sinuous rivers commonly found elsewhere in the Southeast. The main channel of the St. Johns flows through 10 large lakes, and is fed by tributary streams from at least six more. The river basin c an be divided into three primary segments upper, m iddle, and l ower that reflect divergent geomorphology and hydrology (Figure 2 6 ) . Because the St. Johns flows from south to north, the u pper portion is the southernmost reach of the river. The upper St. Joh ns stretches for approximately 120 km from the headwaters near Vero Beach to the vicinity of Lake Harney and the confluence with the Econlockhatchee River, a major tributary. The headwaters of the St. Johns are derived from the Blue Cypress Marsh, an exten sive floodplain marsh in Brevard and Indian Riv er c ounties (Brenner et al. 1999 ; Brenner et al. 2001). The channel of the river is poorly defined and sheetflow through these wetlands dominates for the upper 50 km of the river (Kroening
73 2004). The channel o from t here flows through a series of lakes in broad, shallow depressions in the floodplain. Unlike other central Florida lakes, which are formed by solution processes and karst development (Kindinger et al. 1999; Schiffer 1998), these lake basins are remnants of the estuarine lagoon in which the St. Johns River formed (White 1970). The channel becomes more distinct between lakes Windsor and Poinsett, where the bottom transitions from peat dominated to sandy (DeMort 199 1 :100). From here the river channel adopts an anastomosing pattern: not a single flowing channel, but a series of lakes, lagoons and floodplain wetlands, connected by multiple channels separated by islands and bars. The c hannel is generally shallow, averaging 1 3 m deep, althou gh water level fluctuations are high (Smock et al. 2005). Prior to the twentieth century the vast wetlands of the upper St. Johns covered approximately 1800 km 2 (Brenner et al. 1999) . The soils in th ese wetlands are extremely fertile and well suited to ranching and agriculture. An extensive series of levees, canals, and other water control structures was installed to drain these soils and render them suitable for human use. By the 1980s, approximately 80% of the basin had been converted to pasture or other agricultural use, and 70% of the wetlands had been drained (Brenner et al. 2001; Kroening 2004; Sterling and Padera 1998). These water control efforts and the influx of nutrients from agric ultural ru noff have had numerous impacts on the hydrology and ecology of the region. Restoration efforts were begun in the 1980s by the St. Johns River Water Management District (SJRWMD) and U.S. Army Corps of Engineers to attempt to mitigate some of these impacts.
7 4 The middle St. Johns River is sometimes refer O ffset . he course of the St. Johns jogs west from Lake Harney to Lake Monroe, near Sanford, where it resumes its northward course on the western side of the Crescent City and De land Ridge s . The river jogs back to the east just south of Palatka. This is contrary to what would be expected from fluvial dynamics and the physiography of the region. In a headward consequent course, the river would be expected to flow from the headwater s to the mouth in a relatively straight line , following the late Pleistocene ridges adjacent to the Eastern Valley. However, a s the river makes its westward jog it flows around a block fault and passes between the DeLand Ridge and Geneva Hill, into a geologically older valley (Belaineh et al. 2012:4 75; Pirkle 1971). As a result, the channel is constrained on both sides by ridges and uplands the Crescent City and Deland Ridges to the east and the Marion Uplands and Mount Dora Ridge to the west before r eturning to the Eastern Valley. It is thought that the middle St. Johns formed earlier than t he segments above and below it (White 1970). The most likely scenario is that the middle St. Johns captured the headwaters during an early Pleistocene sea level l ow stand, diverting them from an eastern course into the western offset . Crescent Lake , then, is probably a remnant of the St. Johns channel that was abandoned when the river switched to its current configuration (Miller 1998) . As a result of this divergin g geomorphological history, the middle St. Johns also features distinctive hydrology. In places th e floodplain is more con s trained, and the river quickens as it meanders between well defined , elevated banks, sometime s forming multiple, braided channels (De Mort 1991:101) . In other places, the channel flows
75 through expansive, shallow lakes, including the aforementioned lakes Harney and Monroe, Lake Dexter, and Lake George. Unlike the lakes of the upper St. Johns , th e se reside in basins formed, at least partia lly, as a result of karst development. Seismic profiling of the northern portion of Lake George revealed the presence of subsidence sinkholes beneath several meters of recent lacustrine deposits (Kindinger et al. 1994 ) . The Ocklawaha River is the largest t ributary of the St. Johns, draining an area of over 7,400 km 2 (Livingston 1991:85). Like the St. Johns, the headwaters of the Ocklawaha is a series of wetlands and a chain of lakes (Lakes Apopka, Harris, Eustis, Dora, Yale, and Griffin). The Ocklawaha Rive r lies to the west of the St. Johns in the higher elevations of the Central Valley physiographic province. It roughly parallels the St. Johns over much o f its 113 km course. The Ocklawaha receives significant input from artesian springs and receives overflow from Orange Lake by way of Orange Creek. The spring fed Silver River, a major tributary, provides approximately half of the streamflow in the Ocklawaha (Livingston 1991:88). The river follows a northerly course , winding through extensive floodplai n wetlands before making an eastern turn and debouching in to the St. Johns downstream of Lake George. The l ower (northern) St. Johns River begins a t the confluence with the Ocklawaha River 4 , and, after returning to the Eastern Valley, is bounded to the wes t by the Duval Upland. The channel of t he St. Johns widens rapidly , expanding from 1.5 km wide at Palatka to 5 km wide near Jacksonville , and remains in a single channel without extensive intervening lakes (DeMort 1991:101). On reaching Jacksonville, the r iver 4 Geomorphologically, the l ower St. Johns more prop erly begins where the western offset ends, near present day Palatka and approximately 15 km north of the confluence with the Ocklawaha River.
76 makes an abrupt turn to the east and the channel narrows , essentially forming a drowned estuary reaching to the Atlantic Ocean at the mouth of the river . Both salinity and tidal influence increase as the river approaches the Atlantic Ocean, with high tides causing daily flow reversals. Tidal influence routinely reaches Lake George, some 160 km upstream, and can occasionally have an impact as far south as Lake Monroe, especially when low river flow and particularly high tides coincide (DeMort 1991). Lik ewise, a wedge of saline water commonly reaches 40 km upriver, sometimes penetrating 90 km or more (Smock et al. 2005) . Springs of Florida and the St. Johns River As of 2011, the inventory of springs recorded by the Florida Department of Environmental Prot ection (FDEP) included 1,014 entries (FDEP 201 2 ) . This inventory does not exclusively record artesian springs discharging from the Floridan Aquifer. A s mall number of water table springs and relict man made wells are included. Further, the list collapses s ome springs with multiple, distinct vents into spring groups, wh ereas others are recorded individually. In many cases these vents are hydrogeologically independent, drawing water from different portions of the aquifer and relying on different recharge basi ns , and so it is questionable whether they should be grouped together (Scott et al. 2002). Finally, the list includes some karst features that are not technically springs, such as river rise segments and karst windows. So , the exact number of springs in the state is somewhat nebulous. It is also ever evolving as new springs are documented and new hydrogeological research refines the grouping of spring vents. Within the state of Florida, m ost artesian springs are found in ar e as of well developed karst, w here the Floridan Aquifer is near the surface and overlying sediments are thin or absent (Figure 2 7) . Within these regions, springs tend to be located at lower
77 elevations, such as river valleys and coastal reaches, where the potentiometric surface of the aquifer is higher than the land elevation and water table. Springs are most frequent in the Ocala Karst District, a broad area encompassing most of northwest peninsular Florida (Scott et al. 2002; Scott 2011) . This area is som etimes referred to as the Ocala Uplift District (e.g., Brooks 1981) , an area of karst plains where the Floridan Aquifer is unconfined, with limestone at or near the surface. Springs are also found in abundance in the Dougherty Karst and Central Lakes D istr icts . The Dougherty Karst District is located in northwest Florida and, like the Ocala Karst District, is an area of well developed karst where the Floridan Aquifer is unconfined. The Central Lakes District lies to the east of the Ocala Karst District, whe re xeric sandhills thinly confine the Florida Aquifer. This is an area of active sinkhole dev elopment that encompasses a portion of the St. Johns River b as in (Brooks 1981:6) . The St. Johns River Water Management District maintains an independent digital sp rings inventory, last updated in 2014 (SJRWMD 2015 a ) . Combined, the inventories of the SJRWMD and FDEP document 205 springs in the basin s of the St. Johns and Ocklawaha Rivers (Figure 2 6) . This tally includes 14 water table springs that will be excluded from further discussion . It also includes cases where a number of spring vents are documented individually, despite comprising a recognized spring group. F or the sake of convenience I collapse the se in the following discussion. Treating spring groups as si ngle entities trims the total to 103 spring s . O f these, the majority (60 springs, or 58% of the total) feed into the middle St. Johns River basin. A further 32 discharge into the Oc klawaha River, and 11 into the l ower St. Johns River . No springs have been documented in the U pper portion of the St. Johns River basin .
78 The preponderance of springs in the middle St. Johns is a function of geography and geology. The Hawthorn Formation that confines the Floridan Aquifer in Northeast Florida varies in thickness ac ross the St. Johns River valley (Belaineh et al. 2012; Scott 1983, 1988). Recall that the middle St. Johns is offset to the west of the U pper and L ower thirds. This offset places the river within the Central Lakes District in a geologically older lowland v alley surrounded by uplands and rid ges. In this valley the Hawthorn Formation has been uplifted as a result of faulting and warping, exposing it to erosion. As a result, it is considerably thinner in the middle St. Johns (~ 15 m) than either the upper or l o wer segments, where it is up to 150 m thick. The thickness of the Hawthorn Formation in the upper and l ower St. Johns limits diffuse groundwater discharge and explains the relative lack of springs in these reaches (Connor and Belanger 1981; Spechler 1994). Further, incision of the river during Pleistocene sea level low stands cut through the thin Hawthorn Formation in segments of the middle St. Johns River (Stringfield and Cooper 1951) . The incised channel has since infilled with sediments (Kindinger et al 2000) , but it nevertheless enhances connection to the Floridan Aquifer in th is portion of the valley . Finally, the uplands ris e 10 20 m above the middle St. Johns River valley and are significant recharge areas for the Floridan Aquifer (B elaineh et al. 2012). The river valley itself is a significant discharge area for both diffuse leakage and spring flow, as the potentiometric sur face is 3 6 m above the level of the St. Johns River. This high recharge to the east and west coupled with disc harge within the valley creates a positive head difference that drives groundwater circulation into the valley from both
79 sides (Phelps and Rohrer 1987) . Many spring s are located near the contact between the upland recharge zone and lowland discharge zone. The inventory of springs in the St. Johns basin includes some of the largest in Florida, such as the f irst magnitude Alexander , Silver Glen, and Volusia Blue springs , and second m agnitude Salt, DeLeon, and Wekiw a. There are 13 additional second m agnitude a nd 83 third or lower magnitude springs. Several of the se spri ngs feed spring runs that form major tributaries of the St. Johns River (e.g., Alexander Creek, Wekiw a River) . Taken together, groundwater from the Floridan Aquifer comprises a significant portio n of the water flowing in the river, particularly in the middle St. Johns . Miller (1998:67) estimates that flow from the largest springs in the region (i.e., first and second magnitude) comprises one third the water flowing in the St. Johns River. Similarl y, groundwater flow modeling conducted for the St. Johns River Water Management District (Motz and Dogan 2004:41) suggests that direct discharge of groundwater from spring flow and diffuse upward leakage into the river channel accounts for 25% of the total stream flow in the St. Johns and Ocklawaha Rivers. As outlined in Chapter 1, the significant contribution of springs to the flow of the St. Johns River led Miller (1992, 1998) to conclude that , prior to the establishment of near modern climatic regimes some 6000 years ago , the majority of springs in the St. Johns River valley could not have been flowing. Without this groundwater input the St. Johns itself did not exist in any recognizable configuration . Therefore, the cultural changes evidenced by the co nstruction of freshwater shell mounds, middens, and the like were predicated on the onset of substantial artesian flow in the region. Indeed, they are a direct consequence of it. This chapter has reviewed the geological context ,
80 history, and physical param eters of spring flow , reviewed the evidence for climatic changes in the late Pleistocene and Holocene , and sketched the environmental framework of the St. Johns River valley and its springs . I have done this to provide a platform for the discussion that fo llows and to highlight the enormous geological and hydrological complexity of springs issuing from the Floridan Aquifer . This complexity underscores the deficiency of simplistic models that use global or hemispheric paleoenvironmental reconstructions to in fer the hydraulic status of Florida springs . Also unsatisfactory are red uctive archaeological a rguments that posit lock step stimulus and response dynamics to complex human ecological systems in the past . In the next chapter I unpack this model, evaluating it through multiple lines of evidence and setting the stage for a revised understanding of Florida springs archaeology.
81 Figure 2 1. The home of Mae Rose Williams descends into the crater, May 9, 1981. Photo by Barbara Vitaliano, Orlando Sentinel
82 Figure 2 2. Conceptual diagram illustrating the hydrology of the Floridan Aquifer System. Image courtesy of the St. Johns River Water Manag ement District
83 Figure 2 3. Generalized extent and degree of confinement of the Floridan Aquifer System
84 Fig ure 2 4. Generalized extent of areas recharging the Floridan Aquifer System in Florida
85 Figure 2 5 . Physiographic provinces in the vicinity of the St. Johns River valley , northeast Florida. Divisions follow those of Cooke (1939)
86 Figure 2 6. Major river basins and surface water features of the St. Johns and Ocklawaha River basins
87 Figure 2 7. Distribution of karst districts (following Brooks 1981) and springs recorded in the 2011 inventory of the Florida Department of Environmental Protection. No te: four submarine springs are not shown
88 Table 2 1. Spring m agnitude and e quivalent d ischarge r ate Magnitude Metric Units English Units 1 2.832 m 3 /s >100 ft 3 /s 2 0.283 2.832 m 3 /s 10 100 ft 3 /s 3 0.028 0.283 m 3 /s 1 10 ft 3 /s 4 6.308 l/s 0.028 m 3 /s 100 gal/min 1 ft 3 /s 5 0.631 6.308 l/s 10 100 gal/min 6 0.063 0.631 l/s 1 10 gal/min 7 0.473 3.785 l/min 1 pint/min 1 gal/min 8 < 0.473 l/min < 1 pint/min
89 CHAPTER 3 SPRING ORIGINS In central Polk County, a half kilometer west of the Peace River, a spring sits at the edge of a wooded swamp. The former owner of the spring, Dr. R.H. Huddleston, opened it to the public as a recreational retreat and health spa in 1883 (Figure 3 1). He installed bath houses, cottages, a diving platform, and a sma ll hotel (Bair 2011; Brown 1991:283). Originally known as DeLeon Mineral Spring, Huddleston renamed it renowned for its medicinal springs. At its height in the early twentieth century, the resort drew as many a s 10,000 picnickers, swimmers, and revelers on a single summer day (Figure 3 2). The spring basin held a pool of water some 60 m across with water issuing forth from a cave 6 m beneath the surface. In 1898 the rate of discharge was measured at 20 million g allons of water per day, making it what we would now call a second magnitude spring (Ferguson et al. 1947:140). The vulnerability of Kissengen became apparent in the late 1920s. At the time, it was thought that the Florida platform might contain untapped o il and gas deposits (Hine 2013:107). In July 1927, drilling began on an oil test well less than 100 m from the spring. When the drill had penetrated 70 m into the surface it intercepted the dropped alarmingly and the pool was nearly drained. The well was capped, restoring spring flow, and further testing was moved away from the spring (Peek 1951). A decade later the water level again began dropping, not suddenly , but slowly and persistently. Drilling was not the culprit. By this time oil prospectors had largely lost interest and moved on to greener pastures. The problem was extraction of another kind.
90 Kissengen Spring lies in the heart of the Central Florida Phosphate District, one of the mos t productive phosphate deposits in the world 1 (Hine 2013:157 158). Here, metallic surface sands. Phosphate mining has several detrimental consequences, such as the release of radon gas and accumulation of waste products in massive phosphogypsum stacks, but of particular note is the water intensiveness of mining. Water is a key component in the chemical processing that transforms phosphate into phosphoric acid and, ultimately, fertilizer. The rate of fresh groundwater with drawal in Polk County began increasing progressively after 1937 (Peek 1951). Nearly 70 % of the water used at that time went to phosphate mining. The output of Kissengen Spring declined in step with this until, o n a winter day in 1950 , the final trickle of water pulsed from the spring . What had once been a popular gathering spot for residents and visitors alike was transformed into a stagnant pool choked with dog fennel (Figure 3 3) . And so the death of a spring is not unknown: Kissengen became the first maj or spring in Florida to cease flowing as a result of groundwater withdrawals (Peek 1951; Rosenau et al. 1977:307). But what of the birth of a spring? How does a spring come into being, for example as pressure increased in the Floridan Aquifer over the late Pleistocene and Holocene? The birth of a spring, or rather, of springs writ large, is the subject of this chapter. Johns River valley is demarcated by the appearance of large shell mounds on the 1 Florida accounts for some 30% of global phosphate production. The Port of Tampa exports more phosphate fertilize r than any other port in the world (Hine 2013:159 163)
91 banks of the river some 7,400 years ago (Beasley 2009; Endonino 2010; Randall 2010; 2013; Wheeler et al. 2000). These constructions are generally taken to be an index for the inception of a riverine adaptation and more sedentary settle ment (Milanich 1994:87). It has been argued that shell mounds and the practices that gave rise to them were that today feed the St. Johns River (Miller 1992, 1998). Although surface water gradually increased over the course of the Early and Middle Holocene as sea level rose, the St. Johns River likely did not attain its modern configuration until it began receiving fresh groundwater from artesian springs. Under t his m odel, the inhabitants of the region would have mapped on to the nascent aquatic habitats, exploiting shellfish as a food resource and depositing their shells on the banks of the river. Miller (1998:97) views this as a rapid change in the landscape: T he flo oding of the St. Johns River Basin around 5,000 years ago must have been a sudden event, in geological time. Because the artesian flow warning that the river basin was about to double in size. Interpreted in this way, the initiation of spring flow appears as an ecological founding event a rapid restructuring of the landscape that provided the necessary and sufficient conditions for a specific adaptive strategy. Intuitively this e xplanation makes sense; people could not collect and deposit shells if the habitats for shellfish were not present. However, the problems with it are threefold. First, it perpetuates a perception of hunter gatherers as ahistorical and reactionary. That is, change in human cultures is reduced to a process of adaptation to external stimuli with no consideration of internal dynamics as the driver of stability or change. Second, this explanation assumes that shell mounds are simply middens, the
92 palimpsest refus e of many small meals. By corollary it denies any intentionality or foresight to the shell mounders themselves and unnecessarily imposes our own biases onto our subjects of inquiry. Third, it has simply not been empirically tested. The onset of spring flow is hypothesized to follow Holocene sea level rise and initiate changes in aquatic ecology and human interactions with the river. But the timing and synchronicity of these events and the implications of a sudden influx of freshwater to the system have not been investigated. Recent archaeological work at several springs in the St. Johns River valley is casting light on these questions and causing us to rethink environmental and cultural histories of the region. In this chapter I discuss the results of both f ield investigations and hydrological modeling of the onset of artesian spring flow in the region. Mounting ecology, both in terms of the tempo with which they unfolded and, m ore importantly, the human response they elicited. In the following I first review the archaeology of the Mount Taylor period, with emphasis on the changing form of shell constructions through time and the association of these sites with artesian springs i n the St. Johns River Valley . After this, I offer discussion of the because the eventfulness of artesian and cultural changes in the valley is at issue here. Miller posits the onset of spr ing flow as a (geological) event, but it is not altogether clear if this is commensurate with commonsense understandings of the term or if it can be addressed with archaeological evidence.
93 With this background in place, I then turn to a GIS based model of spring flow in the middle St. Johns River basin. This model examines how springs at differen t landscape positions would be a ffected by a reduced potentiometric surface of the Floridan Aquifer. The model begins with aquifer pressure reduced to such an exten t that spring flow is absent in the middle St. Johns basin. T he potentiometric surface is incrementally increased until pre development levels are reached. The model is a heuristic device and not intended to mimic the reaction of the Floridan Aquifer to sp ecific sea level or precipitation changes. Rather, it allows me to evaluate whether the onset of spring flow valley wide might have been a sudden and widespread event. The implications of this model are evaluated with empirical evidence from two archaeolog ical sites. Site 8MR2322 surrounds Salt Springs in the western portion of the valley. Investigations conducted in 2009 of a subaqueous Mount Taylor midden in the spring channel provide insights into both the timing of initial spring flow and the changing s pring side practices of Mount Taylor people. Fifteen kilometers south of Salt Springs lie s a plethora of archaeological deposits surrounding Silver Glen Springs. I focus here on Locus A at site 8LA1 West, a Mount Taylor shell ridge paralleling the spring r un that is the subject of ongoing field investigations. Recent archaeological evidence points t o pre Mount Taylor activity , and cores extracted from the wetland adjacent to Locus A provide data pertinent to changing hydrological conditions in the spring. I close with a discussion of the implications of these data to the hypotheses described above. Was the onset of spring flow rapid in the middle St. Johns River rapid and widespread? Was it an event that precipitated changes in the lifeways of local
94 inhabita nts? I argue that it was not and set the stage for an alternative explanation, developed in the following chapters, that springs were critically important for people inhabiting the region, but not for reasons traditionally offered. The Mount Taylor Period in Northeast Florida The Mount Taylor archaeological culture encompasses a suite of material objects and practices centered on the St. Johns River valley (Figure 3 4). Regionally, it falls within the Archaic period of the American Southeast (Sassaman 2010) . The Archaic has long been characte rized as a transitional era of mobile, egalitarian hunter gatherer bands gradually adapting to post glacial conditions and leading a simple lifestyle that remained relatively unchanged for some 8 , 000 years ( e.g., Smith 1 986; Steponaitis 1986). Regionally, the Archaic is generally divided into Early (11,500 8900 cal B.P.) Middle (8900 5800 cal B.P.) and Late (5800 3200 cal B.P.) sub periods (Figure 3 5) . These divisions are recognized largely on the basis of shifts in tech nology along with presumed reductions of settlement mobility and increased intensiveness of resource use over time , although the precise timing of these varies considerably both throughout the south eastern U.S. and within the state of Florida. Broad brush strokes generally paint a picture of increasing population, reduced settlement mobility, and subsistence specialization as communities adapted to ameliorating environmental conditions. However, this picture of gradual adaptation is being overtu rned by recent research that emphasizes the importance of sociality, interaction, identity, and history to Archaic communities (e.g., Cl aassen 2010; Emerson et al. 2009; Gilmore 2015; Randall 2015 ; Sassaman 2010). Investigations into the Early Archaic peri od (ca. 11,500 8900 cal B.P.) typically focus on environmental changes at the Pleistocene to Holocene transition, and
95 concomitant human adaptation to new climatic regimes. The Early Archaic is recognized by a shift in the form of diagnostic hafted bifaces. Lanceolate forms, characteristic of the earlier Paleoindian period, were no longer manufactured by approximately 11,000 cal B.P. (Anderson and Sassaman 2012:71 72). In their place appear a variety of side and corner notched forms and bifurcate based haft ed bifaces. The most common of these in Florida are Kirk and Bolen. Early Archaic communitie s in Florida were likely highly mobile and may have been tethered to sources of freshwater and toolstone. Dunbar (2002, 2006) has argued that a multi century dry sp ell termed the Bolen drought , spanned the Paleoindian to Early Archaic transition ca. 11,500 10,800 cal B.P. However, b oth sea level and precipitation increased over the course of the early Holocene, so the constraint posed by freshwater availability likel y lessened , opening up new areas for exploitation (Donoghue 2011; Milanich 1994:62 63). Similarly, Middle Archaic cultural developments regionally have been explained by reference to the warming and drying trend variously referred to as the Hypsithermal, A ltithermal, or Mid Holocene Climatic Optimum (e.g., Brown and Vierra 1983; Marquardt and Watson 2005). After ca. 8000 cal B.P. there was an increased focus on aquatic resources across the Southeast, as evidenced by the appearance of shell middens and mound s along the coasts and interior river valleys (Dye 1996). Also at th at time, earthen m onuments were first constructed in the Lower Mississippi River Valley and Florida ( Piatek 1994; Saunders et al. 2005). The Middle Archaic is marked by the disappearance o f notched hafted bifaces and the appearance of stemmed varieties. Kirk stemmed/serrated is perhaps the earliest of these, in use by approximately 9,000 cal B.P. in Florida (Faught and Waggoner 2012). Following this are a variety of local
96 variants (Levy, Al burial tradition, best known in Florida from the Windover archaeological site in Brevard County (Doran 2002), where at l east 168 individuals were interred in saturated peat deposits sometime between 9000 and 8000 c al B.P. Pond mortuaries from that time have been documented at other locations in Florida as well (Clausen et al. 1979 ; Purdy 1991:167 177; Wharton et al. 1981). The Late Archaic period is marked regionally by the end of the Hypsithermal and establishment of near modern climatic regimes and sea level (Anderson and Sassaman 2012). Regionally, this interval is characterized by long distance exchange and interaction centered on Poverty Point, Louisiana (Gibson 200 1 ; Kidder 2010). The Late Archaic also witnessed technological innovations, notably the initial dissemination of pottery in the region. Pottery appeared by ca. 4600 cal B.P. in Florida ( Gilmore 2014; Sassaman 2004). This pottery, among the earliest in North America, was tempered with Spanish moss fibers and is locally referred to as Orange. Despite the addition of pottery, little else seems to differentiate the Orange period in Florida from what came previousl y (Milanich 1994). Settl ement and subsistence patterns we re largely similar, with a continued focus on the aquatic resources of the coasts and interior rivers and wetlands. However, a handful of locales in the middle St. Johns River valley became the focal points for massive gatherings that drew people from across the Florida peninsula (Gilmore 2014; Randall et al. 2014). The focus of this chapter is t he preceramic Archaic Mount Taylor period of northeast Florida. Mount Taylor was originally defined by Gogg in (1947, 1952) as one of
97 the earliest recognizable cultural complexes in the region. Dating from ca. 7 4 00 4600 cal B.P., it spans the Middle and Late Archaic periods and is traditionally defined by the inception of large, compl ex deposits of freshwater sh ell 2 (Beasley 2009; Endonino 2010; Randall 2010 , 2013 ; Wheeler et al. 2000). The dominant species of mollusk in these deposits is the banded mystery snail ( Viviparus georgianus ). Remains of the apple snail ( Pomacea paludosa ) and freshwater bivalves ( Unionidae sp. ) are commonly present as w ell, often in discrete deposits. A suite of artifact types is characteristic of Mount Taylor assemblages, al though to some degree these crosscut boundaries with the preceding Early Archaic and subsequent Late Archaic Orange period. So it is the appearance of shell sites that delimits the onset of the Mount Taylor period, and the appearance of pottery that signals its terminus. Mount Taylor assemblages are typified by bone, shell, and antler tools, with varying amounts of lithic materials (Wheeler et al. 2000) . Where present, lithic hafted bifaces are largely consistent with the Newnan horizon (Bullen 1975). Bone and antler tools include both decorated and undecorated forms, such as awls, pins, socketed projectile point s, and net fids (Byrd 2011; Wheeler and McGee 1994a). Marine shell was transformed into receptacles, beads, and woodworking tools (e.g., axes and adzes). Later Mount Taylor assemblages often contain items that originated far from the St. Johns River valley , such as soapstone and greenstone from the interior Piedmont, Strombus gigas from southern Florida, and stone beads that evidence connections as far away as Mississippi ( Endonino 2009, 2010) . 2 cally and
98 Recent work in the middle St. Johns River valley has clarified our understanding of Mount Taylor material culture and chronology. One recent development is the definition of two phases within the Mount Taylor period (Beasley 200 9 ; Endonino 200 9 , 5700 cal B.P.) and the Thornhill Lake phase (5700 4600 cal B.P.). This division is based primarily on the recognition of changes in mortuary and exchange practices. Randall (2013) has further refined this vary in content, context, and form. The first two of these correspond to the Early Mount Taylor phase, and the third is commensurate with the Thornhill Lake phase. The Early Mount Taylor Phase (7400 5700 cal B.P.) The Early Mount Taylor phase begins with t he onset of intensive freshwater shell deposition in the St. Johns River valley at approximately 7400 cal B.P. Again, Miller (1992, 1998) attributes this practice to sea level rise and the onset of artesian flow in the St. Johns River valley . Although the hallmark of the Mount Taylor period is large shell mounds and ridges, measuring as much as 200 m long and 5 m high, the earliest shell deposits were less conspicuous on the landscape. These have been studied both in isolated contexts and beneath larger rid ges. The evidence to date indicates that , at the onset of the Early Mount Taylor phase , shell deposits consisted of habitation spaces arrayed, perhaps in a linear fashion, along waterways. The Hontoon Dead Creek Village (8VO215) comprises a series of small , regularly spaced house mounds fronting a relict m in maximum dimension and ~50 cm in height, and consist of thin layers of crushed and whole shell that likely correspo nd to periods of occupation and abandonment. The
99 material inventory is consistent with habitation activities, as food remains, ash, charcoal, and shell tools are a bundant . Beginning approximately 7,200 years ago, some of these domestic spaces were apparent ly covered over and transformed into linear or crescentic shell ridges. These imposing sites once were abundant along the middle St. Johns, numbering in the dozens, if not hundreds, prior to twentieth century mining for road fill and fertilizer. Only a few remain relatively unscathed, but those that do attest to dramatic transformations in depositional practices ( Randall 2010 ; Sassaman 2012, 2013; Sassaman and Randall 2012). The Hontoon Dead Creek Mound (8VO214) is a tear drop shaped ridge, approximately 14 0 m long with an offset summit approximately 5 m high ( Randall and Sassaman 2005:83 106 ). The nearby Live Oak mound (8VO41) is similarly arrayed, measuring 120 m long and 5 m high ( Sassaman 2003 a :69 90 ). Both ridges are located on former channels of the St . Johns that have since been abandoned. The presence of buried shell deposits beneath a meter of muck adjacent to these ridges attests to rising water levels and wetland aggradation over the past seven millennia. Although these are discrete sites, their co nstituents are strikingly similar and I discuss them collectively below. The basal components consist of midden deposits that are partially or wholly saturated today. These are the only portions of either ridge that exhibit evidence of domestic or habitati on related activities. Their disposition beneath the water table hampers recovery and interpretation, but they appear to be similar in content and form to the house mounds of the Hontoon Dead Creek Village. These basal middens we re subsequently covered wit h a thick (1.5 2 m) deposit comprised largely of clean, whole
100 Viviparus shell . This shell layer has a conspicuous lack of both non shell matrix and vertebrate fauna or other artifacts. These thick shell layers are interpreted as intentional s that sealed off former domestic spaces and marked a transition in site use (Sassaman and Randall 2012). Upon this massive cap are relatively diminutive layers of unburned and burned shell. iscent of stacked living surfaces documented elsewhere, but i ndicators of daily living (e.g., vertebrate faunal remains, lithic or shell tools, charcoal) are relatively sparse . This, c ombined with their relatively rapid construction (decades to a few centuries at most) and a lack of indicators of significant depositional hiatuses ( i.e., pedogenesis), has led to the conclusion that these result from a series of restricted, but repeated, communal events aimed at renewing mound surfaces and drawing on pla ces as historical resources to confront unanticipated change ( i.e., changing hydraulic conditions evident in the abandonment of the river channel; Sassaman and Randall 2012). ) unt il approximately 6350 cal B.P. Episode II (6350 5700 cal B.P.) shell deposition consists of the abandonment of earlier sites, likely coincident with wetland aggradation and channel migration facilitated by the relative stabilization of sea level prior to 6 000 cal B.P. Shell sites at this time take on two divergent forms. The first of these are dedicated mortuaries constructed of shell and sand. N either the Live Oak nor Hontoon Dead Creek mounds have produced evidence for mortuary use , although mound cores w ere not penetrated, and Episode I mortuary practices remain unknown. Episode II mortuary practices have been well documented at t he Harris Creek site (8VO24) on
101 Tick Island . The site was first visited by C.B. Moore in the nineteenth century, and subsequent ly excavated by Ripley Bullen in 1961 (Aten 1999; Jahn and Bullen 1978). Aten (1999) has reconstructed the complex depositional history of this site, which includes multiple shell fields and ridges covering five acres . Over 184 Mount Taylor burials were re covered from the largest shell ridge at the site, which measured over 10 m tall. These burials, likely a small fraction of the total, were located in two discrete mortuary deposits (Aten 1999:170). Like the ridges described above, Harris Creek began as a r esidential space that was subsequently capped with shell and repurposed (Randall 2010:319 325). The major mortuary feature is white sand, which was used to inter burials emplaced atop and within the shell deposit. Above this is a charcoal rich layer with p ostholes and other features indicative of a charnel house (Aten 1999:147). Another mortuary component with multiple interments lies atop this, which in turn is capped with deposits of shell and earth. Isotopic analysis indicates that the individuals interr ed in the mound included some who were local to the St. Johns region and others from as far away as Tennessee and Virginia (Quinn et al. 2008; Tucker 2009). The second major form of site documented for Episode II is shell ridges similar in form and scale t o earlier Episode I ridges but with ample evidence for daily habitation. These have been documented at sites such as the Silver Glen (8LA1W) and Thornhill Lake complexes (8VO60 63) and occur along extant bodies of water, rather than relict channels. Locus A at 8LA1W is the remnant of a Mount Taylor shell ridge some 200 m long (Randall 2010; Sassaman et al. 2011; Sassaman and Randall 2012). Although much of the site was destroyed by twentieth century shell mining, surface relief and subsurface exposures have revealed as much as 4 m of stratified deposits. Three
102 macro units can be defined, similar to those of Episode I ridges (Sassaman and Randall 2012). Once again this ridge began as a domestic or habitation space that was subsequently capped. However, in thi s case the cap was not shell but relatively homogenous brown sand. The upper deposits above this cap consist of alternating layers of shell and earth, interspersed with crushed shell surfaces and occasional pit features. These are similar in form and conte nt to the shell nodes documented at Hontoon Dead Creek Village, containing ample amount s of vertebrate fauna and tools of bone, stone, and shell. Notable in these assemblages is the inclusion of marine shell vessels and substantial quantities of lithic deb ris and tools, both of which are lacking in Episode I deposits (Randall 2013). In sum the Early Mount Taylor phase is characterized by changing patterns of shell deposition and landscape use over 1,700 years. Initial habitation spaces were repeatedly capp ed and conscripted for new uses. During Episode I (7400 6350 cal B.P.) this transformation entailed cyclical, communal events of mound surface renewal. Over relatively short periods of time these new depositional practices built up imposing shell ridges al ong abandoned, relict channels of the St. Johns. During Episode II of shell deposition (6350 5700 cal B.P.) earlier shell ridges were abandoned as loci for emplacing shell and new ridges were formed. In Episode II ridges, b asal domestic spaces were likewis e transformed towards diverging ends. In some cases these became the loci for interment of the dead, in others for the ongoing sustenance of the living. Thornhill Lake Phase (5700 4600 cal B.P.) The Thornhill Lake phase is coincident with Episode III shell deposition (5700 4600 cal B.P.) and, like the earlier transition from Episode I to II, marks the cessation of
103 shell deposition of some earlier shell ridges and the establishment of new shell sites (Randall 2013). S hell deposition was apparently restricted to fewer locales than previously, but the resulting mounds and ridges tend to be larger than ever . However, continuity is evident in the arrangement of domestic spaces, reliance on aquatic resources, and use of a similar tool set. The primary distinguishi ng feature of the Thornhill Lake phase is a novel mortuary tradition involving the construction of sand burial mounds and the widening of exchange networks to encompass much of the lower Southeast (Beasley 200 9 ; Endonino 200 9 ). Thornhill Lake mortuary pr a ctices contrast with earlier interments in shell , like those at Harris Creek . Conical sand mounds built at this time are the earliest earthen monuments in the region. Well studied examples are the Bluffton Burial Mound and Thornhill Lake mounds A and B. A t Bluffton a single individual was interred in a conical mound some 20 m in diameter and 5 m in height. The mound was composed of brown sand and shell beneath layers of organic muck and redeposited shell midden (Randall and Tucker 2012; Sears 1960). No arti facts or other inclusions were present with in the burial. A single radiocarbon assay places the burial early in the Thornhill Lake phase, at 5660 5320 cal B.P. (Randall and Tucker 2012). As documented in recent excavations by Endonino (200 9 , 2010 ) , the Tho rnhill Lake Complex consists of multiple shell ridges and at least two conical sand mounds. The earliest ridges began accumulating during Episode II, between 6300 and 5600 cal B.P. The conical sand mounds were initially excavated by C.B. Moore ( 1 894a, b ), who noted a lack of pottery in the mortuary mounds and significant quantities of items of nonlocal origin accompanying burials . The larger Mound A (3.4 m high) contained at
104 least 42 burials in alternating lenses of brown and white sand. Mound A was cons tructed late in the Thornhill Lake phase, likely postdating 4840 cal B . P . Mound B was built atop an Episode II ridge and is composed of brown sand with sparse shell. It contains seven burials and wa s constructed sometime after 5600 cal B.P. Long distance r elationships with denizens of the interior Southeast ar e indicated by the inclusion of items that originated from far flung locales. These items appear in both mortuary and non mortuary contexts in the middle St. Johns River valley . At Thornhill Lake t hese included bannerstones, polished stone beads and pendants, and marine shell beads. Stone items were produced of materials not available in the Florida peninsula (e.g., greenstone, steatite, jasper). Wh ereas the bannerstones likely originated in the Piedmon t region of northern Georgian and Alabama (Endonino 2009:157 158; Sassaman and Randall 2007), polished stone beads are reminiscent of those produced in Mississippi and hint at connections even f a rther afield (Randall 2010:174). The Thornhill Lake mortuarie s thus contrast with the Bluffton Burial Mound in that they encase multiple individuals and exotic grave inclusions and were apparently used over a longer period of time. An earthen mortuary, roughly coeval with Bluffton and Thornhill Lake, has been docume nted at the T omoka Mound Complex (8VO81). Nonlocal items were also recovered there, notably a cache of bannerstones (Piatek 1994). Randall (2010:3 17 321) has suggested that Thornhill Lake mortuary mounds were constructed in a single stage over few individu als, and thus contrast with earlier shell mortuaries and mounds n ot only in terms of the principal construction medium, but also with regard to the temporality and sociality of mound construction. That is, wh ereas
105 earlier shell mortuaries and ritual mounds were apparently communal, integrating affairs erected through cyclical acts of deposition, later Mount Taylor sand m ortuaries appear more eventful and exclusionary, and may have been less labor intensive. Sassaman ( 2012, 2013) has further argued that chan ging Mount Taylor ritual practices are underlain by a common thread, an historical ontology that viewed water as both the source of life and a medium of renewal. Taking the position that water was an important symbolic medium in past ritual practices, Sass aman argues that the shift from Early Archaic pond burials at places like Windover pond (Doran 2002) to shell burials, occurred at a time when sea level was still rising rapidly, and changes in the loc al hydrology (i.e ., inundation of the valley) would hav e been perceptible over the course of a human lifetime. This may have been at odds with elements of cosmogony that held that earth and life emerged from water (i.e., the earth diver myth), and had a profound impact as pond mortuaries became inaccessible un der rising waters. The Early M ount Taylor practice of ritualized shell deposition is a response to the contradictions of rising water inasmuch as it represents a conversion of ponds into mounds. Shell, in this sense, is metaphorical water. Capping mortuari es with shell may have recapitulated the emergence of earth and life from water. The couplet s of clean and burned shell or shell and earth witnessed at Hontoon Dead Creek and Live Oak mounds reenacted this cosmogonic myth at a smaller scale. Ritual action involving shell was thus a strategy for imposing greater control over unpredictable yet symbo lically charged water and became an important medium for community building. The second transformation of mortuary practices, involving the use of sand, occurred as rates of sea level rise attenuated and hydrology became more stable
106 (Sassaman 2012, 2013). Water levels were thus more predictable and environmental change no longer perceptible at the generational time scale. Shell was no longer needed as a ritual medium to intervene against erratic hydrology. In addition to the shifting ritual media, mounds be came smaller in scale and more eventful in their construction, in that they no longer evidence repeated ritual depositions. Further, Sassaman suggests that the inclusion of nonlocal items indicates contact with diverse Thus, shell an d earth may have become signifiers of original/native and newcomer/non native, respectively. Discussion Several observations can be made with regards to the history of Mount Taylor shell deposition outlined above and the role played by springs in the onset of this practice. First, the underlying theme throughout the Mount Taylor period is the covering over of prior habitation spaces and the establishment of new depositional regimes atop them. This indicates a persistent concern with places as historical res ources that were drawn upon in the siting of locales for communal mortuary or renewal rites, or the re establishment of everyday dwelling. Second, at least in the case of Episode I ridges like Hontoon Dead Creek and Live Oak mounds, the triggering event th at precipitated the mounding of shell is not the birth of ecological productivity drawing people to the river, but rather the demise of places . Rising waters, wetland aggradation, and channel migration stranded existing habitation spaces and inundated exis ting mortuary spaces (i.e., pond burials; Sassaman 2012, 2013; Sassaman and Randall 2012:73). As a result, new dwelling spaces were required for both the living and the dead. It is unclear if similar processes or events preceded the erection of later Episo de II shell ridges, or if some other factor is at play.
107 The Mount Taylor concern with history is visible here as well, pre existing habitation spaces were capped with shell or sand prior to renewed habitation or the interment of the dead . This cap marked a transition, the end of one stage in the history of a place and the beginning of another. We currently lack the data to say with certainty whether these habitation sites were in use at the time of their capping, or if they were abandoned and then transform ed at a later juncture. Finally, it is notable that no Episode I Mount Taylor shell ridges have been documented proximate to springs in the St. Johns River basin. This could be explained if springs were not yet flowing, but if so this undermines the hypoth esis that shellfish exploitation is dependent on spring flow. Or it may be that springs were flowing but people simply did not visit them at this time, perhaps because their flow was intermittent or unreliable. This again undermines hypotheses that charact erizes spring flow as eventful, precipitating a rapid transformation in the ecology of the St. Johns. Data pertinent to these alternatives will be brought to bear below. But first, the following and clarifies how I conceive of them here. Events and Non Events Studying events in the deep past is tricky business. The distortion of time and the palimpsest nature of the material world conspire to blur the residues of discrete happenings in the past. Or rather, they blur the totality of events that might make up a thick history of a given people, time, or place. Where individual events have been identified and studied they are typically broad transformations in material culture that have the appearance of rapidity. Often these are correlated with changes in climate
108 (typically temperature and precipitation) at the local or regional scale that assume a ca u s al role in explaining the eventful changes in human behavior. When considering events in the past ou r common sense understanding is often tacitly invoked. A n event is generally considered to be an occurrence or happening that is recognized as having significance. To paraphrase Sahlins (1985:xiv), an event is a happening interpreted and imbued with meanin g. Typically we conceive of events as having a finite duration, though their boundaries can be difficult to demarcate. Events may happen rapidly, even instantly, or they may unfold over a protracted interval. Likewise varying is the spatial scale of events , the number of people who experience d them, and their materiality, with some events leaving little material evidence of their passing and others significantly more. But regardless of scope, it is recognized that something of consequence transpired; it was affective to those who experience and/or interpret ed it. The significance of an event is derived from its appropriation and interpretation in a given cultural order (Sahlins 1991:45). Eventfulness is thus dependent on cultural context. Further, the affect ive quality of events extends them through space and time; their repercussions give them life beyond their experiential temporality (Sahlins 1985, 1991; Sewell 2005). And if the event itself varies in scale and extent, so too do the consequences. In contr commonplace understanding is a useful starting point. To indicate that an occurrence was non eventful is to assert that nothing of consequence transpired. Whatever fallout may result is non descript and insignificant. This again draws our attention to the importance of cultural context. But if an event is a happening interpreted, what, then, is
109 a non event? Can we say that a non event is a happening not interpreted? This clearly canno t be the case. Non events are interpreted happenings as well. They may be happenings that are dismissed, ignored, or simply overlooked, but whatever the case they are interpreted within a cultural context. However, in the case of the non event the happenin g is deemed inconsequential. Fogelson discusses several ways that non events come into being. First, non events may arise from differential recognition. In other words, for any given happening, some may recognize an event where others do not. Similarly, th ough there may be agreement that an event transpired, the significance or consequences of the event may be debated. Another form of non event is the imagined event. This is an event that could or should have happened but did not. Nevertheless, imagined eve nts affect happened in the past or are anticipated in the future. A subtype of imagined events is longer historical explanations, though they are fictions in the sense that the events recounted event is the latent event, an event that has been overlooked because it does not fit into existing questions or narratives. Finally, denied events are events that are so traumatic that their recollection is repressed and they are deliberately forgotten. to provide a typology of non events, but rather to show that there are multiple vehicles, justifications, and circumstances that contribute to the interpretation of any happening as significant or insignificant, and thus
110 as eventful or non eventful. This highlights the relational quality of events and non events, as both are constructed in the interplay between objective conditions and subjective perception and interpretation. The eventful is not solely dictated by what it plucked out of the ether. Events, then, are an emergent quality of the human inhabited world. It is important to emphasize that the interpretation of incidences as significant or ecognized, defined, evaluated, and endowed with meaning differentially in different cultural Indeed, the very distinction between event and non event presupposes multi ple interpreting observers. It is impossible for an occurrence to be recognized as a non event without it also being interpreted as an event by some other observer. These may be contemporaries with competing interests and differentials of power that dictat e their ability to interpret occurrences. Or the observers may be separated in space or time, as in the distinction between events recognized by observers who lived through them and those recognized by researchers or analysts looking in (or back) from the outside [2015:7 8]). The more pertinent question, then, is not whether some occurrence was eventful or not, but rather for whom was it an event or non event? Further, non events can become events retrospectively and thus are influenced by subsequent happenings. The distinction involves an attribution of meaning. What we recognize as events in hindsight may not have been interpreted as such by those who experien ced them. What we see as events, then, may actually have been non events at
111 the time. If we are to avoid grafting our understanding of events onto the past then it is insufficient to simply demonstrate that some happening transpired over a relatively short interval of time (i.e., perceptible within a human generation or two). Rather, it is necessary to demonstrate if and how that happening was acknowledged by and incorporated into existing cultural traditions. These distinctions then, (event/non event and e xperiential/analytical events) help disentangle events that may be recognizable at the geologic time scale from those recognizable at the human perceptual scale. All of this redoubles the challenge facing archaeologists, who must tease out events from thei r material residues while continually guarding against the bias of our happening. One approach is to focus solely on historical event s. As defined by Sahlins ( 1985, 199 1) these are the happenings that cause a rupture in the articulation of structures and thus are moments of significant change. To some these are the only events worth considering (e.g., Beck et al. 2007), although there is some debate whether the re articulation of structures is limited to moments of existential chaos, or if it can be realized in mundane, daily practices (e.g., Gillespie 2007; Gilmore 2015). Further, focusing solely on transformation disregards the eventfulness of structural reproduction, a point Sa hlins (1991) recognized but did not resolve. Lucas (2008) takes a slightly different approach, arguing that archaeologists should focus on the materiality of events. However, the sense of materiality employed by Lucas is not simply the surviving objects or elements, but rather is manifested in the material organizations of things as assemblages. These assemblages can be characterized in terms of two salient characteristics reversibility and residuality. The
112 reversability of an assemblage is the ease with wh ich it can be reorganized or reconfigured. Residuality refers to the potential of an assemblage to leave material traces. These two factors are inversely correlated, so that an assemblage that is easily reconfigured (i.e., high reversibility) is unlikely t o leave material traces (i.e., low residuality). Lucas argues that the material organization of most events is not preserved in the archaeological record because they have high reversibility and low residuality. In contrast, those events that are accessibl e to archaeologists consist of changes in material organizations that are entrenched and carry great inertia they have high residuality and low reversibility. The reconfiguration of such an assemblage is thus an event that is apparent in the archaeological record. From this perspective the construction of large shell mounds and ridges by Mount Taylor people was an event inasmuch as it was the establishment of a material assemblage characterized by high irreversibility. The act of emplacing shell in particul ar locales along the river fundamentally altered the material circumstances of subsequent occurrences at that place (Barrett 1999 ). Once established these deposits could not be easily disassembled and they structured future material organizations. Similarl y we might argue that the onset of spring flow was eventful if it significantly altered the local and regional ecology (i.e., the assemblage of beings, places and objects). However, it is insufficient to assume that this phenomenon (an exogenous event [Sah lins 1991:43]) would be interpreted as significant, and hence eventful, by those who experienced it. In the remainder of this chapter I investigate whether the onset of spring flow was an event to the inhabitants of the middle St. Johns River valley . Below I present a GIS
113 model of the onset of artesian spring flow in the region to explore the temporal and spatial patterning of this process. I then turn to a discussion of recent archaeological evidence and hydrological reconstructions from two major springs in the region. Spring Chronology I: Modeling As detailed in Chapter 2, Florida is home to one of the largest concentrations of freshwater springs in the world. In the middle St. Johns River valley alone 60 springs have been documented. The majority of thes e springs discharge water from the Floridan Aquifer. Artesian flow in springs is pressure dependent (Scott et al. 2004; White 2002). Pressure within the aquifer fluctuates as a result of several factors that vary within and between individual spring basins , notably variations in precipitation, topography, and physical properties of the aquifer. Aquifer pressure is measured and displayed as a potentiometric surface, defined as the level to which groundwater will rise in a tightly cased well. There are essent ially two requirements for spring flow at any given point on the landscape. First, the potentiometric surface of the Floridan Aquifer must be higher than the ground elevation. In other words there must be sufficient pressure in the aquifer to force water u p and onto the surface. Second, there must be a pathway for the transmission of groundwater to the surface. As noted in Chapter 2, the Floridan Aquifer is in most places overlain by a layer of relatively impermeable materials. Where present , these material s confine the aquifer and prevent the flow of groundwater onto the surface. Thus, in order for a spring to flow this confining layer must be either absent or breached. If there is sufficient pressure , but the confining layer is intact, groundwater will flo w to areas of lower pressure , but will not discharge onto the ground surface. Where the confining layer is breached , but there is insufficient aquifer pressure, closed
114 surface depressions may form and the area will serve as a recharge zone, rather than a d ischarge zone. These criteria are obviously satisfied at the springs of the region today. However, it is unclear how long this has been the case. Given that fluctuations in sea level, precipitation, and evapo transpiration can all impact aquifer pressure, spring flow has likely fluctuated significantly in the past. There were probably few, if any, springs flowing during the Last Glacial Maximum (LGM), when both sea level and, concomitantly, aquifer pressure were considerably lower. Or rather, few springs wo uld have been flowing in the locations they do today. However, it is possible, if not likely, that numerous springs were present on portions of the Florida Platform that have since been inundated ( Faure et al. 2002; Scott et al. 2004:13). Regardless, as se a level rose over the course of the late Pleistocene and Holocene , aquifer pressure presumably rose with it, reaching a point at which springs would begin to flow. If the argument put forth by Miller is correct, and spring flow was a prerequisite to both t he establishment of extensive wetland biomes and the appearance of shell mounds in the region, then we should expect that the onset of spring flow was relatively rapid and synchronous across the basin. To explore the tempo of the initiation of artesian spr ing flow, I constructed a model of spring flow under conditions of decreased pressure in the Floridan Aquifer. This modeling effort carries several caveats. First, it assumes that the geometric configuration of the aquifer has not been drastically altered over the course of the Holocene. That is, there have not been significant structural geologic changes that altered the flow of ground water. Second, it assumes that changes in the potentiometric
115 surface of the aquifer would be realized uniformly over the s tudy area. Third, the model is not linked to specific reductions in sea level or precipitation as these do not correlate in a linear or simplistic manner with reductions in aquifer pressure. The individual effects of these variables (in addition to soil pe rmeability, thickness of overlying sediments, and rate of evapotranspiration) are complex and difficult to disentangle. But, reduction in either has essentially the same effect a net reduction of the pressure in the aquifer and thus lowered potential for s pring flow. I constructed the model using a digital elevation model with 15 meter resolution developed by the Florida Geologic Survey (Arthur et al. 2005). The locations of springs in the region were provided by the St Johns River Water Management District . Groundwater withdrawals for domestic, agricultural, and industrial uses ha ve altered the potentiometric surface of the Floridan Aquifer, so for this model the baseline potentiometric surface was derived from pre development estimations produced by the U. S. Geological Survey (Bush and Johnston 1988). A hydrologically correct raster was interpolated from the potentiometric isolines and laid over the digital elevation model. The elevation of the ground surface was then subtracted from the elevation of the aq by cell basis. The potential for spring flow exists where this differential is greater than zero (i.e., where the potentiometric surface is higher than the ground surface). This procedure was repeated for progressiv ely lowered potentiometric surfaces to explore the conditions under which contemporary springs could have begun their flow, thus indicating the pattern of the onset of spring flow in the region.
116 The model produced several interesting results ( Figure 3 6). First, there is a high degree of variability, as the differentials at contemporary springs range from 0.5 meters to over 12 meters. This indicates that even a small reduction in aquifer pressure could cause some springs to stop flowing, wh ereas others woul d be much more resilient. It follows too that there would be significant differences in the onset of flow at different springs. The model indicates that it is unlikely that any springs would have flowed if the potentiometric surface of the aquifer were in excess of 12 meters lower than present. This is equivalent to the water table reduction hypothesized by Watts (1971) for north Florida during the LGM. At 10 meters below present a single spring, Clifton Springs on the southern shore of Lake Jesup, could potentially flow. However, it would remain the only flowing spring in the region until the potentiometric surface increased to 6 meters below present. Several additional springs would begin flowing under these conditions, but the majority of springs would not begin flowing until the potentiometric surface increased to 2 meters below present. Interestingly there is some regularity to the pattern of spring initiation: the springs that would begin to flow fir st are in the southern portion of the middle St. Johns (e.g., Clifton, Wekiw a, and Rock springs), and are not the largest springs in the region. Springs in the northern portion of the study area wo flow until the p otentiometric surface was m uch closer to present conditions. Indeed, some of the largest springs in the area (e.g., Silver Gle n, Blue, and Alexander) would have been among the last to begin flowing 3 . 3 Many of these large springs originate from deep caverns that likely contained fresh groundwater, even if that water did not flow onto the surface.
117 I do not suggest that this model precisely predicts the sequence of spring initiati on in the region, but I do believe it provides some interesting points to consider regarding the eventfulness of spring flow. The GIS model suggests that the onset of spring flow may not have been a wide scale occurrence that suddenly inundated the valley with groundwater. It is therefore also unlikely that the regional hydrology and ecology was rapidly restructured as a result of spring flow. Given that the springs of the region vary with regard to elevation, conduit depth, and the localized expression of the potentiometric surface, initial artesian flow may have been heterogeneous, time transgressive, and punctuated. Further, it is difficult to know how the birth of a spring would be manifested. Many may have been pre existing depressions or groundwater fe d ponds that spilled over their banks as pressure increased. Or conduits may have been plugged with sediment that required significant pressure to flush , or covered by thin limestone ceilings that collapsed suddenly. There are other possible scenarios as w ell, and this was likely highly variable since the ways spring flow could begin are dependent on pre existing structural conditions at different locales. Regardless, the point is that although spring flow at any single spring may have begun rapidly, eventf ully, the regional scale pattern is decidedly more complex. Although the model indicates all of the springs of the middle St. Johns River valley did not begin flowing simultaneously, it is possible that groups of geographically proximate springs may have c ome on line in rapid succession. This was likely the case for the springs feeding into the western shore of Lake George. These six springs have differentials within two meters of one another. Thus, other things being equal, these
118 springs would have respond ed to increases in aquifer pressure roughly contemporaneously. However, the question remains what the timing of this was and whether it had significant ramifications for those who experienced it. Fortunately, recent archaeological investigations along seve ral springs in the Lake George watershed have generated data relevant to these questions. Spring Chronology II: Archaeology and Hydrology The complex of archaeological sites and water bodies surrounding Lake George provides an ideal laboratory for investig ating the timing of increased water availability in the middle St. Johns River valley and the human response to these dynamic conditions. Lake George itself is the second largest lake in Florida, covering an area of more than 190 km 2 (Stewart et. al 2006). The lake sits in a broad, shallow basin bordered on the west by high relict dunes. The eastern flank of Lake George is lower and seasonally flooded (DeMort 1991:101). A verage depth in the lake is only 2.5 m , outside of the dredged navigation channel . It i s thus heavily influenced by wind action, resulting in a well mixed water column with little thermal or chemical stratification. As part of the St. Johns River, Lake George is a flow through lake with an average turnover time of 84 days ( Stewart et. al 200 6 ) . Tidally i nduced reverse flows are periodically experienced because of the low gradient of the river. Seismic profiling of the northern portion of Lake George has indicate d the presence of both subsidence sinkholes and incised fluvial channels beneath s everal meters of lacustrine deposits (Kindinger et al. 1994 ) . The age of these features is unknown, but their presence suggests that the infilled lake basin was once characterized by both sinkholes and channelized water.
119 Abundant fresh groundwater flows into Lake George from six artesian springs on its Western margin (Figure 3 7). These springs lie on the eastern edge of the Ocala National Forest. Juniper and Fern Hammock springs collectively form the headwaters for Juniper Creek, a 15 km long spring run that winds through extensive wetlands. Sweetwater and Mormon Branch springs contribute their flow to the creek approximately 10 km downstream. Salt and Silver Glen springs are the largest springs feeding Lake George, indeed some of the largest in the entire St. Johns valley. Both Salt and Silver Glen springs are located in US Forest Service Recreation Areas and surrounded by significant archaeological deposits. Fieldwork conducted at Salt Springs in 2009 and Silver Glen Springs from 2007 2013 investigated the archaeological and hydrological histories encased in spring side sediments. Salt Springs The Salt Springs Recreation area is situated in eastern Marion County, in the town of Salt Springs . Lake George is located to the east, con nected to Salt Springs by a spring run that meanders for more than 8 km through lowland swamps and wetlands. Lake Kerr lies just to the west, separated from the spring by a na rrow (~ 300 m) isthmus of land. Salt Springs lies at the eastern flank of the Mari on Upland, a relatively narrow ridge characterized by Pleistocene sand dunes that extends from the Mount Dora Ridge to the western shores of Lake George. Approaching the spring from the west affords a panoramic view from the edge of the escarpment as the l and slopes quickly down to the spring basin. Salt Springs is aptly named as the concentration o f dissolved salts is several orders of magnitude greater than other artesian springs in Florida. The spring has been developed as a recreation area, complete wit h picnic pavilion, restroom facilities, and a
120 campground. Salt Springs itself consists of a broad, shallow pool approximately 40 x 60 m in maximum dimension, surrounded on three sides by a concrete retaining wall. The w ater , a relatively constant 23.3 Â° C ( 74 Â° F), issues from several ve rtical fissures in the pool bottom. The average depth of the water is approximately 0.5 m (Scott et al. 2004:237 239). Salt Springs is a second magnitude spring, the fourth largest in the m iddle St. Johns River valley , with a mean discharge of 80 ft 3 /s. Recent Archaeological Investigations In 2009 the USDA Forest Service conducted maintenance and repairs on the concrete retaining wall, extending a portion of it along the northern shoreline to replace an existing timber bulkhead . Prior survey work within the recreation area had documented dense archaeological materials in this vicinity (8MR2322), but the depth and integrity of these deposits was unknown (Figure 3 8). To facilitate the repair and replacement of the retaining wall, the Forest Service installed a coffer dam and pumping system. The coffer dam held the spring water away from the shore while the pumps removed seeping water, effectively drawing down the surface of the water some 2 m in a localized area. The net effect of draining the area behind the coffer dam was the exposure of a saturated midden component in the spring bed. The surficial expression of the midden is a lobe shaped area elevated approximately 1.5 meters over the surrounding channel bed ( Figure 3 9 ). T he elevated area measures roughly 30 x 20 m, with its long axis oriented parallel to the spring run. Historic photographs indicate that this is likely the subaqueous portion of a shell ridge situated on the northern bank of the spring. Twentieth century la nd alterations largely obscured this ridge, although intact subsurface deposits are present, and were investigated by National Park Service (NPS) archaeologists in 2009. The excavation of a construction trench with heavy machinery,
121 and installation of the new section of retaining wall, severed the subaqueous portion of this deposit from its terrestrial component . The Laboratory of Southeastern Archaeology (LSA), University of Florida , 2011). National Park Service testing of the terrestrial component determined that it was largely Mount Taylor aged with well preserved organic materials. Our e xcavation s trategy was designed to address the histories of anthropogenic and fluvial deposition and relate them to fluctuating water levels and shoreline transgression. To this end, a 1 x 8 m trench was laid out near the center of the subaqueous midden deposit, wher e it appeared to be the thickest. Testing of this orphaned midden remnant was facilitated by the Forest Service coffer dam and pumpi ng system, which kept t he test units relatively free of standing water until the drawn down surface of the local water table was intercepted. Excavation of the trench was halted when this surface was reached and stan ding water began to accumulate. The material recovered from Salt Springs is rife with evidence for daily habitation. Organic preservation was excellent in the satur ated deposits, with abundant botanical and vertebrate faunal remains. The vertebrate faunal assemblage is well preserved and diverse, with an emphasis on aquatic resources (Blessing 2011). The botanical assemblage is likewise diverse, and includes a number of economically important species (e.g., bottle gourd, squash, passionflower, grape, and blackberry [Talcott 2011]). The most ubiquitous taxon recovered was elderberry, which may have been used for its medicinal properties. Few pottery sherds were recover ed, and those that were came from the uppermost excavation levels. Marine shell, sharks teeth, lithic
122 debitage and tools, and modified bone were found throughout the trench, but were typically more frequent in the southern half (i.e., TUs 5 8). Four strati graphic units were identified in the trench excavations (Figure 3 10). These comprise two macro stratigraphic units. Stratum I consists of shell bearing deposits in the upper portion of the midden, wh ereas Stratum II encompasses the underlying shell free m idden sands. Stratum I is comprised of three sub str ata (i.e., IA 1, IA 2, IB), with subdivisions based primarily on changes in the abundance and species composition of the shell component, and on c hanges in the non shell matrix. The basal unit in the tren ch consists of organically enriched sands largely devoid of shell ( Stratum II ) . This unit contains contorted layers of stacked and interdigitated sand lenses, varying in color from grey to black. These sands are often stained with colloidal organic matter and contained moderate amounts of vertebrate faunal remains, lithic flakes, and both charred and uncharred botanicals (e.g., wood, hickory nut, seeds, and charcoal). The top of Stratum II is undulating and dips away from the shore . Lying unconformably over the midden sands of Stratum II is Stratum I. This is a relatively thin ( 5 15 cm) layer of grey sand and shell that was observed in TUs 1 6 and in the percussion cores. Overlying Stratum IB is Stratum IA, a shell midden deposit more than 50 cm thick that c ontained very dark brown to grey organically stained sands with abundant Viviparus shell and localized lenses of bivalve shell. Pomacea shell was relatively rare, but several concentrations were encountered. Stratum IA was divided in to Stratum IA 1 and IA 2 on the basis of slight color and textural variations in the matrix. Stratum IA 2 is slightly darker and finer than Stratum IA 1. The contact between
123 the two is diffuse in places, but dips noticeably away from the shore, suggesting that Stratum IA 2 was d eposited partially overtop of an d to the south of Stratum IA 1. C hronology and Depositional History W ell preserved organic materials in near shore anthropogenic deposits generally indicate permanent saturation ( Bleicher and Schubert 2015 ). That is, the abu ndance of these organic materials, particularly in the lower levels of the trench, points to deposition directly in the water . However, the presence of concreted shell, coupled with a lower frequency of organic remains , in the uppermost levels of TUs 1 6 indicate s that some portions of the midden were subject to periodic aerial exposure and drying as water levels in the spring fluctuated . The morphology of Stratum II lends further credence to this interpretation, as the constituent sand lenses have a contorted or rippled appearance, consistent with their mobilization and deposition in an active open water environ ment. Six AMS radiocarbon assays were obtained to investigate the chronology and depositional hi story of the mi dden. Two were selected from each of the major stratigraphic units defined in the trench (i.e., Strata IA, IB, and II). To investigate the sequence of deposition within each strat um one sample was taken from the shoreward (northern) side of the strat um and one from the springward side. Based on 2 sigma calibrated ranges, the anthropogenic deposits investigated in the trench were emplaced over a period of some 400 to 900 years in the interval 6640 5750 cal B.P. The radiocarbon sequence obtained from the tren ch bears out the inferred order of deposition. Alt hough there is slight overlap at the 2 sigma range, none of the dates are out of sequence. Samples from the lowest deposits exposed in the trench, Stratum II, returned the oldest dates: 5710 Â± 50 B.P. (6640 64 00 cal B.P. ) and 5610 Â± 50 B.P.
124 (648 0 6300 cal B.P. ). Slightly younger dates were obtained from Stratum IB, the lowermost shell bearing deposit, at 5460 Â± 50 B. P . (6400 6130 cal B . P . ) and 5230 Â± 50 B . P . (6180 5910 cal B.P. ). Stratum IA, the uppermost un it, was dated to 5150 Â± 50 B.P. (6000 5750 cal B.P. ) and 5130 Â± 50 B.P. (5990 5750 cal B.P. ). The terminus of deposition is not known , as dates were not obtained from the upper portions of Stratum IA. However, several radiocarbon assays from the terrestria l portion of the site excavated by the NPS have a 2 sigma range of 5450 4850 cal B.P., pointing to at least three additional centuries of occupation (Michael Russo, personal communication 2010). Although comparison of the radiocarbon assays between stratig raphic units confirms the observed vertical sequence of deposition, comparison of the assays within each strat um can inform about the horizontal expansion of the deposit. In each case the date obtained closer to the shore is older than the date(s) obtained closer to the spring. Taken together, these data suggest that these anthropogenic deposits prograded outward, away from the shore. This progradation was followed by the establishment of a new depositional regime closer to the shore, over top of the previo us deposits. This pattern is most strongly expressed in Strata II and IB, where overlap between the dates is less than a century. The two dates from Stratum IA are virtually contemporaneous. However, as noted above, the position of Stratum IA 2 relative to IA 1 is indicative of progradational deposition. Thus, the tightly clustered dates of Stratum IA may indicate an increase in the tempo of deposition rather than a change in its mode. To recap, the depositional history revealed in the trench began with the aggradation of a shell free midden deposit prior to the emplacement of shell sometime
125 after 6300 cal B.P. With regard to the eventfulness of spring flow and the changes it precipitated, this is relevant for two reasons. First, it shows that the initial oc cupation of the spring did not involve any detectable deposition of shell. Spring flow is inferred at this time from the character of the deposits, and thus the necessary conditions for shellfish exploitation are presumed to have existed, but this exploita tion is not reflected in local practices. Second, this shell free midden post dates shell sites elsewhere in the St. Johns River valley . Recall from above that accumulations of massive shell are evident by 7400 cal B.P. at places like Hontoon Dead Creek Mo und and Live Oak Mound. The available evidence from Salt Springs indicates that shell deposition did not begin until a millenni um Thus, people were visiting the spring without de positing shel l, despite a 1,000 year history of doing so in the middle St. Johns River valley . Clearly cultural practices, if not hydro ecological conditions, were not spatially homogenous in the region. Hydrological History In addition to the histor y of anthropogenic deposition, I also explored the hydrological history of Salt Springs through the extraction of three percussion cores (Figure 3 9). These were emplaced directly in the spring bed, penetrating the anthropogenic sediments and documenting alluvial sediments b eneath . Two of the cores were successfully recovered (Core 1 & 2) but the third was compromised during extraction (Core 3). The sequence of deposits observed in the cores offered insight into deposits that could not be reached during trench excavations (Fi gure 3 11) . These sub midden units were delineated as Strata III VI. They appear to represent water lain deposits that are largely lacking anthropogenic sediment inputs. Variations in the fluvial regime at Salt Springs are registered by changes in these
126 deposits, which are broadly composed of layers of clean, light grey sand alternating with more heterogeneous layers. Organic materials (i.e., faunal and botanical remains) are well preserved throughout these deposits, al though they are markedly more abunda nt in Strata III and V. Stratum V also exhibit s fine laminations and a slightly silty texture consistent with quiet, slack water deposition. Meanwhile, Strata IV and VI, which are largely clean deposits of light grey sand, may reflect periods of increased flow from the spring and the consequent flushing of organic matter and fine sediment. Alternatively they may represent periods of desiccation under lowered water levels. T wo radiocarbon assays were obtained on samples recovere d from Core 1, which was adjac ent to the distal end of the excavation trench . A large wood fragment at the juncture betw een Strata V and VI at 110 cmbs was uncharred, organically stained and well pr eserved, indicating that it was deposited under water. This sample returned a date of 83 20 Â± 40 B.P. (9460 9140 cal B.P. ). Charcoal from t he basal shell deposit (Stratum IB ; 35 cmbs ) in Core 1 returned a date of 5300 Â± 40 B.P. (6190 59 5 0 cal B.P. ). Thus, 72 cm of fluvial sediment were deposited over the course of some 3000 to 3500 years, yiel ding an average sedimentation rate of 2.0 2.4 cm/century. Although this evidences relatively slow, gradual accumulation of sediment, it is not outside the range of variation recorded in other fluvial settings in North America (Ferring 1986), and can be exp lained by the minimal sediment load being carried by the spring as it emerges from the aquifer. Indeed, much of the sediment load is likely colluvium from the surrounding uplands. These results indicate that water was available at Salt Springs over 9,000 y ears ago . This is more than three m illennia earlier than hypothesized by Miller (1992, 1998)
127 and well before sea level stabilized in a near modern regime. Although there is little evidence of human activity at th at time, wood associated with a hafted end s craper from the terrestrial portion of the site was radiocarbon dated to ca. 8500 cal B.P. (Michael Russo, personal communication 2010). Coupled with the inferences from trench excavation, this suggests a significant lag between the onset of spring flow an d shell deposition, and indicate s neither a rapid nor a dramatic mapping on to the purported ecological ramifications of spring flow. The evidence from Salt Springs is compelling, but it is only a single data point. For corroborating evidence we can turn t o Silver Glen Springs. Silver Glen Springs The Silver Glen Springs Recreation Area is i n eastern Marion County , approximately 14.5 km southeast of Salt Springs . Like Salt Springs, Silver Glen Springs lies adjacent to the eastern edge of the Marion Upland . Lake George is located to the east, connected to Silver Glen Springs by a short (1.2 km) spring run that that averages 60 m in width with a maximum depth of approximately 3 m ( Pandion Systems, Inc. 2003). Silver Glen is a first magnitude spring, one of the largest in Florida, with an average discharge of 102 ft 3 /s. The spring consists of a large pool, measuring 200 x 175 m in breadth, with water discharging from two main vents and a number of smaller vents both adjacent to and within the spring run. The mai n vent of Silver Glen springs sits at the base of a conical depression beneath ~5.5 m of water near the center of the spring pool. A second vent, referred to as the natural well, lies at the southwestern edge of the pool. This is a 12 m d eep vertical shaft , or chimney, measuring 3 5 4.5 m in diameter. An extensive system of
128 caverns and conduits has been mapped at Silver Glen Springs, extending over 600 m from the main vent (Springs Eternal Project 2013). Archaeological Overview Numerous archaeological sites surround the pool and run of Silver Glen Springs (Figure 3 12). These are referred to collectively as the Silver Glen Complex, and have been investigate d since 2007 by the LSA and the University of Florida S t. Johns Archaeological Field School , both under the direction of Ken Sassaman ( Sassaman et al. 2011 ). This program of fieldwork has uncovered a dense archaeological record and generated numerous technical reports, dissertations, and other publications (e.g., Gilmore 2014; Randall 2010, 2011; Randall et al. 2011; Randall et al. 2014; Sassaman et al. 2011; Sassaman and Randall 2012). The Silver Glen Complex includes as many as four Mount Taylor shell ridges, Thornhill mortuary mounds, two massive U shaped shell mounds constructed during the Orange period, Orange and St. Johns era habitation sites, and a St. Johns burial mound. Few of these components are spatially isolated, but rather occur in overlapping deposits at various locations in the watershed. Evidence for late Pleistocene and Early Holocene occu pation in the vicinity is also present, if ephemeral. The Silver Glen Complex will be discussed in detail in the following chapter. For present purposes I focus on the evidence for the onset of both spring flow and shell deposition. The best documented Mou nt Taylor shell ridge at Silver Glen Springs is 8LA1W Locus A. Locus A is located approximately 500 m downstream from the pool of Silver Glen Springs, on the southern shore of the spring run. Immediately across the run from Locus A, on the northern shore, lies another shell ridge dubbed site 8LA4242. Little is known about site 8LA4242 as it has only recently been documented. Based on its size
129 and configuration (185 m long by 85 m wide, in a tear drop shape parallel to the run) it is likely a Mount Taylor sh ell ridge (Randall et al. 2011). Additional Mount Taylor shell ridges are suspected at other locations, notably beneath later U shaped shell mounds surrounding the spring pool (8MR123; Randall et al. 2011) and at the confluence of Silver Glen Run and Lake George (8LA1E; Sassaman et al. 2011). I focus on Locus A as it is the most intensively investigated. Shell Deposition at Locus A Site 8LA1 West Locus A is the remnant of a Mount Taylor shell ridge measuring some 200 m long by 75 m wide (Randall 2010:330 3 3 5, Sassaman et al. 2011:121 170 ). Much of the ridge core was destroyed by mining operations in 1923, leaving behind a halo of discontinuous elevated deposits that mark the outline of the ridge. The pre mining height of the ridge is unknown, but 3 m of inta ct deposits remain. Locus A parallels Silver Glen Run, but is separated from it by a hydric hammock to the north. A linear embayment, possibly a relict seep spring, marks the eastern extent of Locus A. Stratigraphic testing in 2007 and 2008 was designed to expose intact vertical profiles along mining escarpments (Sassaman et al. 2011). Six 2 x 2 m test units spread across three areas of Locus A exposed 12 m of profiles and 24 m 2 of plan excavation. Later block excavations in 2012 and 2015 exposed an additio nal 29 m 2 in plan and over 50 m of profiles. The stratigraphic sequence exposed by these excavations is complex, with individual test units exposing as many as 28 discrete stratigraphic units. However, broad trends in deposition are apparent, allowing thes e units to be grouped into macro stratigraphic units (Randall 2013; Sassaman et al. 2011; Sassaman and Randall 2012). The initial deposition of shell by Mount Taylor inhabitants of Locus A did not involve the
130 mounding of shell , but rather its emplacement i n a number of large pits (Randall 2014 a ; Randall and Sassaman 2012). These pits were typically large and straight walled, measuring 1 to 1.5 m wide and up to 1 m deep. Heavily oxidized sand at their bases suggests they were initially used for roasting shel lfish, before being infilled. Pit fill typically contained multiple species of shellfish, a diverse vertebrate faunal assemblage, and tools of bone and marine shell. Sometime between 6300 and 5940 cal B.P. a cap of fine brown sand up to 40 cm thick was emp laced over the entirety of the ridge (Randall 2013) . Atop this is an upper deposit of shell, composed of accretional layers of earth and shell alternating with surfaces of crushed and burned shell. Like the underlying pit fill, these shell deposits contain ample amounts of material indicative of daily living food remains, ash and charcoal, and bone, stone, and marine shell tools at multiple stages of manufacture and use. In form and content, these deposits are not dissimilar from shell nodes interpreted as house mounds at the Hontoon Dead Creek Village, discussed above. It is likewise similar to Salt Springs, in both content and timing, albeit in a terrestrial rather than subaqueous setting. Recent excavations have also recovered evidence for pre Mount Taylo r activities at Locus A (Randall and Sassaman 2012; Randall 2014 a ). The shell processing pits that mark the onset of intensive shell deposition intrude into a lower deposit of organically enriched soil with small amounts of vertebrate fauna and shell. This deposit is not accretional or pedogenic, but rather is an amalgamation of large pits. These pits measure, on average, 1 m in diameter and depth, and have no evidence of burning. The contents of their fill replicates that of the upper pits, but in consider ably lower density.
131 The function of these lower pits is, as yet, unclear. However, they attest to a history of low level shell deposition that preceded intensive shellfish processing and the building up of the ridge. Radiocarbon determinations from a numbe r of the lower pits place them in the ninth millennia before present, with three assays ranging from 8170 to 8980 cal B.P. Additional testing indicates that these pits are restricted to the footprint of Locus A. In other words, the shell ridge that was ere cted some 2,000 years later is isomorphic with the distribution of these pits. Hydrological History Excavations at Locus A are pointing to an increasingly long history of shellfish exploitation in the St. Johns River valley . But how this relates to the onset of flow from Silver Glen Springs is not clear from the archaeological evidence alone. Given the proximity of Locus A to Lake George, it is entirely conceivable that aquatic resources, such as shellfish, could be obtained without water input from the spring. In order to reconstruct the hydrological history of Silver Glen Springs, six vibracores were extracted along two transects aligned roughly perpendicular to the axis of the spring channel (Figures 3 13 and 3 14). These vibracores were emplaced in th e hydric hammock intervening between Locus A and Silver Glen Run. The hammock is tear drop shaped, widest at the northeast and tapering to the southwest, and mea sures approximately 300 m long by 60 m in maximum width. Transect BC was located near the easte rn margin of the hammock, where the channel of the run widens proximate to the relict seep spring. Two cores along this transect were likely compromised by rodding wherein the core r penetrates but does not capture unconsolidated or flocculated materials as they exhibit short and discontinuous sedimentary records. Transect EF was placed near the middle of the hammock, proximate to the western
132 margin of Locus A. Cores form this transect have greater integrity and are thus the focus of discussion below. Observ ations at similar locations adjacent to wetlands (e.g., Hontoon Dead Creek Mound) and springs (e.g., Salt Springs) indicated that shell deposits are frequently found beneath wetland sediments. As water levels rose in the mid to late Holocene and overtoppe d the aprons of shell ridges, hydroperiods also stabilized. This allowed the accretion of organic sediments atop anthropogenic deposits. So it was expected that our cores would reveal intact portions of Locus A buried beneath wetland deposits. However, thi s was not the case as strata observed in the cores consist of deposits that are lacking anthropogenic sediment inputs. As many as 12 stratigraphic units were observed in cores, differentiated on the basis of color, clastic particle size, and abundance and degree of decomposition of organic material. These are grouped into two macro stratigraphic units: an upper layer of organic soils and sediments (i.e., muck and peat) overlying basal siliciclastic deposits (i.e., sand). Intervening between these was a thin (5 10 cm) zone of mucky sand or intercalated muck and sand. Organic soils and sediments ranged from dark reddish brown peats with many visible fragments of vegetal matter to black muck with few visible fibers. Basal sands were typically grey to greyish br own beneath the contact with organic deposits. These lightened in color and grew finer in texture to white fine sand at the base of most cores. One core, EF 02, penetrated a pale brown loamy, sand with striations reminiscent of clay lamellae at its base. T he shift from mineral to organic sedimentation reflects a change in the hydrology of Silver Glen Springs. The deposition of peat and muck indicates a relatively
133 quiet, slackwater environment , likely on the margins of the channel . The underlying sands may r epresent relict channel deposits, but are more likely derived from colluvium from the surrounding uplands, deposited in a terrestrial setting. Sandy deposits are largely devoid of preserved organic materials, supporting this inference. The change to organi c sediments is indicative of rising water, drowning terrestrial deposits and favoring the preservation of organic detritus. Presumably this registers the onset of flow from Silver Glen Springs, but it may represent channel migration or an increase in the s tage of an already flowing spring. Regardless, the timing of this transition provides a terminus ante quem for the onset of spring flow. A cross section of the spring run and surrounding terrain along core transect EF is presented in Figure 3 14. Organic d eposits were absent at higher elevations adjacent to the remnant shell ridge of Locus A and gradually increased in thickness, extending as far as 2.8 m below surface at the distal end of core transects, adjacent to the Silver Glen Run. Likewise, sa ndy depo sits underlying organic sediments were at a higher elevation proximate to Locus A and dipped towards the spring. Two radiocarbon assays were obtained from Cores EF 02 and EF 03. These assays were run on basal samples of organic sediment to date the transit ion from mineral to organic sedimentation and rising waters. In core EF 03, at the distal end of the transect, a sample taken from the ba se of the organic sediments ( 278 280 cmbs) returned an assay of 7750 +/ 30 B.P. (8590 8450 cal B.P.). Core EF 02 was l ocated approximately 25 m inland of Core EF 03. Here, the base of organic sediments was at ~ 170 cmbs. A sample taken 2 cm above this juncture returned an assay of 5170 +/ 30 B.P. (5950 5905 cal B.P.). Taken together, these results indicate that roughly 12 5 cm of
134 wetland sediment accumulated over 2500 2685 years, or 4.65 5.0 cm per century. This is a considerably higher rate of accumulation than at Salt Springs, but they are in different sedimentary environments. Recall that the cores from Salt Springs were extracted directly from the bed of the spring pool. Mineral sediments were deposited there in an actively flowing channel of variable energy. Deposition at Silver Glen was characterized by the accumulatio n of organic sediments in a low energy wetland envi ronment. The results of coring indicate that the hydric hammock fronting Locus A formed as organic detritus accumulated in a basin with basal sands. This basin contains the channel of Silver Glen Run and wetlands along its margin. Underlying sands likely a ccumulated as a result of colluvial processes prior to inundation, and water levels began to rise no later than 8450 cal B.P. This is in ag reement with the ninth millennium before present pit digging described above and corroborates the dating of spring flow at Salt Springs, albeit with a gap of some 700 years. Discussion To summarize, the hypothesis that intensive shell deposition in the St. Johns River valley was predicate d on spring flow has several implications. As put forth by Miller (1992, 1998), the onset of spring flow is expected to have followed the stabilization of sea level after ca. 6000 cal B.P. and to have been widespread and rapid in the valley. Shell depositi on should follow the onset of spring flow with minimal delay, as inhabitants mapped onto nascent aquatic resources and populations grew. Minimally, the evidence from cores at Salt and Silver Glen springs pushes the chronology of spring flow initiation back several millennia. Salt Springs was apparently flowing by 9140 cal B.P. and Silver Glen no later than 8450 cal B.P. I cannot say with
135 certainty that these dates mark the onset of continuous spring flow, which may have been ongoing for centuries prior to t his. Indeed it is likely that both Salt and Silver Glen springs began as sinkholes or depressions that were water bearing but did not fl ow onto the surface. But by that time water was present in these locations, sufficient that wetland deposits began aggra ding. Likewise, the origins of shell deposition have been relocated in time. Prior work by Sassaman, Randall, and others has shown that intensive shell deposition in the region began as early as 7400 cal B.P. at places like Hontoon Dead Creak and Live Oak mounds. The evidence from Locus A at the Silver Glen Complex points to the need to discriminate between different modes and intensities of shell deposition. The earliest shell deposition do cumented thus far indicates low intensity shellfish exploitation an d deposition underground, in pits. These practices may have been in place as early as 9,000 years ago and pre date intensive shell deposition by over a millennium. These results perhaps indicate a long interval of gradual change, wherein both spring flow a nd shell deposition began as intermittent, low level affairs that slowly increased in intensity over the course of several thousand years. But if all that has been accomplished here is a pushing back or stretching of the chronology of these changes , it wou ld add little to our understanding of either springs or the archaeology of the region. Closer examination reveals not gradual change, but rather a series of disjunctures that appear gradual only when one draws straight lines between them. The available evi dence suggests that at least some springs were present on the landscape of the middle St. Johns River valley long before they became a locus for the intensive deposition of shell. This is not to say that springs have been static; the
136 configuration of sprin gs, intensity of their flow, and quality of their waters have undoubtedly fluctuated. Further, while contemporary experience can inform us on the death of a spri ng, it is difficult to know how the b irth of any given spring would unfold and how it would be received by local residents. Indeed, each spring has a unique ontogeny and history, both hydrologically and culturally. Modeling this history indicates that although the initiation of spring flow appears significant and eventful at first glance, it was not a synchronous event throug hout the St. Johns River valley . Even proximate springs, such as Salt and Silver Glen, have divergent and discontemporaneous histories. This underscores that we cannot uncritically apply the results any one spring to others in th e region. With this caveat in mind, flow from both Salt and Silver Glen springs was initiated prior to 8500 cal B.P., coinciding (or preceding) locally with the excavation of large pits adjacent to Silver Glen Springs and regionally with the pond burial tr adition best known from W indover. Following this, there wa s a gap at these locales. We know little of what transpired at either spring from ca. 8200 cal B.P. to 6500 cal B.P. Likewise, the archaeological record is sparse across the region until 7400 cal B. P., when intensive shell deposition began at Hontoon Dead Creek and Live Oak mound (i.e., Episode I). Thus, the earliest evidence of shell deposition (next to springs) is followed by a gap of at least 800 years, after which intensive piling up of shell beg an (away from springs). Shell mounding adjacent to springs lags behind a further 1,000 years, beginning coincident with the onset of Episode II. Add to this pict ure the presence of basal shell free deposits at Salt Springs, which indicate that Mount Taylor era habitation adjacent to springs did not always
137 involve the deposition of shell. Th is is evident elsewhere in the v alley as well, outside the Lake George watershed. At Blue Springs (8VO43) , Thornhill Lake phase deposits were encountered, dating to ca. 5 300 4600 cal B.P. (Sassaman 2003 a ). These consist of a basal , shell free midden beneath a modest shell bearing deposit. Abundant faunal remains, charcoal, and occasional lithic and marine shell tools attest to intensive daily habitation in this locale, bot h before and after the inception of shellfishing. Future fieldwork will no doubt begin to fill in these gaps, but the work reviewed above calls into question the eventfulness of the onset of spring flow . Spring flow does not appear to have precipitated a r apid reorganization of subsistence and depositional regimes (i.e., shell mounding) that we might recognize as an archaeological event ( sensu Lucas 200 8 ). But I do not mean to suggest that springs were uneventful. Simply that what appears to be significant and eventful to us the onset of spring flow and the ecological potential this brought about does not appear to have had observable consequences with respect to shell mounding and the Mount Taylor tradition. It was thus rather n on eventful. We can instead r ecognize a series of events for example at the digging of shell free pits and shell processing pits, emplacement of a sand cap, and accretion of the shell ridge at Locus A that had material ramifications and structured future activities. The transformation of these places evidenced by the building up of shell was predicated not on spring flow, but on earlier activities that occurred. Alternatively, we might imagine that the birth of a spring could have figured as a prominent event in myths and narratives r ecounted in the past. Indeed it may even have been foundational. But in terms of the perspective outlined above, it did not lead to significant reorganization of material assemblages in any way that is apparent. T he
138 events of significance at springs, at le ast in the middle St. Johns River valley , were not instigated by their changing hydrology, but were constructed through the material engagements that people initiated on their banks. It is no coincidence that spring side shell deposition coincides with the onset of Episode II. This was a time of expanding social geographies, indicated both by the inclusion of items from distant places stone from the panhandle and Gulf Coast of Florida, marine shell from southern Florida and by the interment of people foreign to the St. Johns River valley . Springs became focal points for regional gatherings amidst this widening circumference of interaction. The accumulation of shell at springs marks an a cceleration of these gatherings that would continue in the Thornhill Lake period, reaching an apogee during the Late Archaic Orange period. In the next chapter, I explore the growth and transformation of springs through a more detailed examination of Silve r Glen Springs, gathering place par excellence in northeast Florida.
139 Figure 3 1. Kissengen Spring , Polk County, Florida in 1894 . Photo courtesy of the State Archives of Florida, Florida Memory , https://floridamemory.com/items/show/117843
140 Figure 3 2 . Kissengen Spring ca. 1947 , three years before it went dry. Photo courtesy of the Florida Geological Survey Photo Archive
141 Figure 3 3 . Kissengen Spring, April 2006. Photo courtesy of the U.S. Geological Survey
142 Figure 3 4 . Regional culture historical timescale (after Anderson and Sassaman [2012:Table 1 1]) and local traditions in the St. Johns River valley
143 Figure 3 5 . Distribution of Mount Taylor era sites in the St. Johns River valley , highlighting locations mentioned in the text. Also shown are two pond burial sites, Windover and Gauthier
144 Figure 3 6 . Results of the GIS model, showing areas of potential spring flow under conditions of lower than present potentiometric surface of the Florida Aquifer
145 Figure 3 7. Loc ation of Lake George and surrounding springs
146 Figure 3 8. Location of site 8MR2322 on a subsection of the U.S. Geological Survey excavations relative to the pool of Salt Springs (below )
147 Figure 3 9 . Topographic map of subaqueous portion of the Salt Springs site (8MR2322), showing locations of University of Florida excavation trench and cores and National Park Service excavation block. Elevation relative to arbitrary datum
148 Figure 3 10. Grid west profile of trench at 8MR2322. A) Composite photograph . B) Profile drawing
149 Figure 3 11 . Deposits observed in percussion cores at Salt Springs, Core 1 (left) and Core 2 (right), in relation to trench stratigraphy. Elevation relative to arbitrary datum
150 Figure 3 12 . Map of the Silver Glen C omplex, highlighting loc ations with Mount Taylor era deposits. Site footprints courtesy of Asa Randall (2014b)
151 Figure 3 13 . Deposits observed in vibracores at Silver Glen Springs. Elevation relative to NAVD 1988
152 Figure 3 1 4 . Aerial photograph of Silver Glen Springs and Run (above), showing location of vibracores. Cross section of Silver Glen Run (below) , showing vibracore transect E F in relation to surrounding landforms. Elevation relative to NAVD 1988. Vertical exaggeration x20
153 CHAPTER 4 TRANSFORMATIVE HISTORY AT SILVER GLEN SPRINGS Mount Shasta stands alone, ascending sharply to a peak 3,000 m above acramento Valley. The treeless, glaciated upper slopes of the volcano are visible for hundreds of kilometers on a clear day (Figure 4 1; Huntsinger and FernÃ¡ndez GimÃ©nez 2000; McCarthy 2004). Just beneath the tree line, at an elevation of 2,400 m, lies Pan ther Meadow and, within it, Panther Spring. Springs are sacred to the Winnemem Wintu tribe, a relatively small Native American group indigenous to the McCloud River region of northern California (Theodoratus and LaPena 1994; Winnemem Wintu 2015). Springs a re important for healing, cleansing, and purification rituals and serve as potent conduits to the spiritual realm (McCarthy 2004:175; Theodoratus and LaPena 1994:24). Although all springs are important, Panther Spring is pre eminent, the genesis place wher e the Winnemem Wintu (2015) The Wintu are not the only people who consider Panther Spring and Mount Shasta sacred. It has become a popular destination for New Age religious seekers, spiritual nomads, and vagabonds, all of whom have easy access to Panther Meadow from a parking lot at the end of Everett Memorial Highway (Huntsinger and FernÃ¡ndez GimÃ©nez 2000; McCarthy 2004). While professing to revere the spring, many of the acts of these pilgrims are di sruptive, if not harmful: use of the meadows and springs. Crystals were placed in and near the water, particularly Panther Spring, sacred to the Wintu tribe. Prayer flags were tied to tree branches, and pictures and poems were left on small rock altars in the meadow and near springs. The crisscrossing trails led to altars and to the denuded edges of Panther Spring, where many people
154 came to collect water. Nude sunbathing, drumming, and chanting were frequent activities nearby (Huntsinger and FernÃ¡ndez GimÃ©nez 2000:536). The physical damage to the meadow, co opting of Native American symbology, and desecration of Panther Spring are all understandably troubling to the Winnemem Wintu. But Mount Shasta and Panther Spring lie within the Shasta Trinity National Forest, under the management of the U.S. Forest Service. The desire of the Wintu to safeguard these sacred places against the threat of New Age tourism underscores the difficulty of man aging competing interests, secular or otherwise, on public lands. Despite this conflict, the perceived sacredness of this place by non indigenous groups was crucial in mobilizing resistance to, and ultimately rebuffing, a proposal to develop a ski resort o n the mountain in the 1990s (Huntsinger and FernÃ¡ndez GimÃ©nez 2000). The previous chapter decoupled spring flow from the regional construction of shell mounds as an adaptational process and argued that the use of springs was predicated not (or not solely) on their ecological productivity. It was suggested, rather, that springs became foci for extraregional gatherings in the context of expanding spheres of interaction and exchange. In this chapter I further develop this line of thought through an examination of Silver Glen Springs, which features an expansive and long lived archaeological record. The long term record of visitation at Silver Glen Springs will be used to explore the historical trends and transformations of pr actices taking place there. But they will also be used to address the question of sacredness, both at springs generally and Silver Glen specifically. There are several parallels between the situation of Panther Spring at Mount Shasta and that facing springs in Florida. Both are under threat from development and over use of public lands and subject to competing interests from diverse parties. However, whereas development has been successfully
155 rebuffed at Panther Spring, in Florida springs conservation has met with mixed results (Knight 2015). What Panther Spring illustrates aptly is that the sacredness of a place can congregate the interests of competing parties and buttress conservation efforts. In posing questions of sacredness, I do not mean to recapitulate a simplistic dichotomy between sac red and profane that in itself is problematic, a product of Cartesian dualism in Western thought that may have little purchase in non Western societies (Bradley 2005; BrÃ¼ck 1999). Indeed, in many cultures sacredness can extend into all aspects of life and encompass entire geographies, particularly as they memorialize ancestral or mythic persons and events (Basso 1996 a ; Hubert 1994; Morphy 1995). Certain places are nevertheless marked out as particularly important, special, or powerful. However, the power or sanctity of these places is not given solely from their physical features. Places are not static, objective arenas for human activity or canvases for the interpretive endeavors of human perception and consciousness. They ar e experienced from a culturally mediated , human perspective. But t his experience of place is not unbound or subjective . It must confront the affordances and constraints of the material world. Places are not given, but emerge through mutually constituting relationships with people. Places a ). Sacred places thus become sacred through an emergent, historical process. The question is thus less was this place sacred? than how, or in what ways, did it become sanctified? Sacred Site, Sacred Spri ng The spring rose up from its deep source and smelled of wet earth and the stones at the center of the world. Whatever you believe and whatever God you pray to, a place where clean water rises from the earth is in some way sacred (Frazier 2006:56).
156 Is there an inherent sacredness to springs? Does something of their experiential or aesthetic quality elicit awe or reverence in humans? Strang (2005, 2008) argues that the observable qualities of water, coupled with the cognitive and sensory apparatus common to all humans, overrides cultural and historical particularities to generate recurring themes in water symbolism and meaning. Notable among these are regenerative force; as the substance of social and spiritual identity; and as a symbol of sustaining activities. As such, it is frequently implicated in religious ceremonies and ritual (e.g., baptismal rites initiating one into the congregation , or the anointing with holy water ). Water is mutable, always changing, and so is invoked in metaphors for change and transformation. Likewise, the motion of water is drawn upon in tales of mythical journey and met aphors for the passage of time, while water confluences imply a coming together or gathering. Water can be both creative and destructive, as in a flood that can imaginativ In a similar vein, TaÃ§on (1999; see also Nash 1997) suggest s that certain landscapes are archetypal and considered sacred or special regardless of cultural background. Likewise, it has been argued that: there are broad similarities between peoples from various parts of the such as mountain peaks, springs, r ivers, woods and caves (Carmichael et al. 1994:1). Often these places are transitional, or locales where the upper and lower worlds collide
157 1996 a ; Santos Granero 1998) and Bradley (2000:33 44) argues that greater attention be paid to these places as they are part of the larger landscape of monuments and : where the results of great acts of natura l transformation can be best seen at junctions or points of change between geology, hydrology, and [and] places providing panoramic views or large vistas (TaÃ§on 199 9:37). Springs would seem to satisfy many of these criteria . They are transformative places where the upper and lower worlds intersect as water flows up and out of the ground. Springs are typically located at the base of ridges and uplands that offer panor amic vistas, and lie at ecotones juxtaposing differing geologies and ecologies. They also bring together different kinds of water at the confluence of spring runs with lakes or rivers. Florida springs, in particular, are striking in their stark contrast wi th other water bodies on the landscape. Indeed, in many ways springs are an inversion of these sluggish, blackwater rivers or still ponds, lakes, and wetlands. Variability in spring flow is stable relative to the rivers they feed into, like the St. Johns o r Suwannee (see Chapter 2). Their temperature is constant, and so, relative to atmospheric temperature, spring water is cool in the summer and warm in the winter. The water flowing up from the ground, oft en rapidly, is crystal clear, hypoxic, and harbors s p ecies of animals and plants different from those in other water bodies. But identifying a place as striking or affective is insufficient. A sacred place is more than a piece of geography. It implies a set of beliefs and rules, regulations, and cultural no rms regarding proper behavior at that place and comportments towards the power manifested there (Carmichael et al. 1994; Hubert 1994). Dwyer (1996:163)
158 plants, animals and people and the physical media within which or upon be manif ested in the visible world to varying degrees, particularly at potent places. Often ritualized practices (i.e., formalized and repeated sets or bundles of actions, following Bell ) are enacted with the explicit intent of making connections between th e visible and invisible worlds. Often, sacred sites are avoided or only visited by certain individuals or sects of society. The activities taking place there conform to norms and standards of proper behavior that may or may not involve material traces. In many cases , observing ritualized practices and inferring the rules of behavior may be the only way to identify sacred sites (Hubert 1994). However, often the activities taking place at sacred places d o leave material traces, traces that can be interpreted by archaeologists . Notable among these is the interment of the dead, construction of monuments, and placement of votive offerings. Springs are considered sacred in many parts of the world. Certainly the Panther Spring example is relevant here, but there a re others as well. In England and in Ireland, countless springs are revered as holy wells. Many were pilgrimage sites in the past and continue to be important for healing and fertility rituals (Brenneman and Brenneman 1995; Gribben 1992; Logan 1980; Rattue 1995). In central Kansas is an artesian spring flowing from atop a travertine hill, now inundated by a reservoir. Waconda, or the Great humans to communicate with the animal
159 springs, cenotes , and caves to be portals to the underworld ( e.g., Brady and Ashmore 1999; Lucero and Kinkella 2015 ; Prufer and Brady 2005 ; Woodfill et al. 2015 ). Many of these were pilgrimage sites where votive offerings were left, human remains interred, and temples constructed. Ethnohistoric accounts indicate that Native Ame ricans in the southeastern U.S. also viewed features such as caves, waterfalls, and springs as portals or conduits to the invisible world (Hudson 1976:130). It certainly cannot be assumed that springs were similarly considered to be sacred liminal places, or portals, by ancient Floridians. Fo rtunately, the archaeological record of Silver Glen Springs offers much that is relevant to this question. Situating the Silver Glen Springs Complex Silver Glen Springs was introduced in the previous chapter, in the context of discussing the relationship o f the onset of spring flow to the earliest shell deposition. In brief, Silver Glen is a first magnitude spring in Marion County, one of the largest in Florida. It is proximate to the western shore of Lake George a flow through lake of the St. Johns River a nd the second largest body of water in Florida to which it is connected by Silver Glen Run. As noted in Chapter 2, the St. Johns River experiences flow reversals and saline intrusion as far upstream as Lake George, and so there is a confluence of upstream blackwater, downstream brackish water, and subterranean groundwater from Silver Glen (and other) springs. Terrestrial environs are likewise diverse, as the narrow strip of hydric lowlands fronting the western shore of Lake George abut xeric uplands that ri se rapidly to greater than 25 m amsl. The Silver Glen Complex encompasses a diverse array of dense archaeological deposits surrounding Silver Glen Springs and both the northern and southern shores of
160 Silver Glen Run (Figure 4 2). The Complex straddles the border between Marion and Lake counties, necessitating the identification of multiple sites to satisfy state recording standards. These divisions are somewhat arbitrary, as there is a near continuous distribution of archaeological deposits on the landscape . In addition to straddling two counties, the Complex also crosses multiple ownership and jurisdictional boundaries. The land surrounding the spring and northern shore of the run are part of the Silver Glen Springs Recreation Area, managed by the U.S. Fore st Service. The surface water bodies are under jurisdiction of the Florida Department of Environmental Protection and St. Johns River Water Management District. The southern shore and adjacent land of Silver Glen Run are owned by the non profit Juniper Clu b of Louisville, Kentucky. The largest recorded site in the Complex is 8MR3605/8LA1 (hereafter 8LA1), confined to Juniper Club lands south of Silver Glen Run. Since 2007, this site has been investigated by the University of Florida, Department of Anthropol ogy, Laboratory of Southeastern Archaeology (LSA), through its St. Johns Archaeological Field School, directed by Kenneth Sassaman. Additional testing has been conducted by the LSA in conjunction with researchers from University of Oklahoma. Multifaceted f ield investigations involved bucket auger and shovel test survey, surface collection, remote sensing, and test unit excavation (Gilmore 201 4 ; Randall 2010; Sassaman et al. 2011). To facilitate communication, the site has been divided into eastern and weste rn halves (8LA1 E and 8LA1 W). A linear embayment, or seep spring, intervenes between 8LA1 E and 8LA1 W. Site 8LA1 W has been further subdivided into four archaeological loci based on the density, age, and character of archaeological deposits (Locus A Locu s D; Figure 4 3).
161 Beginning at the east, at the mouth of Silver Glen Run, site 8LA1 E consists of the remnant of a large U shaped shell mound. This mound was noted repeatedly by visitors in the eighteenth and nineteenth century. John Bartram (1769:23) , for example, thick we could hardly pass through them . Jeffries Wyman provides the most detailed description: The one last mentioned is much the larger and consists of three po rtions forming as many sides of a hollow square. The first extending along the shore of the creek, near the mouth of which it has a height of from twenty to twenty five feet by measurement; the second is on the shore of the lake, and measures from a hundre d and fifty to two hundred feet in width, and the third extends inland at nearly right angles to this. Between these ridges is a deep valley, in which the shells are entirely wanting or are only sparingly found (Wyman 1875:39). It was apparently the larg est shell mound observed by Wyman on the St. Johns River, measuring some 300 x 200 m and 8 m high. This mound dates largely to the ceramic Late Archaic Orange period (ca. 4600 3500 cal B.P.), although there is some indication that it was founded on an earl ier Mount Taylor shell ridge (Sassaman et al. 2011). Just west of 8LA1 E lies a remnant Mount Taylor shell mound. This is Locus A of site 8LA1 W, described in the previous chapter, a linear shell ridge measuring some 200 m long by 75 m wide, paralleling Si lver Glen Run (Randall 2010:330 335 ; Sassaman et al. 2011:121 170). Shell was emplaced from ca. 6300 5750 cal B.P. and was thus largely restricted to the Early Mount Taylor phase. A second shell ridge, the Silver Glen Run North Mound (8LA4242) is situated directly across the run from Locus A, on the northern shore (Randall et al. 2011:90 98). Only recently documented, this ridge has been largely destroyed by shell mining, much like Locus A of 8LA1W. S ubsurface testing has not been conducted at 8LA4242, b ut its size (~185 x 85 m)
162 location and the lack of pottery on the surface indicate that it is likely a Mount Taylor shell mound and may be coeval with Locus A. Moving f a rther west, three topographic highs, or ridge noses, overlook the south shore of Silver Gl en Run. Site 8LA1 W Locus B occupies a slight topographic rise some 200 m from Locus A (Gilmore 2011, 201 4 ). This portion of the site is well stratified with both Thornhill Lake phase and Orange period deposits. Locus C of 8LA1 W occupies a ridge nose over looking Silver Glen Springs, immediately west of Locus B. Secondary refuse and habitation deposits are abundant, dating largely to the St. Johns period. The next topographic high to the west is Locus D, a shell free midden roughly contemporaneous with Locu s C. The western terminus of the C omplex includes archaeological deposits consists of lithic and bone artifacts discovered by divers in the early 1960s. These were reco vered 30 45 m back inside the cave system of Silver Glen Springs (Dunbar 1990). More recently a possible dugout canoe was uncovered beneath the sediments of the spring pool (8MR3554 [Porter 2009]). Extensive subaqueous deposits have also been noted along t he southern shore of the spring run, including shell, vertebrate fauna, human remains, and both Orange and St. Johns pottery (Seinfeld 2013). Whether these deposits have been disturbed or remain in situ is not presently known. Terrestrial deposits adjacen t to the spring pool have been documented as site 8MR123. This area surrounding the spring pool. Although heavily reduced by mining, a recent survey and testing program has documented the prese nce of intact deposits proximate to the pool
163 and in the surrounding uplands (Randall et al. 2011). The majority of shell emplacement appears to have occurred during the Thornhill Lake phase and subsequent Orange period. The mound was crescentic to C shaped , covering an area of at least 350 x 200 m (Randall et al. 2011:196 197). Shell has been documented 2 m beneath the current water level of the spring, and as much as 4 m above it, suggesting a vertical shell bluff at one time emerged from the spring to a h eight of 6 m or more. The history of Silver Glen Springs is one of repeated alterations of the landscape, over several millennia, as different materials were deposited in different contexts and towards alternative ends. These landscape alterations register transformations in the occupation and use of the site and its material, historical resources. The following details the various pre Columbian deposits of the Silver Glen Complex, highlighting these transformations. Late Pleistocene and Holocene Antecedents Late Pleistocene settlements of peninsular Florida are recognized by the presence of a series of diagnostic hafted bifaces. In general, hafted bifaces are lanceolate shaped and may be either fluted or unfluted. The earliest of these are general ly classified as a variant of Clovis, dating as early as 13, 200 cal B.P. (Waters and Stafford 2007). Other forms include Simpson, Suwannee, and Dalton. The temporal placement of these latter types is uncertain, but they are generally thought to post date C lovis. argued that Paleoindians may have been semi sedentary and tethered to freshwater sources, such as first magnitude springs and large lakes, during the arid late Pleistocene in Flo rida (Dunbar 1991; Dunbar and Waller 1983; Neill 1964; Thulman
164 2009). Most Paleoindian sites in Florida are located in northwest peninsular Florida, where carbonate bedrock is near the surface and karst terrain well developed (Dunbar and Waller 1983). Only a handful of Paleoindian sites have been recorded in the St. Johns River basin. These include the Paradise Park site (8MR92) adjacent to Silver Springs (Dunbar and Doran 2010 ; Faught 2009; Neill 1958; Hemmings 1975;), and the Helen Blazes site (8BR27) in the upper St. Johns River valley (Edwards 1954 ; Rink , Dunbar, Doran, Frederick, and Gregory 2012 ) 1 . Thulman (2009) identified the area from Lake George to the Ocklawaha River as one locale with a high potential for Paleoindian sites, based on the availability of groundwater in the vicinity. The recently documented Lake George Point Site (8PU1470) lends some credence to this. The site consists of over 40 whole or fragmentary lanceolate hafted biface s , many resembling the Suwannee type (Thulman 2012) . These were purportedly recovered by local collectors beneath the shallow waters of northeastern Lake George. Systematic underwater survey by professional archaeologists failed to recover unequivocal evidence of Paleoindian occupation of the locale, altho ugh Pleistocene aged deposits were indicated by the presence of ex tinct faunal remains. Thulman ( 2012) attributes the lack of definitive Paleoindian artifacts to the intensity of collector activity in the area. At Crescent Lake, some 10 km northeast of Lak e George, near shore survey and analysis of collector assemblages from submerged sites indicated substantial 1 Dun bar and Waller (1983:19 20) include sites adjacent to Salt and Silver Glen springs as St. Johns no documented Paleoindian site adjacent to Salt Springs . The Guest Mammoth Site (8MR130 [Rayl 1974]) in Silver River is another purported Paleoindian site, but the association of the fauna and artifacts is dubious.
165 Paleoindian and Early Archaic activity (Sassaman 2003 b ). Diagnostic hafted bifaces numbering in the hundreds were recovered by collectors along the western margin of Crescent Lake, proximate to the Crescent City Ridge. Sassaman (2003 b :111) argues that Late Pleistocene and Early Holocene occupants likely targeted the edge of the ridge as this would provide a panoramic view overlooking the lake basin, likely a productive wetland biome at the time. He suggests revising extant Paleoindian settlement models to include escarpments and ridges overlooking wetlands in northeast Florida. A similar argument is offered by Thulman (2012), who suggests that Paleoin dians sought vantage points overlooking broad expanses, such as lake basins, in the St. Johns River valley and elsewhere. Evidence for Paleoindian and Early Archaic presence at Silver Glen Springs is pervasive, if ephemeral. No diagnostic artifacts have be en found, but several formal side and end scrapers were recovered during survey of the uplands north of Silver Glen Springs (Randall et al. 2011:205; Sassaman et al. 2011:77). In many cases these are heavily patinated. Lithic debitage, likewise heavily pa tinated, was often found below Middle Archaic diagnostic artifacts. Anecdotal evidence from local collectors indicates that Paleoindian fluted, lanceolate hafted bifaces have been recovered from the spring. Similarly, several of the artifacts illustrated b Cave Site (8MR162) within the cave system of Silver Glen Springs, appear to be Paleoindian or Early Archaic forms. Activities taking place at Silver Glen Springs prior to 9000 cal B.P. were apparently widespread a nd diverse, based on the distribution of lithic artifacts in shell free deposits (Randall et al. 2011; Sassaman et al. 2011). However, radiocarbon assays
166 are lacking and little can be said about the intens ity of activities taking place or the nature of tho se occupations. The potential no doubt exists for a well preserved site beneath the waters of Lake George, perhaps with associated organic materials that might lend some insight into occupational activities. Locus A of 8LA1W contains the earliest well docu mented deposits, where a series of pits were repeatedly excavated and infilled for a variety of purposes (as discussed in Chapter 3). At least two episodes of pit digging are apparent, one from ca. 9000 8000 cal B.P. and a second pre dating 6300 cal B.P. T he earliest of these were eventually capped with sand and a shell ridge emplaced , shifting the pattern of deposition from excavation/infilling to accumulation. Circumstantial evidence suggests that a similar process transpired across the run at 8LA4242. T hornhill Lake Phase An appreciable transformation of the use of space at the Silver Glen C omplex was underway at ca. 5800 cal B.P. Regionally, this is coincident with the onset of the Thornhill Lake phase, marked by the abandonment of many earlier Mount Ta ylor shell ridges, establishment of a new mortuary program, and expansion of regional exchange networks (see Chapter 3). At Silver Glen, this transition manifested itself first with the abandonment of Locus A as a place of daily living by 5700 cal B.P. (Sa ssaman et al. 2011:168) . Excavations have documented an expansive deposit of sand atop the remnant shell ridge, nearly 50 cm thick. Its position at a high elevation relative to surrounding landforms suggests that this wa s not the result of colluvial or all uvial processes. Likewise, its widespread and uniform occurrence argues against it being a byproduct of shell mining in the vicinity. Rather, it appears that inhabitants of Locus A capped the shell ridge when it was no longer used as a habitation place.
167 Th e practice of capping abandoned habitation sites has deep antiquity in the St. Johns River valley and has historical precedents at the Silver Glen Complex. Recall from Chapter 3 tha t places like Live Oak Mound, Hontoon Dead Creek M ound , and Harris Creek or iginated as habitation spaces that were then capped with deposits of whole shell before being transformed into loci for communal ritual and/or the interment of the dead. Similarly, shell processing pits underlying the Locus A shell ridge were capped with b rown sand prior to the accumulation of living surfaces and midden deposits (Randall 2013) . Whether there was a functional or symbolic difference between shell caps and sand caps is unclear, but regardless there was clearly a need to physically mark the ces sation of one activity at a place prior to the initiation of another. After ca. 5700 cal B.P. L ocus A was no longer the site of significant deposition, whether of shell or other materials , although later Orange and St. Johns era materials have been recove red in small amounts. At approximately the same time, Thornhill Lake era inhabitants began transforming previously unmodified portions of the Silver Glen Springs landscape. This is evident at Locus B of 8LA1 W, situated a mere 200 m southwest of Locus A an d 80 m south of Silver Glen Run. The basal component of Locus B consists of a series of repeated settlements and depositional hiatuses spanning the Thornhill Lake phase, from ca. 5700 4600 cal B.P. (Gilmore 2011; 201 4 :145). These are evident stratigraphica lly as a number of stacked, crushed shell habitation surfaces with layers of organically enriched soil between them. Shell layers are comprised principally of crushed bivalve and apple snail, and contain lithic tools and debitage, marine shell, bone tools, unmodified vertebrate fauna, and paleofeces. Many small conical or basin shaped pits also originate from these surfaces. The intervening
1 68 sand layers, in contrast, are devoid of both shell and artifacts. Gilmore (2011:25) estimates that a minimum of four s equences of habitation and abandonment are represented in these deposits. The end result of these practices is an elevated dome of shell, approximately 1 m high, which is reminiscent of the shell nodes documented at the Hontoon Dead Creek Village site (see Chapter 3 ). The habitation related activities evident in Thornhill Lake deposits at Locus B are not dissim ilar from those that preceded them at Locus A. However, coincident with the onset of habitation related deposition at Locus B, inhabitants initiated a radically different transformation of the landscape immediately surrounding the pool of Silver Glen Springs (Randall et al. 2011). Although the land surrounding the spring has been heavily modified by twentieth century shell mining and recreation al activities, testing by researchers from Florida State University documented intact archaeo logical shell deposits rising 3 m above the modern surface (Marrinan et al. 1990; Stanton 1995). Three radiocarbon assays indicated that the deposits date primarily to the Thornhill Lake phase, ca. 5650 4620 cal 20 10 and 2011 excavations into the mining escarpment approximately 40 m north of the spring pool were expected to document similarly deep shell deposits (Randall et al. 2011:112 121). Instead, o nly a thin layer of intact shell (Stratum III) was found beneath modern overburden. This shell overlies a deposit of brown and yellow sand (Stratum IV) within which human remains were discovered. Based on the slope of the contact between Stratum II and Str atum IV, this was likely a mounded mortuary deposit (Randall et al. 2011:116 117). A radiocarbon assay indicates that the earthen mortuary was emplaced between 5850 and 5590 cal B.P. The shell overlying the
169 mortuary consists primarily of bivalve and apple snail s in an ashy matrix with abundant charcoal, deposited between 5590 and 5320 cal B.P. Many of the bivalves were closed and paired, indicating that they were not cooked or trampled. Randall and colleagues thus suggest that shell was used to cap the mort uary, either immediately following its emplacement or at a slightly later date. The mortuary was heavi ly impacted by mining, and so i ts original size is uncertain . However, this is not an isolated interment (Randall et al. 2011:37) ; Potter (1935) noted tha t human remains were found in great numbers when shell was mined from the site. Further, photographs published by Norman (2010) show flexed burials in a similar context to that documented by Randall and colleagues (i.e., in a dark earthen matrix beneath sh ell). Finally, human remains of unknown age and provenance have been documented beneath the water by both visitors to the spring and avocational and professional archaeologists (Seinfeld 2013 ; Willis 1972 ). The mortuary mound was capped with shell after 5 600 cal B.P. Shell was simultaneously being deposited in two other locations nearby. As noted above, the 1990 Florida State University investigations found 3 m of intact shell deposits, approximately 100 m east of the mortuary (Marrinan 1990). Like the mor tuary, this deposit abuts the terrace edge, but this shell was emplaced directly atop an A horizon, perhaps as early as 5650 cal B.P. Randall and colleagues (2011) also documented subsurface shell along Silver Glen Run, about 100 m east of the mortuary. Th e presence of well preserved, uncharred botanicals indicates that this shell was deposited subaqueously (Randall et al. 2011:88). A radiocarbon assay on nutshell from the base of this deposit was identical
170 to that from the shell cap over the mortuary (5590 5320 cal B.P.). The implication is that shell was being deposited in three distinct modes capping an earthen mortuary, atop the ground surface, and within the water penecontemporaneously near the onset of the Thornhill Lake phase. These deposits ar e array ed on a line roughly 200 m long that parallels the spring run, and thus likely represent basal portions of a nascent shell ridge. Over the next millenni um, shell continued to be deposited, likely forming a ridge that abutted the terrace edge and prograd ed into the water, partially or wholly filling wetlands on the margin of the spring run and reaching a height of 3 m or more (Randall et al. 2011:203; 2014:27 28). How much of this shell deposition was related to daily living, and ho w much to mortuary acti vities is not clear from the available data. However, human remains were encountered in a shovel test proximate to the subaqueous shell deposits (Randall et al. 2011:78 79). Further, there is circumstantial evidence that a second, coeval mortuary existed a t the mouth of Silver Glen Run (Randall 2010: 219; Sassaman et al. 2011). Here, a single human interment was exposed by erosion and subsequently reburied. It was found amidst densely concreted shell s that likely represent basal deposits of the monum ent at site 8LA1 E (Gilmore 2014 ; Sassaman et al. 2011). Concreted shell is frequently found at Mount Taylor era shell ridges and microliths, often associated with Thornhill Lake phase deposits, and ha s also been documented at site 8LA1 E (Sassaman et al. 2011:79 ). The distribution of concreted shell on the surface and in bucket augers (a linear expanse approximately 150 m x 50 m paralleling the spring run [Sassaman et al. 2011:38 42]) is likewise indicative of a remnant Mount Taylor shell ridge.
171 In addition to th e deposition of shell in new locales, and for alternative ends, during the Thornhill Lake phase, there is evidence for coeval shell free deposition in the uplands surrounding the spring (Randall et al. 2011; 2014). Close interval shovel testing documented at least two concentrations of lithic debitage and tools in the absence of shell. Although no datable samples were recovered, these activities likely date largely to the Mount Taylor period as 95% of the hafted bifaces recovered were Florida Archaic Stemme d variety and nearly half of the debitage was thermally altered (Randall et al. 2011:205). Whether these remains derive from Early Mount Taylor or Thornhill Lake phase activities is unclear, although a similarly isolated lithic production area was document ed at the Lake Monroe Outlet Midden, which dates to the Thornhill Lake phase (Archaeological Consultants, Inc. and Janus Research 2001). Thus, the onset of the Thornhill Lake phase at Silver Glen Springs is reflected in the capping and abandonment of the E arly Mount Taylor shell ridges at 8LA1 W Locus A and (presumably) the Silver Glen Run North Mound (8LA4242). Mortuaries of earth and shell were emplaced at both ends of Silver Glen Run north of the spring pool (8MR123) and at the confluence with Lake Georg e (8LA1 E) coincident with (or just prior to) the establishment of linear shell deposits in the same locales and a habitation space at 8LA1 W Locus B, halfway down the run. Near the end of the Thornhill Lake Phase, the habitation space at Locus B was cappe d with shell and abandoned for a sufficient duration for pedogenesis to proceed on the surface, perhaps a century or two (Gilmore 2011, 201 4 ). The onset of the Orange period brought with it a new technology pottery as well as further transformations in the practices taking place at Silver Glen Springs.
172 Orange Period The onset of the Orange Period (4600 3600 cal B.P.) is marked by the initial appearance of pottery in the St. Johns River valley and greater northeast Florida. Orange pottery is tempered with p lant fibers, typically Spanish moss ( Tilland s ia usneoides) , and has affinities to Stallings wares from the Savannah River and Atlantic Coast of Georgia and South Carolina (Sassaman 1993). Current interpretations indicate that Stallings pottery spread from the Atlantic Coast along pre existing exchange routes into northeast Florida no later than 4500 cal B.P. (Sassaman 2004). Orange pottery is most abundant in northeast Florida, although related fiber tempered wares have a much broader distribution. Early i nterpretations of the Orange period suggested that pottery arrived with little fanfare and that life remained largely unchanged from the earlier Mount Taylor period (Milanich 1994:88). Miller, for example, writes: By this time, adaptation to the river envi ronments must have been firmly established, and indeed, despite this new technological development, it is likely that the settled or semipermanent occupation of the Late Archaic hunters, gatherers, and shellfish collectors continued with little change unti l the introduction of domesticated plants (Miller 1998:70 71). However, more recent investigations into Orange period developments revealed several dramatic transformations in settlement patterning, technology, mortuary ritual, and shell mounding (e.g., Gi lmore 2014, 2015; Randall et al. 2014; Russo 2004; Sassaman 2010; Sassaman and Randall 2012; Saunders and Russo 2011). The introduction of pottery to the region brought with it a decline in the use of both lithic and marine shell tools, which are found in much less abundance at Orange period than earlier sites (Gilmore 2011, 201 4 ; Sassaman 2003 c ). Likewise on the decline were exotic exchange objects, such and bannerstones and beads, which were
173 often interred in Thornhill Lake phase mounds. Instead, ornatel y decorated pottery appears to have become ensconced as the exchange medium of choice (Gilmore 201 4 ). Perhaps the most dramatic change is evident in the location and disposition of Orange habitation sites and shell mounds. Whereas Mount Taylor habitations were typically arrayed in a linear fashion and accreted in a tell like manner to form shell ridges, Orange habitations are arrayed in an arcuate or circular pattern and tend to be thin and ephemeral (Sassaman 2003 a ). The stacked accumulation of living surf aces evident at many Mount Taylor sites have not been documented in the Orange period. Further, there was a marked concentration of shell mounding. Nearly all Mount Taylor shell ridges were abandoned, and the piling of shell was limited to a few shell moun d complexes (Gilmore 201 4 ; Randall and Sassaman 2010). Similarly, wh ereas plain Orange pottery is widely distributed, decorate Orange wares are limited to mounded contexts. Four of these complexes have been documented, roughly equidistant from each other a long the middle and upper St. Johns River. Each of these was founded upon an earlier Mount Taylor mortuary mound that was used as the base for building massive, multi lobed shell complexes. From south to north these are Orange Mound, Old Enterprise, Harris Creek, and the Silver Glen Complex. Orange period activities at Silver Glen Springs included both expansion of existing shell ridges and the establishment of new deposits. However, not all earlier deposits were treated equally. The two Early Mount Taylor shell ridges 8LA1 W Locus A and 8LA4242 continued to be avoided, as they were during the Thornhill Lake phase. In contrast, the Thornhill Lake mortuaries established at the spring pool
174 (8MR123) and, presumably, at the confluence with Lake George (8LA1 E) w ere significantly expanded to form two of the largest pre Columbian constructions in the entire St. Johns River valley . As noted above, the mound at the mouth of Silver Glen Run, 8LA1 E, was described by Wyman (1875) as the largest shell mound on the St. J ohns River. It was a massive U shaped construction, open toward the west, on the south bank of the run. Although largely destroyed by twentieth century shell mining, subsurface survey effectively documented the footprint of a mound extending some 300 m alo ng Silver Glen Run and 200 m along Lake George (Sassaman et al. 2011). Wyman indicates that the northern arm of this construction was broader and higher than the southern arm, reaching a maximum elevation of approximately 8 m in the northeast corner. Bucke t auger survey along the shorelines in the vicinity of the north ridge indicates that shell, frequently concreted, is present as much as 2.5 m beneath the modern ground surface. This is likely a remnant of a Mount Taylor mortuary mound that was subsequentl y covered over by Orange period shell deposition. Test units excavated in the footprint of the north ridge indi cate that near surface deposits were heavily reworked by mining. Intact deposits were located, but are largely inaccessible beneath the current w ater table. Artifacts recovered along the surface and in near shore locales consist largely of highly decorated Orange wares. Multiple radiocarbon assays on soot adhering to pot sherds and Spanish moss fibers on their interior suggest that this ridge was e mplaced relatively early in the Orange period, ca. 4600 3800 cal B.P. (Gilmore 201 4 ; Sassaman et al. 2011).
175 Testing of the smaller southern ridge of 8LA1 E indicate s a notably different depositional history. Intact shell deposits were found to overl ie basa l sands indicating that, rather than being emplaced over an existing shell ridge, the southern arm was built up directly on the ground surface (Sassaman et al. 2011). The artifact inventory was relatively sparse, but the ceramic assemblage was dominated by plain Orange pottery, in contrast to the predominantly decorated wares from the north ridge. It also appears that the southern ridge may have been a slightly later addition to the C omplex. A basal pit feature was radiocarbon dated to 4060 3830 cal B.P., s uggesting that the bulk of the southern ridge was built up late in the Orange period. Importantly, Orange period mortuary remains have not been documented at this or any of the other Orange shell mound complexes, indicating that the Mount Taylor mortuary m ounds were re purposed for a different use. A second, slightly smaller, shell mound was also described by Wyman, surrounding the pool of Silver Glen Springs itself (8MR123). Survey and testing by Randall and colleagues (2011) documented the remnant of this mound, which covered an area of 350 x 200 m and reached perhaps 6 m in height. The Thornhill mortuary mound north and east of the pool was, similarly to 8LA1 E, renovated and repurposed by Orange inhabitants. That Orange activity here was intensive is att ested to by the abundance of decorate d Orange pottery collected from the spring pool (Gilmore 201 4 ). Shell mounding during the Orange period was focused on the spring pool, accreting south and west from the Thornhill ridge. Subaqueous shell east of the spr ing pool, apparently deposited directly in the water, document s the progradation of shell into the spring run during the Orange period. The existing Thornhill shell ridge was expanded
176 some 60 m south, into the spring run, and 40 m west to the edge of the c urrent spring pool by 4520 4240 cal B.P. (Randall et al. 2011:87). Shell deposits in the same locale, but at a higher elevation, were radiocarbon dated to 4410 4080 cal B.P., indicating that some 2.5 m of shell was emplaced rapidly during the Orange period . This subaqueous shell deposition effectively re oriented the outflow of the spring from a relatively straight, eastward flow to the south then east dogleg present today (Figure 4 4). S hell continued to accumulate, wrapping around the northern and western shore and surrounding the In addition to the building up of shell at either end of Silver Glen Run, Orange peoples also engaged in subterranean manipulations in multiple locales on the S liver Glen landscape (Gilmore 2011, 201 4 ; Sassaman et al. 2011). Notably, Locus B of 8LA1 W was transformed from a habitation site to a specialized shellfish processing facility. Discussed above, Locus B occupies a relatively flat ridge nose approximately 80 m south of Silver Glen Run (Gilmore 2011, 2015). Deposition began during the Thornhill Lake phase as a series of crushed shell living surfaces interspersed with accumulations of shell and artifact free sand. These are interpreted as repeated small scal e occupations and abandonments, and wo uld eventually form a small (~ 1 m high) crescentic shell node. Prior to the Orange period, a thin mantle of shell was emplaced and Locus B was abandoned until, sometime after 4600 cal B.P., the directionality of deposi tion shifted form building up to digging down. Gilmore (2011, 201 4 :154 157) identified three depositional patterns and attendant transformations at Locus B. The first pattern is the Thornhill Lake phase occupation already described. Depositional pattern 2 involved the excavation of scores
177 of massive pits geared ostensibly towards shellfish roasting. These pits are densely concentrated at Locus B, frequently intersecting and overlapping one another , and forming a complex stratigraphic record. Thermal alterat ion at the base o f many pits indicates that they contained fires, likely for the purpose of roasting shellfish . Pits are typically large, ranging from 70 120 cm in diameter and greater than 1 m deep, and of variable shape. Pit fill is likewise variable, wi th some containing sparse shell and others abundant shell. Likewise the diversity of shellfish species, the degree of crushing and/or burning, and the internal configuration of the pit fill are highly variable. After ca. 3900 cal B.P., pit digging ceased a t Locus B and a cap of clean, whole Viviparus was emplaced (Gilmore 201 4 :157). This cap, measuring 30 50 cm thick, was largely devoid of artifacts, contained little non shell matrix, and was coextensive with the pits. This points to rapid deposition that c overed over and flattened the p ocked surface of Locus B. Evidence for habitation or domestic activities is largely absent at Locus B during the Orange period . However, arcuate or circular arrangements of materials and features indicative of habitation have been documented elsewhere at Silver Glen Springs. At both 8 LA1 E and in a bait field north of Locus B, ground penetrating radar (GPR) survey located an arcuate pattern of anomalies. At 8 LA1 E, circular anomalies 5 10 m across were arrayed in an arc that , if projected to a circle, would measure 80 m in diameter (Sassaman et al. 2011). These anomalies were found to be thin subsurface deposits of shell with moderate amounts of vertebrate fauna and plain Orange pottery. Anomalies in the bait field north of L ocus B likewise consisted of subsurface shell deposits, primarily small pits (Gilmore 201 4 :168 170). Unlike those from Locus B, these pits were smaller and contained ample evidence for (small scale) habitation. Two
178 radiocarbon assays suggest these occupati ons took place over the interval 4150 3900 cal B.P. Randall and colleagues (2011:206 208) documented a similar arcuate pattern in the uplands north of Silver Glen Springs. However, there was a notable lack of shell at this location . Rather, the distributio nal pattern of Orange pottery suggested a series of household clusters arranged in a circular pattern around interior plazas. These household clusters were, in turn, arrayed in a circular or arcuate pattern around a larger central plaza. In sum, Orange per iod landscape alterations at the Silver Glen Complex included the transformation of two Mount Taylor mortuary mounds into massive U or C shaped monuments, the digging of numerous, large, shellfish processing pits, and establishment of several small habita tion sites. Gilmore (201 4 ) argued that the activities comprising these deposits resulted from the initiation of massive extraregional gatherings at Silver Glen Springs. Technofunctional, petrographic, and elemental analys e s of pottery w ere brought to bear to differentiate locally produced pots from those imported from afar, and to discern patterning in stylistic and manufacturing techniques. These analyses indicate that pottery from the mounds (8LA1 E and 8MR123) ha s an appreciably higher p roportion of ornately decorated Orange pottery than any contemporaneous, non mounded context. Further, mound centered gatherings frequently involved pottery manufactured in distant locales, perhaps as far as Charlotte Harbor in southwest Florida (~200 km a way). In contrast, pottery from non mounded locales, such as Locus B, was almost exclusively local and undecorated (Gilmore 201 4 :313 314). These mounds, then, were the site of ritualized feasts aimed at
179 gathering and integrating participants from across a broad swath of peninsular Florida. Meanwhile, the massive shellfish processing pits at Locus B likely served to provision these feasts wh ereas ephemeral, circular encampments point to intermittent occupation rather than a sustained resident population. In (201 4 :313) estimation, the preponderance of decorated and nonlocal pottery at mounds points to the cultural diversity and ritual significance of gatherings : public, ritually charged consumption events that contrasted markedly with the small scale of the routine subsistence practices evinced elsewhere in northeast Florida during the Orange period . St. Johns Period The St. Johns period in northeast Florida is the local manifestation of the Woodland (3200 10 00 cal B.P.) and Mississippian (1000 500 cal B.P.) periods in the Eastern U.S. In the Southeast, the Woodland period is generally characterized by an increased reliance on pottery and horticulture and the appearance of widespread mound construction and cer emonialism (Anderson and Sassaman 2012:112 151). The onset of the Woodland in the Southeast is marked by the dissolution of regional interaction networks and a shift from large habitation sites to smaller, dispersed settlements. Fiber tempered pottery was no longer manufactured by this time, and was replaced by a variety of wares with differing tempering agents and decorative motifs. The most widespread of these in Florida are Deptford, Swift Creek, Pasco, and St. Johns (Milanich 1994:104 274). In the St. J ohns River valley , the Woodland period is the St. Johns I period, dating to 3500 1200 cal B.P. The dominant pottery type in this period is the speculate tempered St. Johns plain pottery. Florida archaeologists emphasize continuity with the
180 preceding Orange period, despite the fact that the interval 3500 2800 cal B.P. is poorly understood in the region (Milanich 1994:254; Miller 1998:77 86). This continuity is pa rtially a consequence of the frequent presence of St. Johns I components in the same locales as e arlier Orange period components and technological and stylistic overlap in Orange and St. Johns pottery (e.g., spicules in the paste of Orange pottery, St. Johns pottery with Orange decorative motifs [Cordell 2004; Gilmore 2014]) . There is an increase in s ite frequency at this time, continuing the trend with earlier periods, and thus a presumed increase in population. This trend continues in the subsequent St. Johns II period (1200 500 cal B.P.). St. Johns period sites in general are understudied, and littl e is known about the s ize or organization of villages and special use sites . The Mississippian period denotes the era when large, highly stratified societies emerged in the Southeast (Anderson and Sassaman 2012:152 190). Many of these would be classified a s chiefdoms in cultural evolutionary nomenclature. Individual polities were widespread at this time, but were not persistent and many political centers went through cycles of emergence, fl orescence, and collapse. Maize agriculture was widespread in the Sou theast, although the degree to which it was practiced in Florida is debated (Ashley and White 2010) . Monumental architecture, with numerous mortuary and platform mounds arranged around plazas, hierarchical settlement patterns, stratified social organizatio n, and regional exchange and interaction, perhaps in the context of shared religious ideology, all characterize Mississippian societies in the Southeast. T he degree to which communities in Florida participated in, or were peripheral to, and the socio political transformations it entailed is debated
181 (Ashley and White 2010). In the St. Johns River valley , the archaeological culture coincident with the Mississippian period is referred to as St. Johns II (Milanich 1994:262 274; Miller 199 8:86 87). The shift from St. Johns I to St. Johns II is largely recognized by the appearance of check stamping as the dominant surface treatment on pottery . Again, in terms of basic lifeways, the St. Johns II period is largely characterized by continuity w ith what came before, but with an increase in population. Despite this continuity, some changes are evident. Hafted bifaces became smaller as seen in the Pinellas, Ichetucknee, and Tampa types and may be indicative of bow and arrow technology . Burial mound s became larger and, late in the St. Johns II period, some took the form of large pyramidal mounds, similar to those at Mississippian sites elsewhere in the Southeast. However, maize agriculture was likely not practiced, or was not widespread, and the soci al structure of St. Johns II co mmunities is poorly understood (A shley and White 2012). It is not known to what extent St Johns people used or added to the shell monuments that reached their zenith during the Orange period. Survey and testing at 8MR123 fail ed to document significant post Archaic deposits. However, a small deposit of shell on the western margin of Silver Glen Springs contained small St. Johns plain sherds and elongated specimens of Viviparus (Randall et al. 2011:90). The nature of this deposit, and whether it was redeposited by shell mining, is unclear. In contrast, reconnaissance survey of 8LA1 W recovered a large amount of plain St. Johns pottery, widely distributed across the landform (Sas saman et al. 2011:117 118). However, the densest concentration was at the western edge of the survey tract, and downslope, adjacent to the water of Silver Glen Run. Check stamped St. Johns
182 pottery was likewise concentrated to the west, particularly at Loci C and D, the two upland landforms (ridge noses) overlooking Silver Glen Springs. The St. Johns era deposits at Loc i C and D were inv estigated intensively from 2011 to 2013. The materials derived from these investigations are undergoing analysis and synthe sis by Elyse Anderson for her forthcoming dissertation. Presented here are the broad details of these investigations as they relate to the ongoing transformation of the Silver Glen landscape. Locus C is the larger of the two ridge noses, a relatively flat, well drained promontory with dense midden deposits. Reconnaissance survey in this area documented a ring of shell midden surrounding a shell free zone approximately 20 x 40 m in plan (Sassaman et al. 2011:114 115). Test unit excavation supported the infer ence for a circular St. Johns era occupation that apparently increased in intensi ty from the St. Johns I to the St. Johns II period s (Anderson 2015:83). Numerous pits were recorded, and a series of postholes arrayed in an arc suggests the use of circular s tructures. Downslope of Locus C, adjacent to the spring run, a dense shell midden was documented (Anderson 2015:84 100). This contained abundant vertebrate fauna, lithic debris, and other indicators of habitation activities. This midden apparently accumula ted over the course of the St. Johns period, although St. Johns plain pottery is more prevalent than check stamped. St. Johns pottery has also been documented within the spring run (Seinfeld 2013). A basal pit feature dating to the Orange period attests to earlier activities as well. Locus D of 8LA1 W comprises a smaller ridge nose west of Locus C. This landform is largely devoid of shell. Instead, a shell free midden was encountered th at
183 contrasts sharply with Locus C. In addition to the absence of shell, pit features documented were rich with charcoal and apparently served a different function from those at Locus C (Anderson 2015:137). No evidence for architecture was recovered, and vertebrate faunal remains were few. Likewise, there is no evidence for St. Johns I occupation. Instead, an abundance of large St. Johns check stamped sherds w as found. It is thus possible, if not likely, that specialized activities took place here. Whether this was a provisioning workshop or a locale for structured or ritualized deposition of check stamped pottery in pits is, as yet, unclear. A single radiocarbon assay indicates that these pits were dug early in the St. Johns II period, possibly predating St. Johns II deposits at Locus C. An additional deposit, presumably of St. Johns age, is notable. A small, conical sand mound lies approximately 150 m southeast of Locus C (Sassaman et al. 2011:103). Measuring approximately 20 m in diameter and 1.5 m high, it is similar to other St. Johns era burial mounds in the region. Given th e potential for recovering human remains, this mound has not been intensively tested. However, a high density of St. Johns plain pottery was recovered from shovel tests in the vicinity, indicating that it may have been constructed during St. Johns I times (Sassaman et al. 2011:118). In sum, St. Johns era deposits have been documented in numerous places in the Silver Glen Complex, but the best documented are restricted to landforms on the southern shore, opposite the spring. Activities during the St. Johns p eriod apparently did not involve significant shell mounding. This is consistent with shell mounds elsewhere in the St. Johns River valley , which received little additional shell after the Archaic period (Randall 2013 ). Occupation during the St. Johns I per iod appears to have been small -
184 scale, with little indication of the extraregional gatherings that characterized the preceding Orange period. St. Johns plain pottery is widely distributed across the C omplex, but intensive deposition during the St. Johns I p eriod has been documented in only the spring side midden. A mortuary mound was established southeast of the spring, possibly early in the St. Johns period. St. Johns II deposits include a circular village and plaza overlooking the spring (Locus C), downslo pe midden deposits, and a special use site immediately to the west (Locus D) that may have involved votive offerings in pits. Discussion: Transformative Histories at Silver Glen Human use of Silver Glen Springs went through several transformations over the course of some 8,000 years. The evidence for human visitation of the spring d uring the Late Pleistocene and e arly Holocene is tantalizing but thin. Extant settlement models certainly indicate that Silver Glen Springs would attract people as both a source of fresh water and for the panoramic views afforded from the nearby uplands. However, those early visitors have left few clues to the frequency, duration, and nature of their visits. Lithic debitage , microliths, unifacial tools, and the occasional hafted b iface certainly attest to their presence, but they are hardly representative of the richness of life and the suite of activities that likely took place. The presence of Paleoindian aged hafted bifaces placed more than 30 m back in the cave system (Dunbar 1 990) may point to votive offerings marking the spring as special or sacred. As water rose in the ensuing millennia people continued to gather at Silver Glen , bringing with them stone from the interior of Florida. By 9000 years ago, they began to dig into t he earth in at least one locale, beneath Locus A. The function of these pits is not clear; they contain the full suite of artifacts found in later deposits, but in much lower
185 density, and are largely devoid of shell (Randall 2014 a ). Gilmore (201 4 :174) argu es, in the context of later pit digging at Locus B, that, much like springs, the surface of the earth constitutes a boundary between the present and past, or between upper and lower worlds in many cultures (see also Knight 1989; Thomas 1999). Digging into the earth thus transgresses that boundary, and the removal and emplacement of materials in the earth, through pit digging and infilling, constitutes an exchange or interaction with the ancestors. Pit digging continued, likely intermittently, into the Mount Taylor period. At approximately 6300 cal. B.P., the pits were capped with a mantle of sand. This sand cap marked the end of digging down and the onset of building up as a series of shell nodes and associated refuse middens were formed. Over some 500 years , these deposits conglomerated into a linear shell ridge . Atop this ridge is a mantle of sand, possibly a final cap (Sassaman et al. 2011:169) . This practice of capping a habitation with sand or shell to mark the cessation of inhabitation or instigate a tr ansformation of use is not uncommon during the Mount Taylor period (see Chapter 3). Indeed, as it is the replacement of preexisting places of habitation with ceremonial platforms or This is the striking factor about Locus A, and other Mount Taylor places: later deposits are often isomorphic with earlier ones, mapping onto them (Randall 2010). The structuring influence of ancient deposits goes some way towards explaining the location of Early Mount Taylor s hell ridges at the Silver Glen C omplex. That is, why were they not emplaced adjacent to the spring pool but halfway down the run? The shell ridge at
186 Locus A was emplaced atop Mou nt Taylor aged pits, which are in turn co extensive with earlier, pre Mount Taylor pits. Thus, the activities of past inhabitants several millennia removed had a profound effect on later depositional practices. This same process may have transpired across Silver Glen Run, at 8LA4242. Whether these two ridges were coeval remains to be determined, but seems likely given there disposition and location. The influence of the past is not limited to Locus A, nor to the Mount Taylor period. Evidence of ancestral activity seems to have been the primary structuring force dictating the projects undertaken across the Silver Glen landscape. The paired Mount Taylor shell ridges along Silver Glen run were avoided after 5700 cal B.P., with the onset of the Thornhill Lake phase. In their stead, a small habitation was founded just to the southwest, at Locus B. At the same time, linear mounds were emplaced at the head and mouth of Silver Glen Run. Like the earlier shell ridges, t hese were paired and on opposite sides of the run, albeit not directly across from each other. In further contrast, these do not appear to have accumulated in a tell like fashion through repeated or sustained habitation. Rather, as evidenced by 8MR123, the y were first constructed as earthen mortuary facilities that were later capped with shell. The ephemeral nature of occupation at Locus B, punctuated by periods of abandonment, suggests that Thornhill Lake occupation of the Silver Glen Complex was intermitt ent and that mound building likely occurred during periodic gatherings attending mortuary rites. The spatial scale of these gatherings is attested to by the inclusion of objects from as far away as Mississippi in mortuary deposits elsewhere. Thus Silver Gl en was transformed from a place of relatively sustained, or at least intensive, habitation for the living during the Early Mount Taylor phase to a repository for the ancestors during the Thornhill Lake
187 phase. Earlier deposits (i.e., shell ridges) were appa rently avoided, but no doubt provided historical warrant for the construction of a mortuary here. Near the terminus of the Thornhill Lake phase a shell cap was emplaced atop the Locus B habitations (Gilmore 2014:145) . A period of abandonment followed, perh aps a century or two, until the bearers of Orange pottery arrived, encountering a landscape rife with residues of the past. Like earlier Mount Taylor residents , Orange period inhabitants of the St. Johns River valley largely abandoned earlier habitation si tes and established residential spaces in new locales. We know little of their mortuary practices, but these apparently did not involve interment in mounds of shell or earth. Rather, the act of emplacing shell was largely restricted to four locales best ex emplified by the Silver Glen C omplex in the context of massive social gatherings that drew in people from most of peninsular Florida. These gatherings took place at preexisting Mount Taylor mortuary mounds and involved their renovation and expansion into m assive, multi lobed shell constructs. The cultural diversity and ritual significance of these gatherings is attested to by the high frequency of nonlocal and elaborately decorated Orange pottery at mounds, relative to contemporary sites (Gilmore 201 4 :152). Thus at 201 4 :146). There is a persistent theme of duality in the Archaic co nstructions a t the Silver Glen C omplex. As noted above, the Early Mount Taylor shell ridges were paired, as were the later Thornhill Lake mortuary mounds. Whether these reflect a dualistic social org anization or the coalescence of distinct groups of people is unclear. But, these were
188 later renovated into paired U or C shaped mounds by Orange folk, adding a layer of complexity and introducing a duality to the internal structur e of the mounds. For example, it is know n E that the northern ridge was larger than the southern ridge. There is also evidence that it may have been constructed earlier, with the southern ridge a late addition, and it has a higher proportion of decorated Orange pottery (Gilmore 201 4 :152 153). This points to the poss ible coalescence of two distinct groups of people, perhaps descendant Mount Taylor people indigenous to the river valley and Orange pottery bearing immigrants from the coast (Sassaman and Randall 2012). Regardless, and in contrast to earlier Thornhill Lake phase mortuary gatherings that highlighted the status of individuals or groups, Orange period gatherings were integrative affairs. Indeed the abundant historical resources at Silver Glen Springs most conspicuously the paired Early Mount Taylor shell ridge s and paired Thornhill Lake phase mortuary mounds were likely an important factor in the siting of Orange gatherings, drawn upon to integrate diverse cultural groups and build communal bonds through the assertion of a shared history or ancestry (Gilmore 20 1 4 :322). The liminal qualities of Silver Glen Springs no doubt played a role as well, particularly when one considers the physiographic diversity in the immediate surroundings. That is, the spring is bordered by dry uplands and feeds into one of the larges t water bodies in the state, Lake George. At Lake George, the flow of the river frequently reverses direction as ocean tides push up the river. This brings with it a commingling of fresh river water and saline sea water. Thus Silver Glen gathers not only p eople and their attendant artifacts, but land both high and low, water dark and clear, fresh and saline.
189 Historical resources were likewise a significant factor motivating pit excavation at Locus B (Gilmore 201 4 :161 164). Excavated atop an existing Thornhi ll Lake phase shell node, the earliest pits excavated encountered residues of the past. Over the course of several centuries, as pits were excavated across Locus B, they began to intercept not only the Thornhill Lake shell nodes, but pits excavated earlier . Much like 4 :164). Over time, the exposure of ancient residues likely became an anticipated r esult of pit digging. Likewise, pit fill became increasingly complex, combining different species of shellfish in varying conditions (whole/crushed, burned/unburned) in potentially meaningful ways. Gilmore (201 4 :162) argues that these were, in effect, mate rialized historical narratives; evidence of past feasts and gatherings interred for the benefit of future inhabitants. After ca. 3 5 00 cal B.P., shell mounding appears to have ceased at both Silver Glen Springs and the mouth of Silver Glen Run. The excavati on of large pits at Locus B had likewise been discontinued by this time, and Orange pottery was no longer in use (Gilmore 201 4 :324). Early St. Johns period (i.e., St. Johns I) activities at the site apparently involved small scale habitation. St. Johns pla in pottery is prevalent, but dense midden deposits are restricted to the downslope portion of Locus C. Further, only a handful of the radiocarb on assays f r o m the Silver Glen C omplex fall in this interval, suggesting that occupation may have been ephemeral or intermittent. Regardless, at this time Silver Glen once again became an appropriate place for the interment of the dead, as evidenced by the sand mortuary mound southeast of Locus C.
190 The causal factors at play in these transitions are not altogether cle ar. There is some indication of increased precipitation and/or sea level at this time, but the evidence is contradictory (e.g., Filley et al. 2001; van Soelen et al. 2012). W ithin the Silver Glen C omplex, it is striking that St. Johns I deposits frequently contain shells that appear Viviparus like, but that are markedly more elongated than the typically globular forms (Randall et al. 2011:204). Whether this is a result of shell plasticity in the face of local changes to hydric habitats (e.g., hydroperiod, f low velocity, water chemistry) is unclear. Alternatively, these may be a genetically distinct subpopulation of Viviparus with morphological changes resulting from genetic drift, or they may not be Viviparus at all, but a distinct species of mollusk. Regard less, additional examples and analysis are needed to confirm the association with St. Johns I deposits and to untangle the roots of these unique shells. Morphological changes in shellfish populations may point to an ecological explanation for the cessation of shell mounding and the breakdown of extra regional gathering. However, Gilmore (201 4 :324) suggest ed that the roughly coincident ascendance of Poverty Point, in northeast Louisiana, as a gathering place may have disrupted the networks of interaction cen tered on Silver Glen Springs. Poverty Point was at the nexus of a pan regional sphere of influence that involved the importation of raw materials and finished goods from distant locales including the Great Lakes, the Midwest, the Gulf Coast, and the Appala chians (Gibson 2001; Kidder 200 2 , 2010 ; Ortmann and Kidder 2013; Spivey et al. 2015). The presence of St. Johns pottery at Poverty Point certainly implicates northeast Florida denizens in the gatherings there
191 (Hays and Weinstein 2004). Over the next 500 years, interaction and exchange networks throughout the Easter n U.S., including Poverty Point, would decline. Later St. Johns II period deposits at Silver Glen Springs involved the establishment of a circular village and special use locales. It is possibl e that the shell free deposits at Locus D include votive offerings in the form of check stamped St. Johns pottery interred in pits. Locus D is located on a ridge nose directly south of the spring pool, and would have faced the opening of the shell amphithe ater there. That the placement of the Locus C village was predicated on the location of the spring also seems likely. It lies on a high promontory overlooking the spring pool, but if proximity to the spring at elevation were the primary motive behind the p lacement of the village, it surely would have been constructed atop the shell mound encasing the spring pool. Rather, the village was placed opposite the spring, at the midpoint of a line connecting the spring and the sand mortuary mound to the southeast. This suggests that the location of the village has less to do with quotidian concerns than proper disposition with respect to the residues of the past. Silver Glen Springs thus witnessed a succession of alterations and transformations in deposition and us e. Often these transitions are marked by rapidly emplaced caps of sand or shell that obscure residues of the past and/or by periods of abandonment. Through a coarse lens, Silver Glen was initially occupied as a habitation space before being converted to a mortuary facility. The mortuary mounds were later co opted as historical warrants for extra regional gatherings and communal feasts. Following the breakdown of these interaction networks, a conical sand mortuary mound and votive deposits were emplaced, aga in absent evidence for sustained daily living.
192 Silver Glen again became a locus for intensive inhabitation during the St. Johns II period, with the construction of a circular village overlooking the spring. Taken in isolation, many of the pre Columbi an dep osits at the Silver Glen C omplex can be found Mount Taylor shell ridges, Thornhill Lake mortuaries, and villages and burial mounds of the St. Johns period are all located adjacent to springs. However, when looked at in its entirety, the Silver Glen C omplex is unmatched in the region. It is the gathering of these disparate elements that offers the best evidence for the sacredness of Silver Glen. Was Silver Glen Springs Sacr ed? At the outset I questioned whether springs in the St. Johns River valley are likely to have been considered sacred by default. That is, whether springs are archetypal feelings of awe, power, majestic 37) regardless of cultural background. Springs have many features in common that distinguish them from other water bodies in the region. The clarity of spring waters lends them certain optical qualiti es that enhance their striking ness but distort some features beneath the water. This has been extensively remarked upon with reference to Silver Springs the largest spring in the region and subject of the following chapter where the refraction of light ex aggerates the depth of the water and casts iridescent hues over plants, rocks, and animals. All of this is to suggest that springs are liminal places, where expectations of normality are called into question. Indeed, it is remarkable how many early America n visitors refer to Silver Springs as a striking, transformative place. For example, John
193 Or Major General George McCall (1974:150), who described his impression on first visiting the spring thusly: O! how my heart swelled with astonishment as we neared the centre of he line of demarcation between the waters and the atmosphere was invisible. Heavens! What an impression filled my mind at that moment! Were not the canoe and its contents obviously suspended in mid air like Mahomet's It is obviously inappropriate to assume that pre Columbian visitors to a spring would be so str uck by the experience. But, one cannot escape the fact that, particularly during the Late Pleistocene and e arly Holocene, Silver Glen Springs must have been an even more visually striking pla ce than it is today, with water emanating from a cave and wind ing down a valley into the expansive basin of Lake George. This basin may have been an emergent wetland rife with aquatic flora and fauna, or a dry prairie punctuated with channels and sinkholes , each containing water of its own. Particularly if approached from the west, from the xeric scrub of the Marion Uplands, this would have unfolded in a panoramic vista . The construction of shell mounds adjacent to springs added to this strikingness, or dua lity, further highlighting the depth of the spring by accentuating the height of the surrounding terrain. Indeed this may have been a desired outcome. If springs themselves were an unlikely source for the shell deposited on their banks, then the motivation to gather, transport, and deposit shells from other locales may have been to enhance the visual impact of springs, and create a contrast between high and low. This contrast and diversity is evident in the various deposits that people emplaced along Silver Glen Run the paired Early Mount Taylor shell ridges, paired Thornhill Lake
194 phase mortuary mounds, the south facing 8MR123 and west facing 8LA1 E again pointing to a coming together of distinct entities. Votive offerings at springs may also be evidence tha t they were sacred or special places. Hafted bifaces and carved bone pins were recovered from 30 45 m deep inside the cave of Silver Glen Springs, dating primarily to the Paleoindian and Archaic periods (Dunbar 1990). Whether these are votive offerings is unclear, but it is difficult to imagine anything but purposeful placement given their position on bare limestone deep within the cave. Later votive offerings consisting of ornately decorate St. Johns pottery interred in pits may also be present on the upla nds overlooking Silver Glen (Anderson 2015). Dugout canoes have frequently been documented at springs in the region , perhaps purpos efully sunken beneath their waters. However, this may be more a function of visibility than intent, as many canoes have been recovered from non spring locales (Wheeler et al. 2003). But, if springs writ large are potentially sacred by nature of their qualities, something else differentiates Silver Glen Springs. Perhaps more important than the physical parameters of the place is the 8,000 year record of landscape alteration and construction undertaken by pre Columbian visitors to Silver Glen Springs. It was the that exerted such a powerful draw to ancient Floridians and that they manipulated in the co ntext of gatherings for mortuary rites and communal integration (Gilmore 201 4 :316). This drawing on the past is not unusual, nor is the human made sacred places modeled on a core set of natu ral places but embellished with unique
195 Silver Glen Springs is steeped in history and the entirety of the surrounding landscape is sedimented with the residues of p ast human presence. The land 4 :173) by people representing multiple cultural groups from a broad geographical expanse. It is t he gathering, or entanglement, of so many disparate materials, objects, persons, and qualities both natural and cultural that makes Silver Glen unique. What is illustrated so well at Silver Glen Springs is the importance of history and sociality in the san ctification of place as social memory is materialized: this land is sacred to my people. Every hillside, every valley, every plain and grove has , of the northwest coast Duwamish tribe, in 1854, quoted by Turner [1989:192]). Indeed, the material residues of the past endure and are confronted and accommodated by people medium through which people inhabit and make sense of the world, and not merely a static representation of the past (Barrett 2001). Our contemporary engagement with springs is a case in point. We are drawn to these places in part because of the physicality of springs and their cool, clear water. But this physicality is in part the result of past human action for example as roads were built, shell mounds cleared, retention walls installed, and parks established. Take Weeki Wachee Spring, with its underwater theater and mermaid shows. Or Silver Spr ings, where state agencies are investing significant funds to remove artificial features and
196 Many are also drawn to springs for more intangible reasons, as the place of family ga therings, bellwethers of ecological vulnerability, and icons of an authentic Florida (e.g., Belleville 2011). They are the subject of personal recollections and public narratives. But springs are not merely backdrops that we endow with meaning and signific snorkeling in them, painting pictures of them, studying them, protesting their degradation, writing regulations about them that generate meaning and significance. The transformative history of Silver Glen Springs did not end 500 years ago, and Chapter 6 will pick up this thread again. But first, I examine one of the material gatherings manifested at Silver Glen Springs in more detail. Stone used in the manufacture of lithic tools is not found in the S t. Johns River valley . Wherever it is recovered in the valley , it points to an origin at a place far removed. In the following chapter I turn my attention west, to the gargantuan Silver Springs and the heartland of lithic quarries in Florida. I do so to ex plore the ways that springs are enchained across the landscape through the movement of humans and stone.
197 Figure 4 1. View of the northwest face of Mount Shasta. Photo courtesy of the U.S. Geological Survey ( http://3dparks.wr.usgs.gov/shasta/html2/shasta125.htm )
198 F igure 4 2 . Location of the archaeological sites comprising the Silver Glen Complex, on
199 F igure 4 3 . Aerial photograph of Silver Glen Springs, highlighting the archaeological deposits surrounding the spring and run
200 F igure 4 4 . Current topography ( bottom ) and reconstructe d, pre mining topography (top ) of the Silver Glen Complex. Reconstruction after Randall (2014 b )
201 CHAPTER 5 SPRINGS ON THE MOVE Springs are defined by motion . This is most obvious as water flows from passages in limestone with such force as to roil at the surface of azure pools. This water flowed through the limestone from recharge areas, whe re it fell to the surface as rain, worked its way down through dry sandhills, and entered the aquifer below. After emerging from its subterranean voyage, the water continues, flowing on the surface from spring pools into sometimes lengthy spring runs. And from these runs into lakes and rivers, and from those places into the ocean. Springs appear as discrete points on the landscape. But they are connected to places near and far by the motion of water both on and beneath the surface. T he social places that pe ople establish at springs are likewise defined by motion . Indeed, all places are the loci for gatherings of disparate elements , both material and immaterial. As Casey (1996:26) wrote, place is a catalyst for the collection, as well as the recollection, of all that occurs in the lives of sentient beings, and even for the trajectories of inanimate things. Its power consists of gathering these lives and things, each with its own space and time, into one arena of common engagement. This is particularly true of places like Silver Glen Springs, described in the previous chapter. The import of Silver Glen was derived from the gathering of people and things both contemporary and historical. Gilmore has effectively argued that the movement of pottery to Silver Glen S prings in the context of extra regional social gatherings, feasting, maintenance of a regional scale macrocommunity based on mound centered
202 But pots wer e not the only things gathered at Silver Glen Springs. Massive quantities of shell were mobilized and deposited in pits and mounds adjacent to Silver Glen Springs and its run. I have argued elsewhere that, because of the physical and chemical paramete rs of spring water, this shell wa s unlikely to have come from the 0 ). Indeed, shell bearing deposits are uncommon at all but the largest springs in the St. Johns River valley 2013). Likewise, the archaeological deposits adjacent to Silver Glen Springs contain an appreciable quantity of lithic debitage and flaked stone tools. The St. Johns River valley itself is stone poor, and outcroppings of material suitable for the manufact ure of stone tools (i.e., chert) are absent. Thus, whatever lithic artifacts are recovered from archaeological sites in the St. Johns River valley must have been transported from elsewhere at some point in their history . Chert source areas are located prim arily in northwestern peninsular Florida and the panhandle in a number of discrete deposits (Austin 1997; Endonino 2007; Upchurch et al. 1982). Not coincidentally, these are roughly coextensive with the areas of well developed karst in the state (see Chapter 2), and thus with the d ensest concentration of springs. The title of this chapter borrows from an article by Barbara Bender (2001), in which she attempts to reconcile diverging anthropological treatments of movement through landscapes that focus either on socio political and historical factors or on personal engagement and the affective qualities of place. Lithic materials in the St. Johns Riv er valley were transported from source areas more than 70 km distant. This entailed the movement of people across long distances, and likely through unfamiliar
203 terrains. Discussing the relationship between familiar and unfamiliar places, Bender (2001:85) w rote : the rooted, familiar place is never only that, but always surrounded and affected by unfamiliar spaces and attenuated relationships. Moreover, the be on the move. Thus some of the nomadic communities in Mongolia create an ego mundi the joining of earth to sky is recreated through the smoke that rises from the campfire at each resting point along the way. the mo ve always involves relationship, or familiarity, with place. The movement of people involves the movement of things too, and relationships with place are created, in part, by the material practices t hat people effected there. The movement and circulation of things has been addressed by a number of recent theoretical developments in anthropology and related disciplines (e.g., Joyce and Chapman and Gaydarska 2007; following distribution of deliberately fragmented things builds relationships and positions people in networks of interaction. Chapman built on the work of Mauss (1990  ) , Weiner (1992) , and others who elucidated the referential qualities of things , t hat is, the capacity of things to act as material mnemonics and retain tangible links to far removed times, places, persons, and events. Minimally, objects are indexical of t heir place of origin. Chapman argued that fragments of broken things retained the referential qualities of the circulated, the people who possessed them became linked (ench ained) through these referential networks.
204 Below, I review the results of recent archaeological investigations at Silver Springs, the largest spring in Florida. Silver Springs is located proximate to lithic source areas and is the western most spring with a direct hydrological connection to the St. Johns River valley . I argue that Silver Springs, and others like it, facilitated the movement of lithic materials to Silver Glen Springs and the greater St. Johns River valley . A t the same time the movement of st one effectively enchained people and springs a cross the peninsula of Florida , contribut ing to the social integration of gatherings at places like Silver Glen . Silver Springs Approximately 40 km west of Silver Glen Springs, across the upland scrub and wet pra i ries of the Ocala National Forest , is the famous Silver Springs. Silver S prings is the large st spring in Florida, spilling nearly 500 million gallons of water per day (7 35 ft 3 /s) from the Floridan Aquifer (Munch et al. 2006; Scott et al. 2004) . Silver Springs is not a single spri ng, but a group of 30 springs of varying size arrayed along a 1 km long linear expanse that forms the headwaters of the Silver River (Figure 5 1 ; see Butt et al. 2006 ) . The main spring, Mammoth Spring, lies at the western edge o f this expanse, in a steep walled pool that measures some 90 m across and 10 m deep (Scott et al. 2004:243 246). Water issu es from a 42 m wide oblong vent in the northeast portion of the pool and accounts for approximately 45% of the total discharge of Sil ver Springs . T he Silver River flows for 8 km through cypress swamp to the Ocklawaha River, and from there into the St. Johns River north of Lake George , making Silver Springs the westernmost major spring feeding into the St. Johns River valley . If springs are gems of the Florida landscape, then Silver Springs is surely the crown jewel.
205 Archaeological Sites at Silver Springs Silver Springs has been an important destina tion for tourism and industry sinc e the nineteenth century, and i s renowned for glass bottom boat rides over cool, clear waters . Attractions at Silver Springs have included hotels, railway stations, the Ross Allen Reptile Institute , various other zoological exhib its, concerts, jungle cruises, and fireworks shows. But the draw of Silve r Springs has not been limited to the recent past. Several pre Columbian archaeological s i tes have been recorded proximate to Silver Springs (Figure 5 2 ) . Within the spring itself, site 8MR59 encompasses historic and pre Columbian artifacts found on the fl oor and in the main cavern of the spring. This includes a fragment of a Paleoindian aged Suwannee point and remains of mastodon and mammoth (Neill 1964). Also within the spring is a canoe (8MR3173) that has been radiocarbon dated to 920 700 cal B.P. (Wheel er 2001). Site 8MR33 is a pre Columbia n burial mound excavated by Moore (1895:521 52 5). According t o Moore, the mound was located approximately 1.5 km east of the main spring, and measu red 15 m in diameter and over 1 m high . It consisted of unstratified ye llow sand and contained remains of multiple individuals, numerous pieces of pottery and lithic debris, as well as items of mica, copper, and shell. The mound was completely excavated by Moore, and so its age and precise location are unknown. Site 8MR1081 i s also recorded as a pre Columbian burial mound, al so of unknown age and location. Along the northern shore of the Silver River lie the Franklin 15 (8MR1082) and Lost Arrow (8MR3519) sites . Th ese are expansive, but low density , lithic and ceramic scatter s (Belcourt et al. 2009; Southeastern Archaeological Research 2008). Denser archeological deposits have been recorded along the southern shore. The Paradise Park site (8MR92) contains Paleoindian aged materials beneath more recent
206 preceramic Archaic and cera mic era deposits. These materials were exposed along the edge of a borrow pit, approximately 150 m long, excavated into an aeolian dune located south of the canal. The site was first investigated in the mid 20th century by Neill (1958), who excavated 11 te st units into the side of the borrow pit. He recovered fluted and unfluted lanceolate points and lithic debitage nearly 2.5 m below the surface, beneath more recent materials. Two stratigraphic units were recorded : an upper layer of homogenous sand and a l ower layer of sand with clay laminations or lamellae. This site is one of only two known in Florida with fluted points in stratified context. Hemmings (1975) later revisited the Paradise Park site, assessing its potential for further excavation. He confirm ed the stratigraphy noted by Neill, but reported fewer artifacts and no organic preservation. More recently, the site has been visited by Faught (2009) and Dunbar and Doran (2010). In 2003 Faught constructed a topographic map of the site and excavated thre e trenches into the wall of the pit. Although the stratigraphy of the site was again recapitulated, few artifacts were recovered (none below a meter) and diagnostic Paleoindian strata could not be identified. Analysis of sediment samples confirmed that the deposit was aeolian. Dunbar and Doran extracted two vibracores for ana lysis and OSL dating, bu t these have not yet been analyz ed . Given that none of the subsequent visits has recovered diagnostic Paleoindian artifacts, Dunbar and Doran (2010:14) speculated that this component of the site may have been completely excavated by Neill, or destroyed by mining . Site 8MR83 was recorded as a possible midden, just east of Paradise Park, adjacent to a dredged pool off of Silver Springs run (Thompson 1964). Recovered artifacts include historic ceramics, aboriginal ceramics, and lithic artifacts in a dark
207 sandy loam. Site 8MR93, initially visit ed by Griffin and Bullen in 1952 was recorded as diffuse camping debris revealed in several ditches that had been dug for garbage disposal. The boundaries of the site were later expanded after a survey conducted by SouthArc in 2002 (Dickinson and Wayne 200 2). This survey intercepted a high density of aboriginal artifacts primarily lithic debris and tools with some ceramics and further work was recommended to determine the pote ntial significance of the site. The documented arc haeological sites and historical resources indicate a virtually continuous occupation of Silver Springs from the Late Pleistocene to the present. It was in this context that t he University of Florida, Department of Anthropology, Laboratory of Southeastern Archaeology (LSA), conducted a C ultural Re sources Assessment Survey of an area within Silver Springs Park . This was done in advance of the transfer of park management to the Florida Department of Environmental Protection and the creation of Silver Springs State Park. Although the land wa s owned by the State of Florida, the park had previously been managed by a variety of private organizations, most recently Palace Entertainment, which negotiated a buyout of its lease in 2013. The reconnaissance survey undertaken by the LSA encompassed an area of approximately 104 acres immediately to the east of the main pool of Silver Springs, encompassing both administrative and recreational facilities slated for redevelopment. The project area was divided into n orthern (16.5 acres) and southern (87.5 ac res) tracts separated by the Silver River (Figure 5 3 ) , and included all or part of four archaeological sites: the Franklin 15 (8MR1082) site in the norther n tract, and sites 8MR 83, 8MR92, and 8MR93 in the sout hern tract.
208 Field operations entailed 25 m int erval systematic shovel testing acro ss the project area . A total of 469 shovel tests were excavated , 85 in the smaller northern parcel and 384 in the southern parcel . Of these, 406 were positive (i.e., contained cultural materials) while 63 were negative . The northern parcel feature d most of the recreational facilities, and ha d been subject to extensive land alteration. This is evident both in current conditions and historic aerial photographs. I t wa s thus not surprising that shovel tests frequently encount ered disturbed sediments and/or modern fill. However, in many cases the fill was relatively shallow and intact sediments were intercepted. Fifty five of the 85 shovel tests in this parcel were positive. These were concentrated in the northern and western p ortions of the parcel. Artifacts were uniformly distributed across the positive shovel tests; no obvious concentrations or voids are apparent. This resulted in the expansion of the Franklin 15 site boundary to cover a significant portion of the parcel. Mod ern disturbance and fill in the southern parcel w ere sparse and discontinuous. Most areas exhibited intact soil profiles beneath the upper 30 cm. Artifacts were found throughout the southern parcel. Of the 384 shovel tests excavated only 33 were negative. These were widely scattered and typically in disturbed areas. In general, artifacts were most densely concentrated in the nort hwestern portion of the parcel a nd least densely concentrated in the southeast. Along north south transects , artifacts were most numerous near the river and decreased in frequency away from the river, to the south. As a result of this survey, the boundary of site 8MR93 was expanded to encompass the e ntirety of the south ern parcel, including sites 8MR83 and 8MR92 .
209 The amount of mate rial recovered from the positive shovel tests varied considerably, with some containing only a single artifact and others containing several hundred. Artifact density in the shovel tests was as high as 4,200 artifacts per m 3 . By far the most frequently recovered class of artifact was lithic debitage and tools ( n =10,7 65 ; see Table 5 1 ). Several diagnostic bifaces were recovered, primarily Florida Archaic S temmed ( n =17), but also individual specimens of the earlier Kirk and Bolen types and later Hernando and Pinellas types. Pre Columbian pottery was infrequently recovered and was primarily sand tempered and plain, although several check stamped sherds are also present in the inventory . Other types recovered include Orange plain, Pasco plain , and St. Johns plain and stamped . Vertebrate faunal remains were found in only trace amounts. The abundance of lithic material, preponderance of Florida Archaic Stemmed hafted bifaces, and relative paucity of pottery indicates that the bulk of activity re corded took place during the pre Ceramic Archaic Mount Taylor period . The presence of pottery certain ly indicates later visitation to Silver Springs, but the s mall amount recovered suggest s that this activity was intermittent or ephemeral. Likewise, no ind ication of intensive habitation (i.e., midden deposit s ) was recovered . Notwithstanding the previously recorded sand burial mounds, neither of which could be relocated, mounded deposits of shell or sand were absent as well . Indeed, deposits of freshwater sh ell of any kind midden, mound, pit, or otherwise were lacking and only a small amount of freshwater shell was documented (Table 5 1 ) . It is notable that w e recovered nearly eight times more marine shell than freshwater shell (by weight), at a site some 70 km from the Gulf coast and 100 km from the Atlantic.
210 The paucity of pottery , vertebrate fauna, and shell in the inventory from Silver Springs is striking. The absence of intensive habitation remains is likewise remarkable, and unique to Silver Springs amon gst the largest springs in the region . These patterns could be a consequence of sampling error, but this seems unlikely given that more than 100 acres were surveyed on both sides of the river. If the lack of shell is explicable by the constraints po sed by clear, hyp oxic spring water detailed 2010 ) , it is nevertheless the case that all of the other first magnitude springs , and m ost of the second magnitude springs , in the adjacent middle St. Johns River valley have substantial shell dep osits adjacent to them . Further, numerous shell bearing sites have been documented along the Ocklawaha and Silver rivers, with the closest less than 2 km downstream of Silver Springs (sites 8MR53 and 8MR3266) . It seems likely, then, that the absence of she ll proximate to Silver Springs i s not coincidental, but reflective of de cisions made in t he past and a consequence of discursive activities t hat took place at Silver Springs. The abundance of lithic artifacts at Silver Springs is even more striking than the dearth of other materials. This is apparent when compared to the large springs in the middle St. Johns River valley . Table 5 2 summarizes the results of archaeological surve y at Alexander (Willis 1995), DeLeon (Denson et al. 1995), Salt (Dickinson and Wayne 1994), and Silver Glen (Randall et al. 2011) springs and test excavations at Blue Spring (Sassaman 2003 a ). Sample size was estimated roughly from the number and size of sh ovel tests excavated, and assumes a mean depth of 1 m. In the case of Blue Spring, the sample volume was determined from the number, size, and maximum excavated depth of test units. This sample volume was used to c alculate the density of
211 lithic artifacts a nd pottery. Blue Spring is notable for its dense pottery assemblage an d relative lack of lithic artifacts . However, it should be noted that these results reflect the testing strategy, which focused primarily on Orange era deposits. The remaining results su mmarized derive from reconnaissance surveys of relatively broad areas surrounding springs that encountered archaeological deposits of multiple ages. T he most notable pattern is the very high density of lithic artifact s and low density of pottery at Silver Springs. Although the specific densities calculated are likely skewed by errors in the estimated sample volume, differences of this magnitude are unlikely to be attributable to calculation error alone. This is confirmed by the ratio of lithic artifacts to pottery, which disregards excavated volume and thereby bypasses that source of error. With this metric as well, Silver Springs has a considerably greater proportion of lithic artifacts in the assemblage, relative to the other springs. Taken together, the a vailable evidence indicates that far different activities were taking place at Silver Springs than at springs in the St. Johns River valley proper. These activi ties were focused on the reduction of lithic materials . No r adiocarbon dates are available from these deposits, but the dearth of pottery indicates that these activities likely occurred largely during the pre Ceramic Archaic Mount Taylor period. As noted in the previous chapter, there is a notable decline in the frequency of lithic artifacts during t he succeeding Orange and St. Johns periods, so this is perhaps not unexpected. But, if this is so, then the lack of post Archaic deposition at Silver Springs is confounding and requires explanation. Alternatively, if these lithic reduction activities are n ot restricted to the pre ceramic Archaic, but reflect activities taking place in later periods as well , then it
212 is equally confounding that Orange and St. Johns activities would not include intensive deposition of some sort, as they do at other springs in the region. Regardless of the timing of these activities, it is clear that the primary activity taking place at Silver Springs was the mobil ization and reduction of lithic raw materials . T he geographical position of Silver Springs may explain th i s. Silver Springs is located on the western frontier of the St. Johns River valley , proximate to these lithic source areas. G iven the density of lithic materials recovered and its connection to the St. Johns River valley through the Silver and Ocklawaha rivers Silver Springs may have served as a gateway for the movement of lithic materials from west to east. The hypothesis that Silver Springs w as a gateway for transporting stone into the St. Johns River valley was evaluated through a combination of least cost analysis, lithic provenance, and debitage analysis. Debitage analysis allow s inference of the reduction activities taking place at Silver Springs and other locales. If stone w as moved from Silver Springs to sites in the St. Johns River valley , then li thic reduction activities occurring at Silver Springs should reflect early stage reduction and blank preparation. In this case, debitage recovered from sites downstream should be indicative of later stage tool manufacturing and retouch. Datasets comparable to Silver Springs are available for Mount Taylor era deposits at Silver Glen Springs (8MR123 [Randall et al. 2011]), the Lake Monroe Outlet Midden (8VO53 [Endonino 2007]), and the Thornhill Lake Complex (8VO58 60 [Endonino 2010]). Lithic provenance studie s indicate where lithic materials originated. I f lithic assemblages at sites in the St. Johns River valley proper e xhibit a similar distribution of source areas as Silver Springs , this would support the hypothesis that Silver Springs was a gateway. The alt ernative hypothesis is that the lithic materials
213 recovered from each site are dominated by the nearest source area rather than reflecting the redistribution of raw materials from Silver Springs. However, it is ce when determining the most proximate lithic source. Rather, least cost analysis , taking into account transportation costs and impedim ents to movement , provides a better indication of the nearest lithic source area. Further, least cost analysis can provid e an indication of the optimal or most economically rational routes from lithic source areas to the St. Johns River valley . If these routes pass through Silver Springs it would provide further support, and some explanation, for the role of Silver Springs a s a gateway. In the following, I begin with the least cost analysis to establish a baseline for the relative co st of transporting lithic materials from each source area to each of the four sites sampled here and to determine the routes by which this transp ort might have taken place. After this, the provenance determinations for lithic debitage and tools are presented and the patterns discussed. Finally, the results of debitage analysis are used to further evaluate the gateway hypothesis. Least Cost Paths As noted above, sources of chert are restricted to northwestern peninsular Florida and the panhandle. Chert is a microcrystalline quartz that typically forms as a replacement mineral. In Florida, chert is found in Lower Oligocene through Middle Eocene carb onates (Scott 2011; Upchurch et al. 1982). The weathering of clay rich Miocene sediments overlying these carbonates introduces large amounts of silica, dissolved in the groundwater. Under the right geochemical conditions this silica replaces the calcium ca rbonate in limestone, forming beds or nodules of chert. Relatively few geologic formations in Florida contain chert deposits that would have been available for
214 human exploitation. Those that do are the Ocala Limestone, Suwannee L imestone, St. Marks Formation, and the Arcadia (Tampa Member) and Peace River formations of the Hawthorn Group (Austin 1997; Endonino 2007; Scott 2001; Upchurch et al. 1982:23). extending north in an arc from Tampa Bay, along the Gulf Coast, and into the panhandle. A quarry cluster is must have been used by early man, and in which the chert is expected to be relatively uniform in f abric, composition These deposits are distinctive enough that samples can be discriminated on the basis of microscopic examination. The quarry clusters of Florida were originally defined by Upchurch and collea gues (1982), and later revised by Austin (1997) and Endonino (2007). The analysis presented below follows the quarry cluster delineation of Endonino (2007). Lithic materials from quarry clusters in the panhandle (Marianna, Wacissa, reme northern Florida ( Alapaha River, Swift Creek Swamp, White Springs, Upper Suwannee) have not been documented in the assemblages under consideration here , and so are excluded from further discussion. The remaining twelve lithic source areas are (from no rth to south) the Santa Fe, Gainesville, Lower Suwannee, Ocala, Lake Panasoffkee (East and West), Brooksville, Upper Withlacoochee, Caladesi, Hillsborough River, Turtlecrawl Point, and Peace River quarry clusters . In the following, Lake Panasoffkee East an d Lake Panasoffkee West are treated as a single quarry cluster because of their proximity and to simplify the analysis.
215 In order to explore th e hypothesi s that lithic materials moved through springs, it is first necessary to derive an accurate understandin g of the proximity to , and cost of , transporting lithic materials into the St. Johns River valley . Although a Euclidean, or straight line, measure of distance provides one indication of proximity and cost, it fails to consider several mediating factors, su ch as mode of transportation and terrain heterogeneity (e.g., boundaries, elevation, and land cover). A more realistic model of proximity and travel cost can be derived through least cost analysis. Least cost analysis is a spatial analytical technique for determining optimal travel routes over a terrain ( Bell and Lock 2000; Connol y and Lake 2006; Howey 2007 ; White and Surface Evans 2012 ) . Typically this is done within a Geographic Infor mation System (GIS) using a variety of environmental inputs. Calculation of least cost paths is essentially a three step process. The first step is to develop or define a cost of passage or friction surface that approximates the cost of travelling across a terrain. This is usually a raster surface with each cell value represen ting the cost of travelling across that cell. The second step uses this friction surface to derive a cost surface that models the cumulative cost of traveling outward from an origin point. The final step is to trace the path of least cost from a destinatio n or destinations back to the origin point. Accuracy of the final model depends on both the user defined input (the friction surface) and the algorithms used to derive the cost surface and cost paths. Least cost analysis was conducted to examine the optima l routes between the source areas of lithic raw materials and four archaeological sites for which lithic provenance data are available. This was done to determine the proximity of the source
216 areas to the sites (i.e., which source area is least costly?) and to determine t he optimal route for moving lithic materials into the St. Johns River valley . A friction surface was developed for peninsular Florida encompassing the St. Johns River valley and lithic source areas. Lithic sources in the Florida panhandle were not included in this analysis as they have not been found in any frequency at sites within the St. Johns River valley . Two variables were incorporated into the friction surface: slope a nd surface water. Primary digital datasets included a 15 m resolution Digital Elevation Model (DEM) and shapefiles containing surface water features digitized from USGS 7.5 minute topographic quadrangles. These datasets were obtained from the Florida Geogr aphic Data Library (FGDL). I assumed in this analysis that travel by canoe was a more likely mode of transport than walking, given the density of vegetation across much of peninsular Florida and the potentially prohibitive weight of the lithic materials be ing transported. Therefore, the friction surface was biased towards water travel. The surface water shapefil e was processed to remove modern features (e.g., canals, retention ponds). Swamps and wetlands were then removed from the dataset as these were thou ght to be impediments to travel an d/or not present during the middle Holocene. Finally, smaller water bodies (ephemeral streams and lakes smaller than 0.25 km 2 ) were clipped out of the dataset. However, smaller water bodies named in the Geographic Names In formation System were retained. The resulting hydrological shapefiles were then merged and converted to a raster with 15 m resolution. All cells containing surface water were assigned an arbitrary travel cost of 1.
217 The cost of terrestrial travel was based on the slope of the terrain. Slope was derived by geoprocessing the DEM in ArcGIS, and ranged from 0 79 degrees. I then used an equation proposed by Bell and Lock (2000) to model the relative cost of traversing terrains of different slopes. They demonstrat e that the relative cost C can be calculated as the ratio of the tangent of the slope angle ( s ) to the tangent of 1 degree of slope: C = tan s / tan 1 (5 1) This results in a nonlinear relationship where steeper slopes become exponentially more difficu lt to travel. Applying this equation to the slope raster yielded values ranging from 0 200.69 . I then scaled this using map algebra to set the cost of travelling across flat land at 5. This terrestrial friction surface was then combined with the hydrologic al friction surface to yield travel costs ranging from 1 205.69 across the peninsula. Under this friction surface the cost of travelling across one 15 m cell of water was 20% of the cost of travelling the same distance across flat land. The friction surfac e was then used to generate four cost surfaces, one originating at each of the four archaeological sites. Cost paths were then traced from each of the eleven quarry clusters to the archaeological sites, resulting in 44 least cost paths. The least cost anal ysis i ndicates that, in terms of cost distance, the Ocala and Lake Panasoffkee quarry clusters are most proximate to Silver Springs (Figure 5 4 ). Least cost paths for Silver Springs indicate that materials from quarry clusters to the north or south would l ikely be transported to Silver Springs by way of the Ocklawaha River. Southern quarry cluster materials (i.e., Lake Panasoffkee, Brooksville, Upper Withlacoochee, Caladesi, Hillsborough River, Turtlecrawl Point, and Peace River) enter
218 near the headwaters o f the Ocklawaha River and travel downstream to the confluence with the Silver River. Northern quarry cluster materials (i.e., Santa Fe, Lower Suwannee, and Gainesville) enter the Ocklawaha River at Orange Springs before moving upstream to the Silver River. Only Ocala quarry cluster materials are likely to have been transported primarily over land. Likewise , least cost paths to Silver Glen Springs primarily follow the Ocklawaha River, entering the St. Johns River north of Lake George and proceeding south fro m there (Figure 5 5 ) . Again, Ocala and Lake Panasoffkee materials would be the least costly to transport . These paths also lend support to the hypothesis th at Silver Springs was a gateway into the S t . J ohn s River valley , as routes from 7 of 11 quarry clust ers pass through or near Silver Springs. Materials from the northernmost quarry clusters l ikely entered the Ocklawaha River downstream of Silver Springs, at Orange Springs, wh ereas Peace River materials would have take n an easterly route. Least cost paths to the Lake Monroe Outlet Midden and Thornhill Lake C omplex are identical (Figure 5 6 ) , and indicate that Lake Panasoffkee and Upper Withlacoochee quarry cluster materials would be least costly. Further, all materials except those originat ing in the Ocala quarry cluster would bypass Silver Springs on their optimal routes to these sites. The least cost analysis indicate s which quarry clusters are the least costly source areas for the transport of lithic materials to the St. Johns River valle y and the likely paths those materials would take. This analysis indicates that if Silver Springs was a gateway, it was not so because of optimal travel routes. Only least cost paths to Silver Glen Springs would pass through Silver Springs for the majority of quarry clusters.
219 Least cost paths to Lake Monroe Outlet Midden and Thornhill Lake largely bypass Silver Springs. Thus, if other lines of evidence suppor t the hypothesis that Silver Springs was a gateway, we can conclude that this was not a result of op timizing transport costs. Th e analysis also yields some predictions that, if confirmed by provenance data , would lend support to the least cost model. Silver Springs and Silver Glen Springs should be dominated by Ocala an d Lake Panasoffkee materials wherea s the Thornhill Lake Complex and Lake Monroe Outlet Midden should be dominated by Lake Panasoffkee and Upper Withlacoochee materials. Also, there should be a fall off curve between Silver and Silver Glen springs for those materials originating from quarry clusters that would pass through Silver Springs on their way to Silver Glen (Ocala, Lake Panasoffkee, Brooksville, Upper Withlacoochee, Caladesi, Hillsborough River, and Turtlecrawl Point). Lithic Provenance Determination Lithic provenance determination wa s accomplished using techniques developed by Upchurch et al. (1982) . Th is quarry cluster method uses a suite of visual characteristics including rock fabric, fossil content, and secondary inclusions (crystal lined voids, sand, porosity, etc.) to discrimi nate between materials from different chert bearing geological for mations (Upchurch et al. 1982). Provenance de termination was conducted by Jon Endonino at Eastern Kentucky University u sing a stereoscopic microscope with magnification of 0 .7 40X and a dire ct LED light source. Analysis began with the identification of geologic formation followed by a specific quarry cluster or the most li kely quarry cluster of origin. Fossil content was the principal criterion for determining provenance as these are most dia gnostic of both
220 formation and qu arry cluster. Rock fabric and secondary inclusions were also considered , particularly when foss il content was non diagnostic. For a detailed discussion of the microfossils , rock fabric , and secondary inclusions diagnostic of the various quarry clusters see Upchurch and colleagues (1982), Austin (1997), Austin and Estabrook (2000), and Endonino (2007). At Silver Springs, provenance was determined for 1,146 pieces of debitage. This is an a pproximately 10% sample of the total assemblage ( n = 10,6 74 ) of debitage. Provenance was also determined for all tools recovered ( n = 91). Similar sample sizes are available from Silver Glen Springs ( n = 1 , 481 debitage, 107 tools), and Lake Monroe Outlet M idden ( n = 1 , 270 debitage, 302 tools), whereas the Thornhill Complex sample is slightly smaller ( n = 418 debitage, 29 tools). The results of proven ance analysis are presented in T able 5 3 through Table 5 10 and F igure 5 7 . No tably , 9 of 11 source areas are represented in the sample, suggesting a wide catchment area for lithic raw materials in the St. Johns River valley . Only the Lower Suwannee and Turtlecrawl Point quarr y clusters are not represented, and only one specimen originated from the Peace River qu arry cluster . E ach site is dominated by one or two quarry clusters, with a range of others found in minor amounts. The assemblages at both Silver Springs and Silver Glen Springs are comprised primarily of Ocala quarry cluster materials. Likewise, the Thorn hill Lake assemblage is primarily composed of Upper Withlacoochee materials. The most abundant lithic material at the Lake Monroe Outlet Midden was not chert, but silicified coral. Silicified coral cannot be sourced using the quarry cluster method, but is most abundant in the Upper Withlacoochee area (Endonino 2007). Amongst the chert, Upper Withlacoochee
221 and Lake Panasoffkee materials are most abundant. The provenance of lithic tools replicates the distribution of source areas in the debitage assemblages . By and large these results bear out the implications of the least cost analysis. The lithic materials at each site are primaril y derived from one of the least costly sources of procurement . There are, howev er, some discrepancies that are not explained by t he model. First, there is a notable deart h of Lake Panasoffkee materials at all sites except the Lake Monroe outlet midden, despite its low cost of acquisition at all sites . Ocala and Peace River materials are also underrepresented at the Thornhill Lake Co mplex and Lake Monroe Outlet midden. Whether this represents a problem with the least cost model , sampling bias, or preferences for stone from certain source areas over others is unclear. If Silver Springs was a gateway into the St. Johns River valley , the distribution of source areas represented in the lithic assemblage there should be mirrored at the other sites. This is generally not the case. Both Silver Springs and Silver Glen are dominated by materials from the Ocala quarry cluster, wh ereas Lake Monroe and Thornhill are dominated by Lake Panasoffkee and Upper Withlacoochee materials, respectively. But, t here is some indication that Silver Springs did serve as a gateway, at least for materials moving into Silver Glen Springs and (perhaps) the north ern portion of the St. Johns River valley . Although Ocala quarry cluster materials dominate at both sites, there is a drop off in the relative frequency at Silver Glen. This is mirrored in the Upper Withlacoochee materials. However , Santa Fe and Gainesville materials are more frequent at Silver Glen, suggesting that these were procured directly rather than being funneled through Silver Springs.
222 Debitage Analysis Lit hic debitage was studied using an individual flake analysis approach in which eac h flake was examined and a series of attributes r ecorded. These attributes i nclude: flake form ( following Sullivan and Rozen 1985), technological categories (biface thinning flake, notching flake, shatter, etc.), stage of reduction (early, middle, late), str iking platform preparation (prepared, unprepared, indeterminate), bulb of percussion type (salient, diffuse, indeterminate) , presence of cortical material, presence of patination, raw material type, presence of thermal alteration, size (in square cm), and weight (g). Singly , many of these attributes are useful in discriminating between different lithic reduction activitie s. However, used in combination multiple lines of evidence provide a more robust interpretation of activities associated with the producti on and maintenance of stone tools. For purposes of this discussion, three metrics are highlighted in T able 5 1 1 through Table 5 1 4 . The thinning index (TI) is calculated as ratio of the weight to size grade. Although debitage weight and size grade themselv es can be used to infer reduction activities, these data can be biased by, for example, a preponderance of broken flakes. One solution would be to weigh and measure only complete flakes, but this would significantly reduce the sample size. The thinning ind ex is a useful alternative. This measure was originally proposed by Johnson (1981) as a proxy for stage of biface reduction (see also Beck et al. 2002). The logic behind it is that this ratio decreases from early to late stage manufacturing as the knapper attempts to minimize thickness while conserving surface area. This same logic can be applied to lithic debitage, as early stage debitage is likely to be thicker for a given planar area than late stage debitage. Further, since this is a ratio of the two va lues, it can be applied to both whole and broken specimens. The amount of cortex adhering to dorsal surface s of
223 debitage is instructive with initial core reduction and early stage tool production resulting in relatively more cortex and later stage reductio n less. Cortex was recorded as the proportion of the dorsal surface a rea on an interval scale from 0 to 5. This corresponds to 0%, 1 25%, 26 50%, 51 75%, 76 99%, and 100%. Technological flake type (e.g., core reduction, biface thinning, notching) is likewi se indicative of reduction activity. A higher proportion of biface thinning flakes in the assemblage is reflective of later stage tool production and maintenance. Taking the aggregate assemblage data, it is clear that, relative to the other sites, reductio n activities at Silver Springs involved larger packages at an earlier stage of reduction . Thinning i nde x and cortex amount are highest here, and the percentage of bi face thinning flakes is lowest. The other sites examined all have a lower thinning index, l ess cortex, and a greater proportion of biface thinning flakes in their assemblages. This is again supportive of the gateway hypothesis, but is also consistent with the proximity of Silver Springs to lithic source areas generally. As noted above, it may be the case that Silver Springs was a gateway for lithic materials moving into the northern part of the study area, particularly to Silver Glen Springs. Relative to Silver Springs, there is a drop off in frequency of both Ocala and Upper Withlacoochee materi als at Silver Glen Springs. Specimens from these quarry clusters at Silver Glen Springs also tend to have less cortex and are more frequently biface thinning flakes. The Silver Springs Gateway? The evidence from these analyses is somewhat conflicting. Over all, it suggests that proximity was the driving factor in lithic acquisition in the St. Johns River valley . There is little to support the idea that lithic raw materials were redistributed from Silver Springs to the Lake Monroe Outlet Midden and Thornhill Lake Complex. However ,
224 patterns do emerge that indicate the movement of lithic materials from at least some source areas thro ugh Silver Springs to reach Silver Glen Springs. This is understandable when considered in the context of large scale social gatherings taking place at Silver Glen, discussed in the previous chapter . However, t he least cost paths indicate that some materials found in appreciable frequency at Silver Glen from the Gainesville and Santa Fe quarry clusters appear to have bypassed Silver Springs along thei r way. This is supported by the provenance determinations inasmuch as materials from these quarry clusters are more abundant at Silver Glen than at Silver Springs . However, the technological data indicate an earlier stage of reduction at Silver Springs, at least for the Santa Fe specimens. There is only one example of Gainesville quarry cluster material at Silver Springs, precluding inference from the technological attributes . If lithic materials did bypass Silver Springs on their way to Silver Glen, they would nevertheless enter into the St. Johns River basin, via the Ock lawaha River , at the ju ncture with Orange Spring, a smaller third magnitude spring downstream from Silver. The pattern that emerges is one of distinct spheres of circulation ( Figure 5 8 ), whereby lithic materials entering the northern portion of the middle St. Johns River valley (i.e., the vicinity of Lake George) were obtained primarily from the northern quarry clusters and moved by way of Silver, Orange, and perhaps other springs. Lithic materials entering the southern portion of the middle St. Johns (i.e., the vicinity of lake s Monroe and Jesup) derive primarily from southern quarry clusters and entered via the numerous water bodies of the Central Lakes District. It is notable that the se southern least cost paths likewise enter the St. Johns River valley by way of springs
225 Hills borough River chert via Wekiwa Spring and Lake Panasoffkee and Upper Withlacoochee chert via a concentration of a dozen springs in the Seminole National Forest. How these patterns of exchange and interaction manifested at intervening locales for example at Alexander, Blue, or DeLeon springs is unclear but could be elucidated with similar analyses of these assemblages. Rather than Silver Springs operating as the gateway into the St. Johns River valley , it appears that it may have been one of many springs tha t facilitated interaction and movement. If the patterns described above are murky, and the evidence contradictory, this is partially because of the lack of chronological control in the samples studied. The lithic assemblages from the Lake Monroe Outlet Mid den and Thornhill Lake Complex were recovered during controlled excavations of relatively discrete Mount Taylor aged deposits. The lithic assemblages from both Silver and Silver Glen , in contrast, were derived from reconnaissance survey . Although the major ity of the lithic artifacts recovered at these sites is likely of Mount Taylor age, the assemblages nevertheless collapse millennia of activity proximate to the springs. If other springs beyond Silver functioned as conduits or gateways for movement into the St. Johns, it remains to be determined if they exhibit similar material records (i.e., a dense lithic assemblage and dearth of pottery, shell, and evidence of intensive habitation). Neither Orange, Wekiwa, nor the springs of the Seminole State Forest have been subject to intensive archaeological investigation that might address this. However, dense lithic assemblages adjacent to springs ha ve been documented elsewhere, ou tside the St. Johns River valley . It is worth discussing these briefly to consider how they articulate with the springs examined above.
226 Beyond the St. Johns In 2014, t he LSA conducted a Cultural Resources Assessment Survey at Otter Springs Park , located in the Suwannee River valley of northwest peninsular Florida. This was done to aid the Suwannee River Water Management Di strict in advance of infrastructural repair and maintenance dredging. Otter Springs Park is a 636 acre park and campground located in wes tern Gilchrist County, near the bo rd er with Dixie County (Figure 5 9 ). Otter is a second magnitude spring consisting of two separate spring po ols connected by a 75 m artificial channel that was excavated in the early 1960s. Otter Spring #1 is the main (ori ginal) spring, and has a pool measur ing approximately 40 m in diameter with a concrete retaining wall on the southern margin. Otter Spring #2 is loca ted approximately 45 m northeast of the ma in spring. It has a 21 m diameter pool surrounded by a sand cemen t retaining wall. It flows southwest into the pool of the main spring, through the artificial channel. Prior to the excavation of this channel Otter Springs #2 did not flow onto the surface. It is visible on historic aerial photographs as a water bearing s ink or karst window. The spring run flows generally westward for approximately 1.3 km (0.8 mi), where it joins the Suwannee River. T he Otter Springs site (8GI12) had previously been recorded as encompass ing both spring pools and the surrounding land, measu ring approximately 4.8 acres in extent. This site was originally recorded in 1958 by John Goggin, on the basis of his examination of a private collection of material recovered from the spring. Li ttle detail is given in Florida Master Site File records, but the co llection consisted of potsherds and lithic tools. In 1974, two divers reported the recovery of potsherds and faunal remains from the cave system of Otter Springs, approximately 90 m back from the vent at a depth of 17 m .
227 In total, 1 9 STPs were excav ated during the course of th e LSA survey (Figure 5 10 ) . Subsurface disturbance was found to be extensive, with modern fill present over much of the project area. However, cultural materials were recovered from undisturbed contexts in all but one shovel tes t. As at Silver Springs, lithic artifacts we re by far the most frequent ly recovered material (Table 5 1 5 ) , comprising 8 7.1 % of the total ( n = 474 out of 544 total). Pottery w as the second most frequent (6.6 %; n = 36), followed in abundance by vertebrate fauna (6. 3 %; n = 34). These materials we re spread across the project area , but artifact density was greatest proximate to and south of the spring . Artifact density was lower along the eastern and northern periphery of the project area. The artifact invento ry is dominated by lithic debitage ( n = 467). The bulk (73%) of this debitage consists of chert, although silicified coral is also well represented. The flakes are generally small (mean weight = 0.57 g) and lacking in dorsal cortex. Chert debitage is, on a verage, larger than silicified coral debitage, with a mean weight of 0.65 g for chert versus 0.33 g for coral. We lack the contextual control to infer the full range of lithic reduction activities that occurred at Otter Springs , but the diminutive size of individual specimens in the assemblage is indicative of late stage production and/or retouch. Much like Silver Springs, t he bulk of the pottery assemblage consists of plain or eroded sand tempered wares (72.2%). These have a wide spatial and temporal bread th, diminishing their utility as diagnostic artifacts. Diagnostic types recovered include Deptford Check Stamped, St. Johns plain or eroded, and Lochloosa Punctated. Both Deptford Check Stamped and St. Johns plain indicate a Woodland period
228 occupation (ca. 2500 1800 B.P.), w hile the Lochloosa Punctated pottery indicates post Woodland utilization of the site. In sum, activities at Otter Springs are similar in many respects to those documented at Silver Springs and were largely centered on lithic reduction. T he density of lithic artifacts at Otter Springs (99.8 artifacts/m 3 ) is comparable to that at Silver Springs (91.9 artifacts/m 3 ) . Although pottery was considerably more dense (7.6 artifacts/ m 3 at Otter Springs vs 1.1 artifacts/m 3 at Silver Springs), it is s till less than all but one of the large springs of the St. Johns River valley (cf., Table 5 2 ) indicating a lack of intensive habitation. This inference is supported by the lack of midden deposits and mounded features at Otter Springs. Likewise, the ratio of lithic to ceramic artifacts (13.17) is considerably higher here than at St. Johns River valley springs. Where Otter Springs differs from Silver Spring is that the lithic assemblage is consistent with late stage reduction and tool maintenance. This is so mewhat confounding given the proximity of Otter Springs to chert outcrops. The survey at Otter Springs was not extensive, but limited to the area immediately surr ounding the spring. So caution is warranted when drawing conclusions from these samples . Dense midden deposits may be present, but slightly f a rther removed from the spring. However, an intensive survey at Fanning Springs, located just 7 km downstream from Otter Springs , recovered similar results. Fanning is a first magnitude spring consisting of a single main vent and numerous sand boils and seep springs . The pool measures some 60 m in diameter, 6 m deep , and is connected to the Suwannee River by a short (~ 140 m) spring run (Scott et al. 2004:212). Two intensive surveys (Bland and Chance 2000; Dicki nson and Wayne 2003) resulted in the
229 excavation of 507 shovel tests surrounding the spring and run . Although the overall density of artifacts was lower than at Otter Springs, lithic artifacts ( n = 1,464) were similarly far more frequent than pottery ( n = 8 8) and consistent with late stage reduction. The ratio of lithic artifacts to pottery (16.63) was even higher than at Otter Springs . Likewise, the density of pottery (1.06 artifacts/ m 3 ) was lower and more consistent with that documented at Silver Springs. It is tempting to infer from these surveys that springs in the Suwannee River valley frequently feature dense lithic assemblages absent midden deposits, pottery, or other indicators of intensive habitation. However, this is a small sample of the scores of springs along the Suwannee River. Indeed, at Manatee Springs, 12 km downstream from Fanning Springs , Bullen (1953) documented an intensively occupied village dating to the Early Weeden Island period. This underscores the observation that, as in the St. Joh ns River valley , a diversity of activities took place at springs in the western part of the peninsula . The inferences derived from a single spring do not necessarily apply to other springs, locally or regionally. Nevertheless, the lithic assemblages at Otter and Fanning springs are analogous to those at Silver Springs, if not as expansive or dense. This provides a clue to the importance of these deposits, which I discuss in the concluding section. Summary and Discussion As discussed at the outset, the St . Johns River valley is bereft of outcrops of stone suitable for the manufacture of flaked stone tools. Lit hic raw materials were transported to the St. Johns River valley from source areas in northwest peninsular Florida. Recent archaeological survey at S ilver Springs recovered a dense lithic assemblage absent pottery, fauna, shell deposits, or other evidence of intensive
230 habitation. These results, combined with the proximity of Silver Springs to lithic source areas, led to the hypothesis that Silver Sprin gs was a gateway or conduit for the movement of stone into the St. Johns River valley . This hypothesis was evaluated with least cost modeling, lithic provenance determinations, and technological analysis of lithic debitage from Silver Springs and three sit es in the St. Johns River valley proper. These analyses indicate that if Silver Springs was a gateway, this was primarily for movement of lithic materials to Silver Glen Springs. The other sites under consideration Lake Monroe Outlet Midden and Thornhill L ake Complex lie at the southern end of the middle St. Johns and have lithic assemblages derived largely from different source areas than Silver or Silver Glen Springs. If Silver Springs was a gateway for movement to Silver Glen and the pan regional gatheri ngs that took place there, the absence of shell at Silver Springs is nevertheless confounding. However, similarly dense lithic deposits have been documented at springs further west, for example at Otter Springs in the Suwannee River valley. This, I argue , provides a key to understanding the deposits at Silver Springs. As people , bearing stone, moved across the Florida peninsula towards the gatherings at Silver Glen, springs were a touchstone, a place at once unfamiliar but reminiscent of home. Through the process of flake removal people l eft behind fragments of this stone along its journey. These fragments of stone indexed not only their place of origin the locality and geologic formation from which they were acquired but the events of fragmentation and sto ppage at spring side waypoints along their route. T he
231 fragmentation and deposition of stone effectively enchained places and people across the Florida peninsula , in the context of social gatherings like those at Silver Glen Springs. Likewise, as this pract ice was repeated at springs along the journey from the Gulf coastal lowlands, across interior highlands and scrub, and into the St. Johns River valley people effectively re created familiar places in an unfamiliar landscape through the fragmentation and de position of stone from the homeland . In this way springs the communities across the Florida peninsula.
232 Table 5 1 . Inventory of p re Columbian c ultural m aterials r ecovered from STPs excavated at Silver Springs (8MR93 and 8MR108 2 ) Object Class n Weight (g) Lithic d ebitage 10,674 14,646.5 Lithic t ool s 102 8,775.6 Orange, plain/eroded sherd 5 22.0 Pasco, plain/eroded sherd 14 26.5 S t. Johns , plain /eroded she rd 37 64.3 St. Johns, checkstamped sherd 6 60.9 St. Johns, incised sherd 5 10.6 Sand tempered, plain/eroded sherd 63 213.5 Sand tempered, checkstamped sherd 4 14.7 Vertebrate fauna 76 61.4 Marine s hell 46 48.7 Freshwater s hell 1 3 6. 3 Historic 90 1,389.0 Total 11,134 25,339.7 Table 5 2 . Frequency of l ithic a rtifacts at s elected a rchaeological s ites a djacent to s prings in the St. Johns River valley . Sample v olumes and a rtifact d ensity are a pproximate (Denson et al. 1995; Dickinson and Wayne 1994; Randall et al. 2011; Sassaman 2003a; Willis 1995) Spring Name/Site Number Lithic artifact s (n) Pottery (n) Estimated sample volume (m 3 ) Lithic density (artifacts/m 3 ) Pottery density (artifacts/m 3 ) Lithic to pottery ratio Alexander Spring (8LA71) 41 100 4.3 9.6 23.5 0.41 Blue Spring (8VO43) 46 4,047 21.0 2.2 192.7 0.01 DeLeon Spring (8VO30) 78 82 11.7 6.7 7.0 0.95 Salt Springs (8MR2322) 2,025 1,874 137.3 14.7 13.6 1.08 Silver Glen Springs ( 8MR123 ) 1,395 762 58.0 24.1 13.1 1.83 Silver Springs (8MR93) 10,776 134 117.3 91.9 1.1 80.42
233 Table 5 3 . P rovenance of lithic debitage from Silver Springs (8MR93) Table 5 4 . P rovenance of lithic debitage from Silver Glen Springs (8MR123) Quarry Cluster Relative Cost n % Weight (g) % Santa Fe 1.94 43 3.8% 49.1 3.7% Gainesville 1.86 1 0.1% 0.6 0.0% Lower Suwannee 2.38 0 0.0% 0.0 0.0% Ocala 1.00 868 75.7% 1053.9 80.4% Lake Panasoffkee 1.19 3 0.3% 3.1 0.2% Brooksville 2.25 0 0.0% 0.0 0.0% Upper Withlacoochee 1.72 141 12.3% 146.7 11.2% Caladesi 2.92 0 0.0% 0.0 0.0% Hillsborough River 2.24 1 0.1% 0.1 0.0% Turtlecrawl Point 3.57 0 0.0% 0.0 0.0% Peace River 2.46 0 0.0% 0.0 0.0% Silic. Coral -25 2.2% 29.8 2.3% Indet. -64 5.6% 27.5 2.1% Total -1 , 146 100.0% 1 , 310.6 100.0% Quarry Cluster Relative Cost n % Weight (g) % Santa Fe 1.13 325 21.9% 157.9 19.4% Gainesville 1.09 64 4.3% 54.9 6.8% Lower Suwannee 1.38 0 0.0% 0.0 0.0% Ocala 1.00 818 55.2% 455.3 56.0% Lake Panasoffkee 1.05 0 0.0% 0.0 0.0% Brooksville 1.66 1 0.1% 0.1 0.0% Upper Withlacoochee 1.35 37 2.5% 45.4 5.6% Caladesi 2.03 0 0.0% 0.0 0.0% Hillsborough River 1.65 4 0.3% 2.1 0.3% Turtlecrawl Point 2.41 0 0.0% 0.0 0.0% Peace River 1.59 1 0.1% 0.3 0.0% Silic. Coral -171 11.5% 75.2 9.2% Indet. -60 4.1% 21.5 2.6% Total -1 , 481 100.0% 812.5 100.0%
234 Table 5 5 . P rovenance of lithic debitage from Lake Monroe Outlet Midden (8VO53) T able 5 6 . P rovenance of lithic debitage from the T hornhill Lake Complex (8VO58 60) Quarry Cluster Relative Cost n % Weight (g) % Santa Fe 1.50 0 0.0% 0.0 0.0% Gainesville 1.45 11 0.9% 27.4 3.1% Lower Suwannee 1.76 0 0.0% 0.0 0.0% Ocala 1.36 38 3.0% 28.4 3.2% Lake Panasoffkee 1.00 258 20.3% 242.2 27.1% Brooksville 1.64 38 3.0% 20.3 2.3% Upper Withlacoochee 1.31 162 12.8% 138.4 15.5% Caladesi 1.99 14 1.1% 23.8 2.7% Hillsborough River 1.56 10 0.8% 7.3 0.8% Turtlecrawl Point 2.36 0 0.0% 0.0 0.0% Peace River 1.40 0 0.0% 0.0 0.0% Silic. Coral -664 52.3% 391.5 43.9% Indeterminate -75 5.9% 13.3 1.5% Total -1 , 270 100.0% 892.4 100.0% Quarry Cluster Relative Cost n % Weight (g) % Santa Fe 1.46 0 0.0% 0.0 0.0% Gainesville 1.42 0 0.0% 0.0 0.0% Lower Suwannee 1.71 0 0.0% 0.0 0.0% Ocala 1.34 0 0.0% 0.0 0.0% Lake Panasoffkee 1.00 1 0.2% 0.2 0.1% Brooksville 1.59 19 4.5% 4.1 2.3% Upper Withlacoochee 1.29 282 67.5% 130.4 72.4% Caladesi 1.92 0 0.0% 0.0 0.0% Hillsborough River 1.52 33 7.9% 18.3 10.2% Turtlecrawl Point 2.26 0 0.0% 0.0 0.0% Peace River 1.26 0 0.0% 0.0 0.0% Silic. Coral -31 7.4% 9.4 5.2% Indet. -52 12.4% 17.9 9.9% Total -418 100.0% 180.2 100.0%
235 Table 5 7 . P rovenance of lithic tools from Silver Springs (8MR93) Quarry Cluster Bifaces % Other Tools % Santa Fe 2 3.7% 0 0.0% Gainesville 2 3.7% 1 2.7% L ower Suwannee 0 0.0% 0 0.0% Ocala 42 77.8% 25 67.6% Lake Panasoffkee 0 0.0% 0 0.0% Brooksville 0 0.0% 0 0.0% Upper Withlacoochee 2 3.7% 4 10.8% Caladesi 0 0.0% 0 0.0% Hillsborough River 1 1.9% 0 0.0% Turtlecrawl Point 0 0.0% 0 0.0% Peace River 0 0.0% 0 0.0% Silic. Coral 1 1.9% 1 2.7% Indet. 4 7.4% 6 16.2% Total 54 100.0% 37 100.0% Table 5 8 . P rovenance of lithic tools from Silver Glen Springs (8MR123) Quarry Cluster Bifaces % Other Tools % Santa Fe 9 22.5% 16 23.9% Gainesville 5 12.5% 4 6.0% Lower Suwannee 0 0.0% 0 0.0% Ocala 11 27.5% 31 46.3% Lake Panasoffkee 0 0.0% 0 0.0% Brooksville 0 0.0% 0 0.0% Upper Withlacoochee 2 5.0% 1 1.5% Caladesi 0 0.0% 0 0.0% Hillsborough River 1 2.5% 3 4.5% Turtlecrawl Point 0 0.0% 0 0.0% Peace River 0 0.0% 0 0.0% Silic. Coral 9 22.5% 12 17.9% Indet. 3 7.5% 0 0.0% Total 40 100.0% 67 100.0%
236 Table 5 9 . P rovenance of lithic tools from Lake Monroe Outlet Midden (8VO53) Quarry Cluster Bifaces % Other Tools % Santa Fe 0 0.0% 0 0.0% Gainesville 1 2.8% 4 1.5% Lower Suwannee 0 0.0% 0 0.0% Ocala 6 16.7% 12 4.5% Lake Panasoffkee 10 27.8% 76 28.6% Brooksville 1 2.8% 5 1.9% Upper Withlacoochee 0 0.0% 29 10.9% Caladesi 0 0.0% 3 1.1% Hillsborough River 1 2.8% 6 2.3% Turtlecrawl Point 0 0.0% 0 0.0% Peace River 0 0.0% 0 0.0% Silic. Coral 14 38.9% 123 46.2% Indet. 3 8.3% 8 3.0% Total 36 100.0% 266 100.0% Table 5 1 0 . P rovenance of lithic tools from the T hornhill Lake Complex (8VO58 60) Quarry Cluster Bifaces % Other Tools % Santa Fe 0 0.0% 0 0.0% Gainesville 2 8.7% 0 0.0% Lower Suwannee 0 0.0% 0 0.0% Ocala 0 0.0% 0 0.0% Lake Panasoffkee 3 13.0% 0 0.0% Brooksville 3 13.0% 1 16.7% Upper Withlacoochee 4 17.4% 1 16.7% Caladesi 0 0.0% 0 0.0% Hillsborough River 0 0.0% 2 33.3% Turtlecrawl Point 0 0.0% 0 0.0% Peace River 0 0.0% 0 0.0% Silic. Coral 11 47.8% 1 16.7% Indet. 0 0.0% 1 16.7% Total 23 100.0% 6 100.0%
237 Table 5 1 1 . Summary r esults of debitage analysis from Silver Springs (8MR93) Quarry Cluster n Thinning Index Mean Cortex % Biface Thinning Flakes Santa Fe 43 0.66 0.3 44.2% Gainesville 1 0.30 0.0 100.0% Lower Suwannee 0 ---Ocala 868 0.70 0.3 50.6% Lake Panasoffkee 3 0.48 0.7 100.0% Brooksville 0 ---Upper Withlacoochee 141 0.63 0.2 52.5% Caladesi 0 ---Hillsborough River 1 0.07 0.0 100.0% Turtlecrawl Point 0 ---Peace River 0 ---Silic ified Coral 25 0.68 0.3 44.0% Indeterminate Chert 64 0.28 0.2 34.4% Total 1 , 146 0.67 0.3 49.7% Table 5 1 2 . Summary r esults of debitage analysis from Silver Glen Springs (8MR123) Quarry Cluster n Thinning Index Mean Cortex % Biface Thinning Flakes Santa Fe 325 0.33 0.2 65.2% Gainesville 64 0.48 0.2 71.9% Lower Suwannee 0 ---Ocala 818 0.37 0.2 67.8% Lake Panasoffkee 0 ---Brooksville 1 0.10 0.0 0.0% Upper Withlacoochee 37 0.69 0.1 83.8% Caladesi 0 ---Hillsborough River 4 0.30 0.0 100.0% Turtlecrawl Point 0 ---Peace River 1 0.20 1.0 100.0% Silic ified Coral 171 0.29 0.1 67.8% Indeterminate Chert 60 0.28 0.2 41.7% Total 1 , 481 0.36 0.2 66.8%
238 Table 5 1 3 . Summary r esults of debitage analysis from the Lake Monroe Outlet Midden (8VO5 3) Quarry Cluster n Thinning Index Mean Cortex % Biface Thinning Flakes * Santa Fe 0 ---Gainesville 11 0.98 0.0 -Lower Suwannee 0 ---Ocala 38 0.43 0.0 -Lake Panasoffkee 258 0.54 0.0 -Brooksville 38 0.34 0.1 -Upper Withlacoochee 162 0.55 0.0 -Caladesi 14 1.16 0.0 -Hillsborough River 10 0.44 0.0 -Turtlecrawl Point 0 ---Peace River 0 ---Silic ified Coral 664 0.37 0.0 -Indeterminate Chert 75 0.15 0.0 --Total 1 , 270 0.44 0.0 -* D ata on frequency of biface thinning flakes not available . Table 5 1 4 . Summary r esults of debitage analysis from the Thornhill Lake Complex (8VO58 60) Quarry Cluster n Thinning Index Mean Cortex % Biface Thinning Flakes Santa Fe 0 ---Gainesville 0 ---Lower Suwannee 0 ---Ocala 0 ---Lake Panasoffkee 1 0.13 0.0 100.0% Brooksville 19 0.15 0.0 89.5% Upper Withlacoochee 282 0.32 0.1 72.7% Caladesi 0 ---Hillsborough River 33 0.35 0.2 84.8% Turtlecrawl Point 0 ---Peace River 0 ---Silic ified Coral 31 0.25 0.1 71.0% Indeterminate Chert 52 0.25 0.1 48.1% Total 418 0.30 0.1 71.3%
239 Table 5 1 5 . Inventory of Cultural Materials Recovered from STPs excavated at Otter Springs (8GI12) Object Class n wt. (g) Biface, Chert 2 5.5 Hafted Biface, Chert 2 8.5 Core, Chert 1 15.4 Modified flake, Chert 2 20.7 Debitage, Chert 342 223.3 Debitage, Silicified Coral 125 42.1 Deptford Check Stamped sherd 1 29.8 Lochloosa Punctated sherd 6 44.5 St. Johns, eroded sherd 1 2.2 Sand tempered, plain sherd 25 112.8 Sand tempered, stamped sherd 2 8.4 Sand tempered, eroded sherd 1 9.3 Vertebrate fauna 34 13.6 Total 5 44 5 36.1
240 Figure 5 1 . Location of spring vents at Silver Springs. Image courtesy of the St. Johns River Water Management Distric t (http://floridaswater.com/springs/images/silver_ventlocation_aerial_lrg.jpg )
241 Figure 5 2 . Area surveyed by the LSA and previously recorded archaeological sites, Silver Springs , Florida
242 Figure 5 3 . Results of the archaeological survey conducted at Silver Springs, Florida
243 Figure 5 4 . Results of the least cost an alysis from lithic quarry clusters to Silver Springs (8MR93)
244 Figure 5 5 . Results of the least cost analysis from lithic quarry clusters to Silver Glen Springs (8MR123)
245 Figure 5 6 . Results of the least cost analysis from lithic quarry clusters to th e Lake Monroe Outlet Midden (8VO53) and the Thornhil Lake Complex (8VO58 60)
246 Figure 5 7 . Comparison of lithic source use for Silver Springs (8MR93), Silver Glen Springs (8MR123), Lake Monroe Outlet Midden (8VO53), and the Thornhill Lake Complex (8VO58 within each site assemblage
247 Figure 5 8 . Aggregated least cost paths, limited to those quarr y clusters comprising greater than 5.0% of the lithic assemblage for a given site
248 Figure 5 9 . location of the LSA Otter Springs survey
249 Figure 5 1 0 . Shovel test pits excavated at Otter Springs . All STPS were positive
250 CHAPTER 6 SPRINGS ETERNAL Hope springs eternal in the hum an breast: Man never is, but always to be blest: The soul, uneasy, and confined from home, Rests and expatiates on a life to come. Sees God in clouds, or hears him in the wind; His soul, proud science never taught to stray Far as the solar walk, or milky way; Yet simple Nature to his hope has given, Behind the cloud Some happier island in the watery waste, Where slaves once more their nativ e land behold, No fiends torment, no Christians thirst for gold. Alexander Pope An Essay on Man S prings have been significant places on the Florida landscape since the first human colonization of the peninsula, and they continue to be today. Several springs were important inland ports and conduits for travel prior to the widespread construction of railway lines and the advent of automobile travel (Figure 6 1 ; see also Berson 20 11 ), and many were developed as health spas and p r ivate tourist attractions in the nineteenth and twentieth centuries (Figure 6 2 , Figure 6 3, and Figure 6 4 ). For modern Floridians , s prings are popular recreation areas whose waters provide respite from the oppressive heat of Flo rida summers. And today private campgrounds, U.S. Forest Service recreation areas, and state parks surround many springs, affording ample recreational opportunities. Springs are as well, the subject of paintings, photographs, tr avelogues, poems, and literary works (e.g., Bartram 1996; Carr 1996; Douglas 1967; Earl 2009; Lanier 1876; Le Conte, 1861; Rawlings 1938; Tolbert 2010).
251 Annually more than two million people visit the Florida State Parks that feature springs, generating se veral million dollars in revenue for the state (Florida Department of Environmental Protection [DEP] 2014 ). In addition to the revenue generated by park admission fees, springs provide jobs and financial stimulus to surrounding areas. Estimates of the econ omic benefit to the communi ties around just four springs Homosassa, Ichetuckn ee, Volusia Blue, and Wakulla range from $10 million to $23 million each (Bonn and Bell 2003), and Silver Springs alone contributes in excess of $60 million annually to the local economy (Bonn 2004). Ecologically, springs are seen as uniquely Floridian habitats in need of protection from the impacts of development and over use (Florida Springs Initiative 2007; Pittman 2012 a ). Springs are unique hydrological systems, with exceptiona l water clarity and near constant temperature and chemistry. They were critical to the development of systems ecology, through the work of Howard Odum beginning in the 1950s (Knight 2015; Odum 1957a, 1957b). As steady state systems, springs continue to be 2014:2010). The economic, cultural, and scientific value of springs hinges on the physical properties of spring water its clarity, purity, and aesthetic quality and health of spring ecosystems. These are related , as negative impacts to spring ecosystem health have detrimental effects o n their aesthetic appeal. There exist two primary threats to springs ecological and aesthetic integrity: the proliferation of filamento us algae and decline of spring flow. Reports of dense algal mats covering spring bottoms and floating on the surface began in the mid 1980s and have become increasingly common ( Florida
252 Springs Task Force 2000 ; Stevenson et al. 2007 ). These algal mats choke out springs flora and fauna and negatively impact their aesthetic appeal. According to reporter Doug Struck: t he algae is a black fuzz that coats the bottom and sucks up all the light. The luxurious waving eel grass is pretty patchy, the schools of fish are mostly missing. The Wakulla Springs of my childhood swimming hole, the Wakulla Springs of jeweled luminescen ce, now exists only in memories (quoted in Shockman 2015). Algal proliferation is largely attributed to increases in nitrate concentrations in s pring waters. Nitrate, a form of nitrogen, is the most commonly elevated pollutant in the Floridan Aquifer and is primarily introduced from fertilizers used in agricultural and residential applications (Brown et al. 2008; Jones et al. 1996; Katz 2004; Knig ht 2015; Phelps 2004). In 1950, nitrate concentration in Florida springs generally ranged from 0.05 0.1 mg/L. In 2004, the average nitrate concentration in Florida springs was 1.0 mg/L and values as high as 7.5 mg/L were recorded (Heffernan et al. 20 10; St rong 2004). The Florida DEP considers many large springs to be legally impaired because of elevated nitrate (Knight 2015:77) . Whether elevated nitrate causes algal proliferation is debated (see below and Heffernan et al. 2010), but regardless , nitrate has other detrimental effects, including potential adverse health effects for humans and springs fauna, and eutrophication of downstream aquatic environments (Heffernan et al. 2010; Knight 2015). In addition to water quality impairment, many springs have also witnessed reduced flow since 1960 (Weber et al. 2006; William 2006). Knight (2015:57) calculated the average reduction of spring flow in each of four water management districts (WMD) from 1930 to 2009. These range from a 16% average reduction in the Northw est Florida WMD to 48% in the Suwannee River WMD. Flow reductions have negative effects on
25 3 springs flora and fauna, as spring flow is correlated with primary production . Declining spring flow thus adversely impacts wildlife habitat and reduces food availability (Knight 2015:319 333). Some portion of spring flow reduction is likely caused by climatic fluctuations ( i.e., long term reduction in precipitation [M unch et al. 2006 ] ), but groundwater extraction from the Floridan Aquifer system is also a prime factor . These threats, and the perceived cultural, economic, and ecological importance of springs, led the state legislature to provide over $24 millio n and protection from 2001 to 2011, through the auspices of the Florida Springs Initiative. Althou gh this program was terminated in 2011, more recent financial allocations have been made to springs restoration projects by the Florida DEP and state WMDs. The 2015 2016 state budget includes $40 million for 26 Florida DEP springs restoration projects, the highest amount ever allocated (FDEP 2015; Florida Trend 2015). Matching funds from local governments and WMDs bring the total investment in springs restoration to $82 million. The vast majority of the projects funded by this budget involve the constructio n or improvement of wastewater treatment facilities to reduce nitrate loading in springs. This is in line with most m odern prote ction and remediation strategies that are largely concerned with chemical pollution and degradation of water quality from develo pment and agricultural practices and excess groundwater pumping ( Florida Department of Community Affairs and Florida Department of Environmental Protection 2002; Florida Springs Task Force 2000). There is thus conce rted effort to conserve springs, not onl y through public funding but outreach programs, media editorials, art installations, and museum exhibits (e.g., Gainesville Sun and Ocala Star Banner 2013; Knight 2008, 2012, 2013, 2014,
254 2015; Moran 2013; Neeley 2015; Schofield 2013; Tampa Bay Times 2012) . B ut there are barriers and challenges as well. Archaeology, with its long term perspective, should have ample data to provide policy makers, conservationists, and researchers interested in understanding the context of changes observed at springs and the c onsequences of human spring interaction. However, the perceived disjuncture between modern experience of place and that of native Floridians centuries or millennia ago hamstrings attempts at dialogue (Sassaman 2012) , a s does the penchant of archaeologists to equivocate and emphasize the nuance and complexity of individual cases. In this chapter I demonstrate the relevance of what has been presented thus far, of archaeological knowledge, to issues facing springs today. I do so, first, by examining how the pa st is used, or constructed, in accounts of Florida springs as both a mythical Eden that modern humans have fallen from and as target for springs restoration. Then, I inter connections between humans and things in the past and present. There is long term continuity, or at least analogy, in the entanglement of humans and springs that bridges the chasm between ancient and modern experience . As a result, insights from the past a re applicable to springs conservation today, despite appearances to the contrary. I then close with some concluding remarks and indicate directions for future research. The Past in Springs Conservation Narratives The p ast is used in two ways, on two horizons, in narratives of springs conservation. The more recent past, the historical past of the last century or so, is generally used as baseline for the physical condition of springs in a pristine or untouched state .
255 Though not as pristine as i n the days when Hollywood icon Lloyd Bridges Springs is the main draw, and work is underway to reverse decades of inatte ntion that have fed its decline (Thompson 2013 a :2). A centu clear springs drew presidents and millionaires and tourists galore who sought to cure their ailments by bathing in the healing cascades. Now the springs tell the story of a hidden sickness, one that lies deep within the earth (Tampa Ba y Times 2012) . This appeal to the historic grandeur of springs often suggests that springs were pristine prior to the late twentieth century. Jim Stevenson wrote of his childhood visits to springs: that was the 1950s, a time when our springs were still pr istine with strong flow and In addition to a baseline for measuring change, t he recent past and a pristine state is used as a target require returning the spri ngs back to their flows and nitrate levels of the 1950s and A nd a s an NPR story reported : ater this year, Florida's park service will take over Silver Springs and begin working to res tore it to a more natural state . In this effort to restore he material remains of modern humans are considered a 2013) . Where possible, these are removed and remediated, often at considerable cost: time zoo were to be removed sc uttled in favor of a return to a more natural setting (Thompson 2013a). The second horizon in which the past is used is a more distant one, and more rarely invoked. This is the pre historic past of Native Americans. Where it is mentioned at all, the ca. 13,000 years of pre Columbian interaction with springs is largely glossed
256 over. In his recent book Silenced Springs , Robert Knight , perhaps the most vocal and respected scientist advocating for spr ings conservation , summarizes thirteen millennia in less than 300 words (Knight 2015:8 9) . Likewise, the Florida DEP website states: a rchaeological evidence indicates that people have been attracted to made the perfect home for Native Floridians who used the m as a source of water and As the last of the Ice Age came to a close in Florida, many environmental changes were occurring .. .As these drastic changes were taking place, Florida's human inhabitan ts learned to adapt. (FDEP 2014) The next sentence continues: L ater arrivals to Florida, Ponce de Leon, John and William Bartram and other explorers, were drawn to the subterranean discharges of freshwater scattered across central and northern Florida. The narrative jumps f rom Paleoindian to Spanish explorer , glossing over the millennia between. Similarly collapsing history , a CNN report d escrib ed rec overed during dredge operations at Chassahowitzka Springs, with artifacts that represent every period of human occupation in Florida coverage of the Laboratory of Southeastern Archeology (LSA) survey of Silver Springs the renowned site's crystal clear w aters were a draw for visitors long before the first modern tourists arrived 2013b). I do not mean simply to criticize that non specialists are presenting superficial summaries of springs archaeology. Certainly, archaeologists are guilty of this as well, arguing (as I have done) that springs have been important as long as humans have into tractable environmental [and archaeological] narratives, often a necess ary step for
257 narrative of eternity, a history of springs existing largely unchanged : c lear, temperate water flows through limestone channels beneath North Central Flor ida, rising to the surface to form scores of springs that sustained early natives, amazed European explorers and delight us today but this may only be a memory in a generation or two (Gainesville Sun and Ocala Star Banner 2013) . The net effect of these two uses of the past is to recapitulate the perceived rupture, or revo ancient past of springs has been changelessness . Often, the ancient past i s romanticized as an E denic Florida, populated by noble natives living in harmony with nature ille glinting water, and eventually came to re g but t his serves only to remind us of h o w far modern Floridians have fallen. Only the recent past the past documented by written records and the memori es of living people is taken to bear relevance to the threats imperiling springs. So, t he best we can do is look back, and through some process of time reversal, re store springs to their remembered state, before the pace of change quickened. But, the geolo gical origin of springs, and the millennia of human spring interaction described thus far belie any notion of eternal sameness. In the following, I use entanglement theory as a frame for delineating these interactions to show the various interdependencies of humans and springs both clarify our understanding of springs dynamics and provide a place for the archaeological knowledge in modern discourse.
258 Entanglement Entangl e ment theory has recently been developed by Hodder (2011a, 2011b, 2012), and the following summary is drawn from his writings. This is an approach to describe th e complex networks, mixes, and engagement thing interdependence (Hodder 2011a:162). and Old High German ( Heidegger 1971; Hodder 2011a:157, 2011b:177; Olsen 2003). The emphasis is on the dual capacity of things to both appear to human s a s discrete, bounded entities, while simultaneously depending on other things along chains of interdependence in which many other actors are involved human, institutional, legalistic, bureaucratic, 1 1a :157). Entanglement theory focuses on the se dependences and dependencies between humans and things , along multiple axes (Hodder 2011a, 2011b, 2012) . Humans depend on things, for instance, as food, shelter, clothing, water, and the tools used to procure and manufacture these. But much recent literature in anthropology and social theory has revealed the ways that humans depend on things beyond subsistence and technology (e.g., DeMarrais et al. 2004; Gosden 2005; Ingold 2011, 2012, 2013; Meskell 2005; Miller 2 005; Olsen 2010; Olsen et al. 2012; Skibo and Schiffer 2008; Webmoor and Witmore 2008 ) . That is, things are increasingly recognized to facilitate social relations through exchange and interaction, personal and collective identities, negotiations of power a nd prestige, meanings, ideologies, and, indeed, embodiment. Li kewise, all things depend on other things through relations of interdependence. Human crafted things rely, for example, on the tools used to make or repair them, and things that facilitate their use. So a wooden spoon depends on the tree from which the
259 wood was harvested, the axe used to fell the tree, and the woodworking tools used to shape and smooth the wood into a spoon, but also the food that it stirs and scoops , the pot that contains that f ood , and the stove that provides heat . This thing thing dependence may be obvious for human manufactured things, but is also true of natural things , which likewise depend on other things for their existence . T hings depend on humans as well , again most obviously for human made goods , which depend on humans to manufacture and use them. This dependence of things on humans (and other things) creates relationships of dependency, that is, it imposes constraints on human activity. This is so because the things that humans depend on are (Hodder 2011b:180), and so humans have to work to obtain, maintain, and replenish things. The net result is that humans become caught in a bind , things that, in turn, depend on people and other things (Hodder 2011 a:1 6 4, 2012:88) . This is the defining feature of entanglement. Human dependence on things creates obligations as people become drawn into relationships of care . The material properties of things thus create that, because of dependent relations, dictat e and direct what people do. This is entanglement: the dialectical relationship between dependence and dependenc y, between the enabling and constraining reliances of humans and things (Hodder 2012:88 89). This e ntanglement is directional and increases in complexity over time. It is directional in the sense that we humans are so dependent on things that our response to th e breakdown of things is to fix them, tinkering and patching them up
260 to the drawing board innovations and course corrections never occur, but rather that there is a path dependency and historical contingency to human thing entanglement, that is, to social and technological change (Hodder 2012:167 171) . Solutions and innovations things have to fit into existing entanglement s, phenomenal worlds, and conceptual ideologies . The degree of human thing entanglement increase s exponentially over the long term (Hodder 2012:174 177) . The more people come to use and depend on things, the greater their entr apment in relations of dependency. The constant tinkering, maintaining, and repairing of things entails other people and things, expanding the web of entanglement. There is a temporal element to this as well, as h umans and things operate on different tempo ral scales . A s we depend on things we become beholden to their own unique temporalities and this introduces uncertainty in where and when intervention is needed (Hodder 2011a:164, 2011b:181, 2012:84 85) . In this way, humans become caught up in things in v arious ways, through various relationships of reliance and care. And through these entanglements things and people are drawn together across multiple spatial and temporal scales. Things can of course continue to exist absent of human care, but things canno t exist for humans, in the ways that humans want, without human intervention. Resources, wild or domesticated, animate or inanimate, need tending, conserving, protecting, if they are to exist in the ways that humans want them to (Hodder 2011a:162) . This is true of springs as well. T he dependence of humans on springs has entrapped them into relationships of care that necessitate certain actions on the part of humans. The various dependences and dependencies of humans and springs are explored below to illustr ate the depth of their entanglement. This analysis reveals not only the
261 complexity of problems currently plaguing springs health, but also the relevance of archaeological data to potential solutions. Humans Depend on Springs Humans depend on springs in a variety of capacities. Minimally, springs are a source of clean, fresh water. This is most apparent in archaeological narratives of the Paleoindian period. As discussed i n chapters 2 and 4 , current reconstructions indicate that sea level w as some 80 m l ower than present when people f irst inhabited the peninsula. The climate was warmer and drier, with significantly lowered ground water, reduced surface water availability, and xeric vegetative communities (Balsillie and Donoghue 2004; Otvos 2 004; Watts et al. 1996; Watts and Hansen 1988). The fauna of the state was also quite different, with Pleistocene megafauna roaming what were likely arid plains and sandhills. The Oasis Model of Paleoindian settlement, first proposed by Neill (1964) and la ter elaborated on by Dunbar (1991), posits that t hese early inhabitants of Florida were constrained by their need for re liable freshwater . Hydrological modeling of Late Pleistocene Florida indicates that large springs are likely some of the few locales whe re fresh water could be obtained in the interior of the peninsula (Thulman 2009) . As such, Paleoindian populations may have been tethered to these places, frequently revisiting them in the course of their subsistence pursuits. Springs would also have attra cted large game in search of water, thus affording people ample hunting opportunities . Most well documented inland sites of the Paleoindian period occur in association with karst features like springs and sinkholes . For example, substantial late Pleistocene and early Holocene deposits have been recorded in northwest Florida at t he Page Ladson site , submerged sinkhole in the Aucilla River system (Dunbar et al. 1988; Webb
262 2006) , and the Wakulla Springs Lodge site (Rink , Dunbar, and Burdette 2012; Tesar and Jones 2004). Closer to the St. Johns River valley , Paleoindian artifacts and Pleistocene fauna have been recovered from Silver Springs (Neill 1958), and are frequently recovered from the Santa Fe and Ichetucknee rivers in north cen tral Florida (Thulman 2009). The headwaters of the Ichetucknee River are formed by the Ichetucknee Springs group, while the Santa Fe River system incorporates numerous karst features, including sinkholes and several first magnitude springs. This depend ence on springs entailed dependence on other things as well. The presence of springs afforded the opportunity for humans to move into the otherwise arid interior of Florida . However, in many cases water from these springs did not flow onto the surface. For exa mple, Paleoindian materials at Little Salt Spring were recovered from subaqueous deposits nearly 26 m below the current spring surface (Clausen et al. 1979). Similar materials were recovered from organic sediments 13 m beneath the surface of nearby Warm Mi neral Spring (Clausen et al. 1975), including human remains dating to at least 12,000 cal B . P. The recovery of objects from deep within cave systems at Silver, Silver Glen, and other springs suggests that these , too , were more akin to water bearing caves o r cenotes than springs at the time. Th us, obtaining water from springs was no simple matter and in some cases involved considerable risk. This also implicate s dependence on a suite of other things , for example rope, cordage, or vines for descending into the spring and waterskins or other containers to hold the water. However, I have argued at length that the significance of springs was not merely a result of their capacity to supply food and water. In Chapter 3, I evaluated the
263 hypo thesis that the onset o f spring flow was an ecological founding event underwriting riverine adaptation and the inception of shell mounding in the St. Johns River valley . Cores extracted adjacent to Salt and Silver Glen springs demonstrated that the onset of spring flow predated initial shell mounding regionally by millennia. Meanwhile, stratigraphic testing at these and other springs indicated that sustained human activity at springs did not initially involve the deposition of shell. Thus, although people depended on springs, she ll mounds apparently did not, or at least not initially. Rather, the centrality of springs increasingly turned on their role in facilitating movement and interaction across the Florida peninsula. Chapter 4 discussed the centrality of Silver Glen Springs in particular as a n extra regional gathering place that drew visitors from across the peninsula and facilitated interaction and community . I argued that Silver Glen was made sacred by the confluence of dispersed people and things at a spring with an abundant material historical record in the form of pits, shell ridges, and other deposits. People thus depended on springs as places of coalescence and community building. Springs were similarly depended on to facilitat e movement across the peninsula, and in this movem ent became entangled with human dependence on stone . As discussed in Chapter 5, stone suitable for the product ion of tools is absent in the St. Johns River valley . Silver Springs , among others, served as a conduit or way station along the path of ston e from lithic source areas in northwest peninsular Florida to the St. Johns River valley . GIS based least cost analysis demonstrated that the optimal routes taken by sto ne from several distinct quarry clusters passed through Silver Springs on their way to Silver Glen. Provenance determinations and debitage analysis of lithic
264 assemblages confirmed this hypothesis, particularly for Ocala quarry cluster chert, which comprised 56% of the assemblage at Silver Glen. Based on these same analyses, sites fa rther to the south Lake Monroe Outlet Midden and Thornhill Lake Complex are dominated by different lithic source areas and do not appear to have been part of the same circulation network. However, the least cost paths suggest that lithic resources at these sites may likewise have passed through springs (e.g. Wekiwa) along their routes. Human dependence on springs continues today. We rely on springs in much the same way as ancient Floridians did, as sources of freshwater but also as sites for soc ial gatherin gs . Here it is helpful to consider springs not just as points on the landscape , but as windows into the hydrological web woven by the Floridan Aquifer System. As noted in Chapter 2, groundwater s Flo ridan Aquifer supplies the bulk of this potable water and is the primary source of freshwater for agricultural irrigation, industrial, mining, and commercial supply (Marella 2014). Although w ater is not frequently extracted directly from springs , as it was in the past (the bottled water industry notwithstanding [Samek 2004]), over one million artificial wells act as proxy springs, discharging water from the Floridan Aquifer . Likewise, m odern Floridians continue to use springs as arena s for social gathering s, now in the context of recreation. Many canoe or kayak along spring runs in search of tranquility and communion with nature, and trips to the springs for birthday parties, family outings, swimming, tubing, and fishing are highly valued component s of Flor idian life, particularly in northern Florida. As retired Chair of the Florida Springs Task Force and the Florida Springs Initiative Jim Stevenson wrote:
265 Springs are woven into the cultural fabric of rural North Florida. Springs were the gathering places fo r teens, much as malls are today. After a day of working in the fields, people would go to a spring to bathe and cool off. Springs were the social centers for family reunions, church baptisms, and school picnics. Health spas located at spri ngs with hotels attracted invali ds and tourist s from the North expecting to be healed in the medicinal water of White, Wekiwa, and Hampton Springs (Knight 2015:xiii). These social gatherings may seem to be incommensurate with pan regional gatherings that integrated far fl ung people during the Archaic period, but springs today continue to draw a regional crowd (Bonn 2004) . And activities like basking i n the water, feasting on hot dogs, tossing a frisbee, imbibing, and singing by the campfire are not so different from those of the social gatherings millennia ago. Springs Depend on Other Things If humans depend on springs, springs, in turn, depend on humans and on other things. As detailed in Chapter 2, springs depend on a number of materials, places, processes both immediate and far removed in space and time. I defined a spring broadly as a point on the landscape where groundwater flows onto the surface. As a point, or place, a spring thus draws together and is dependent on a number of other things. A spring, for example, is d ependent on the water that outflows and on the void in the stone (i.e., pore, conduit, or cavern) that allows for the movement of water. Indeed, word can refer to the wa Protection Agency 2002:157). Spring water is ultimately derived from precipitation that percolates through overlying sediments to enter and recharge the Florida Aquifer. This recharge is facilitated or hindered by the thickness and hydraulic conductivity of sediments overlying the aquifer, which in turn are a product of ongoing geomorphological processes that
266 illion years ago. These sediments (and atmospheric CO 2 ) also facilitate further karstification of the landscape, as they acidify percolating water and thus enhance dissolution of the limestone , creating and expanding pathways for water to flow from springs . This is also dependent on microbial ac tivity and organic matter in soils , which further acidify the water. Local geomorphology and topography likewise influences the spatial position of springs, which are typically located at inflection points between hi gh and low elevation terrain. The water discharging from a given spring is primarily composed of relative ly young (< 100 years residence time) water that fell on the local springshed the surface and groundwater basins that contribute water to a spring. Th i s spring shed is not apparent from the surface terrain, but a product of largely invisible groundwater flow dynamics directing water to the spring. Likewise, the size and configuration of the springshed changes over time as these forces fluctuate. However, as outlets of the Floridan Aquifer, springs can also discharge water upwelling from deep i n the system. This water recharged the aquifer as much as 30,000 years ago, and can include relict sea water trapped in the system when sea level rose after the Last Glacial Maximum (Moore et al. 2009; Morrisey et al. 2010; Plummer 1993; Toth and Katz 2006). A spring is thus dependent on the motion of water through the cycle of recharge, flow, and discharge. Amongst artesian springs groundwater flow and discharge is dr iven by gradients of pressure (and also temperature) in the confined aquifer. These pressure gradients are a result of elevation differentials between recharge and discharge zones, and fluctuate with changes in precipitation patterning and intensity, and t he level of the
267 seas surrounding the Florida peninsula. The motion of spring water is afforded by the carbonate limestone comprising the Florida aquifer, and the processes that have enhanced porosity and permeability of that stone. The Irish poet philosoph er described the karst landscape of western Ireland the Burren region carbonate limestone platform for med in a shallow marine environment and has been carved by the repeated transgressions and regressions of the sea. It was dependent on a confluence of conditions that favored the proliferation of specific organisms to produce and accumulate carbonate sedim ents. Springs thus depend on, among other things, precipitation, stone, topography and terrain, springsheds, the sea, and the geologic processes and events that formed the Florida platform and Floridan Aquifer. Springs entanglement with other things is ope n ended, and examples can be expanded. For example, over centennial and millennia l scales, springs depend on fluctuations in sea level and precipitation resulting from hemispheric and global scale changes in atmospheric and oceanic circulation patterns, i nduced by post and patterns of solar insolation. More recently, anthropogenic inputs have imposed dependencies on springs. Springs Depend on Humans Over the course of thirteen millennia of human springs interaction, springs have come to increasingly depend on humans. On the one hand, sp rings dependence on humans results from the long history of land alteration and terraforming adjacent to springs. As discussed in Chapters 3 and 4, some of the ear liest anthropogenic
268 deposition at Salt and Silver Glen springs involved both the aggradation and progradation of the shoreline as midden accumulated in both terrestrial and subaqueous contexts during the Mount Taylor period . The elevated deposits facilitat ed access to the spring by covering over low lying saturated areas on the margin of the spring pool and surrounding it with a high, dry platform. The emplacement of thick deposits of large, clastic particles (i.e., shells) likewise stabilized the shoreline of the spring pool. Indeed, as the water level in the spring fluctuated on seasonal, annual, and longer scales, shell deposits became concreted through these repeated cycles of wetting and drying. The spring s (as we know them today) continue to depend on anthropogenic shoreline stabilization and aggradation. Although the shell mound s surrounding Silver Glen Springs w ere removed by mining operations early in the twentieth century, intact deposits extend nearly two meters beneath the current ground surface. The current recreation area with its beach access, volleyball court, g rassy fields, and picnic tables is quite literally built on an Archaic era foundation that continues to facilitate access to the spring by elevating the terrestrial surroundings and stab ilizing the banks. When compar ed to Silver Glen Springs, Salt Springs was h ome to relatively modest pre Columbian anthropogenic deposits. Consequently, modern interventions towards shoreline stabilization and spring access have been more extensive. The pool of Salt Springs is now surrounded by a concrete retaining wall forming three sides of a rectangle. Salt Springs is not alone in this capacity . N earby Juniper Springs is surrounded by a limestone wall , w hile Orange Spring toward the northeast over a 2 ft. (0.6 m) high, man made, limestone waterfall. Past the
269 waterfall, the spring 2004:230) ladders and a concrete sidewalk encircle the spring pool. The spring outflow pours through a concrete weir examples can be multiplied. In addition to shoreline stabilization, shoreline aggradation and infilling is a common feature of springs. At Alexander Spring, access to the pool is provided b y a beach contained by a rock retaining wall. Longtime resident Robert Shepard reported (Figure 6 5) . This was later confirmed by sub surface testing (Willis 1995). Archaeological reconnaissance conducted by the LSA has likewise documented large scale anthropogenic infilling of spring side wetlands , in multiple locales. At Silver Springs, much of the main p ool is surrounded a wooden retaining wall and dock for the glass bottom boats. During the LSA survey, modern fill was found to be discontinuous, but widespread , particularly on the north side of the pool . Fill was indicated by mottled deposits or near surf ace depositional stratification, often coupled with the recovery of modern cultural materials or buried utilities. The area adja cent to the retaining wall contain s thick deposits of modern fill, in some cases exceeding 1 m deep, overlying the former wetlan d surface . Likewise at Otter Springs, fill sand and modern overburden of variable thickness overl ie buried surface ho rizons and wetland deposits (Figure 6 2014) .
270 Weeki Wachee Spring is perhaps th e most heavily modified spring in Florida . Recreational facilities are extensive, and include concessions, gift shops, water slide s a (constructed) b each at Buccaneer Bay Water Park , boat docks for the Wilderness River Cruise, nature trails, animal demonstrations, canoeing and kayaking do wn the Weeki Wachee River , and the famous Underwater Theater built into the west side of the spring pool. As at Otter and Silver Springs, an LSA survey at Weeki Wachee springs documented widespread modern disturbance of variable depth (Figure 6 7 ; see noughue and Sassaman 2013 ), with f ill sand and modern debris frequently overlying truncated soil profiles and buried surfaces . This was sometimes obvious, as in a shovel test pit that had to be t erminated when it intercepted a concrete walkway on a buried surface 85 c m deep . P roximate to the spring and river, fill typically overl ies a dark br own to black buried surface horizon consisting of organically enriched, mucky sands and/or peat deposits. This indicates that sand was used to fill low lyi ng wetlands marginal to the spring. , spring flow is likewise dependent on human activity. As noted above, over 2,500 million gallons per day were pumped from the Floridan Aquifer in 2010 (Marella 2014). Groundwater extraction has cause d significant reduction of water levels and pressure in the Floridan Aquifer over much of its extent. Historically, surface water was the primary source of freshwater use in Florida (Marella 2008) . Before the installation of artificial wells, springs were the primary outlet of groundwater from the Floridan Aquifer (Knight 2015:52). The drilling of wells began in the late nineteenth century, when the City of Jacksonville emplaced two wells in 1884 (Bush and Johnston 1988; Knig ht 2015:277). The rate of groundwater
271 extraction grew steadily over the course of the twentieth century, so that by 1980 groundwater had surpassed surface water as the primary freshwater source in Florida. In 1950, the amount of groundwater extracted for h uman consumption was 614 million gallons per day (Marella 2008) . Sixty years later, in 2010, it had skyrocketed to 4 ,166 million gallons per day, an increase of nearly 600%. Granted, groundwater withdrawals were relatively stable from 2005 2010, but this w as largely a consequence of higher than normal rainfall during this interval (Marella 2014). As a result of this groundwater withdrawal, many springs across the peninsula have witnessed declining flows in recent decades. T here is a spatial heterogeneity t o this, as the greatest reductions in spring flow do not occur in the same areas with the greatest groundwater consumption. The residents and businesses of Jacksonville currently withdraw some 90 million gallons per day of groundwater from the Floridan Aqu ifer, lowering the potentiometric surface of the aquifer in the imme diate vicinity by 12 m. This groundwater reduction is not limited to pumping areas, however , as it creates a gradient of pressure, driving groundwater flow in the Floridan Aquifer from surrounding areas into the pressure sink beneath Jacksonville. The results radiate out into southern Georgia and north central Florida, a ffe cting areas more than 100 km distant and reducing flow in springs of the Ichetucknee, Santa Fe, and Suwannee rivers ( Florida Springs Institute Members 2015; K night 2015 ) . Groundwater extraction is not the only anthropogenic impact reducing aquifer pressure and spring flow . As noted above, spring flow is dependent on precipitation recharging the Floridan Aquifer. Over the course of the twentieth century, the re has been a broad trend towards decreased precipitation and increased daily maximum
272 temperature in Florida. This has been attributed to the drainage and conversion of millions of acres of wetlands to agricultural, residential, and industrial uses. The storage and decreased the temperature and humidity gradients that drive sea breezes and convective summer rainfall (Marshall et al. 2004; Pielke et al. 1999). Thus, twentieth century development and landscape alterations in Florida have altered local climate , particularl y in southern Florida, and reduced recharge t o the Floridan A quifer. Water quality is also dependent on human activity. As noted at the outset of this chapter, increased nitrate nitrogen in springs and the Floridan Aquifer is a threat to both human health and l ocal ecologies. Nitrate has been the pri me culprit in attempts to explain the proliferation of filamentous algal mats covering spring bottoms and floating on the surface. As a form of nitrogen, nitrate is a nutrient essential for plant and animal life. The primary source of nitrate entering the aquifer is fertilizer, but other sources include human wastewater treatment and disposal (whether from municipal facilities or septic tanks), animal waste, and industrial byproducts. The aquifer is particularly susceptible to pollution from surface applica tions in recharge areas where it is unconfined and/or near the surface. It is generally though t that at lower concentrations nutrients fav or native vegetation in springs , including a variety of submerged, floating, and em ergent vascular plants, while at hi gher concentrations nutrients favor fast growing algae, which then out compete native plants (Brown et al 2008; Knight 2015). However, several factors suggest that increased nitrate nitrogen may not be the primary factor driving algal proliferation in spri ngs (Heffernan et al 2010). Experimental and observational studies
273 indicate that there is not a strong correlation between elevated nitrate nitrogen and algal proliferation in springs, and that increases in the two are often spatially and temporally discon nected. Silver Glen Springs, for example, has seen a rapid increase in algal biomass, but without a corresponding increase in nitrate nitrogen levels (Knight 2015:79 ; Pandion Systems 2013; Pittman 2012b ). This implicates other forcing mechanisms and has le d to alternative hypotheses. Researchers from the University of Florida have argued that invertebrate grazers, such as aquatic gastropods, exert a strong predatory control on filamentous algae (Heffernan et al. 2010; Liebowitz et al. 2014). Grazer populati ons have been reduced in recent decades as a result of declining dissolved oxygen levels in spring water. This decrease in dissolved oxygen is attributed to groundwater pumping and the concomitant incorporation of deeper, older, more oxygen depleted water in spring discharge. The depression of grazer populations has allowed algal populations to proliferate, eventually reaching a tipping point where grazers can no longer effectively control algal populations. While acknowledging the role of grazers in algal control, Knight (2015:77 78) suggests that a complex interaction of physical, chemical, and biological forces control the balance of vascular plants and algae. Rather than focusing on factors that favor algal growth, Knight emphasizes changes that have negatively impacted native plants. These include increased nitrate nitrogen and reduced dissolved oxygen and grazer populations, but also reduced flow velocity as a result of groundwater extraction, trampling from recreational activities, and the application of herbicides to control
274 invasive plant species. In this view, those springs m o st vulnerable to algal proliferation are those in which native plants are heavily stressed by these activities. Although the precise mechanisms for increased algal dominance are debated, human activity is universally implicated (Heffernan et al. 2010; Knight 2015; Liebowitz et al. 2014). Whether from nutrient enrichment, grazer populat ions, or some combination of factors, it seems clear that pollution, pumping, and physical damage resulting from human activity are responsible for reduced flows and degradation of water quality in springs. Absent future interventions, springs will continu e to decline and thus depend on humans for their continued existence. Springs Entangled Entanglement involves not only dependent relations, like those outlined above, but also dependencies as humans a re entrapped in the downstream reliance s of things . Huma ns depend on springs that themselves depend on humans and other things , a nd so humans are entrapped in relationships of care. Springs are not eternal. Springs fall apart. They require maintenance and repair , imposing constraints on and directing human acti vity . At Silver Glen Springs, aging recreational facilities have to be replaced and eroding midden deposits along the mining escarpment and pool entrance required stabilization with clean fill (Randall et al. 2011). At Salt Springs, a portion of a relict t imber retaining wall was removed and replaced with concrete in 2009. As described in Chapter 3, the coffer dam installed to facilitate this repair exposed subaqueous midden Alexander Spring: Erosion and sedimentation have been a constant maintenance lot and roads courses through the recreation area during downpours that
275 he beach due to intense foot traffic and the constant movement of 76,000,000 gallons of water per day downstream from the boil (Willis 1995:11). This has required repeated dredging of introduced sand from the spring and constant replenishment of the beach with fresh fill sand. This is a consequence of several factors, including the removal of vegetation that would otherwise stabilize slopes, the infilling of the wetland that acted as a sediment trap, and the introduction of unconsolidated sediment through b each construction in the 1950s. The beach at Weeki Wachee Spring necessitates similar dredging and rejuvenation (Spicuzza 2005 ) , and facilities like the Underwater Theater require maintenance and upkeep for both safety and aesthetics . ition to a state park in 2013 brought with it several million dollars of repairs and renovations , paid for by the former lessee, Palace Entertainment, Inc. This included rehabilitating walkways and boardwalks, replacing rotted wood and painting structures, removing invasive plants, and demolishing zoological exhibits , as noted above . However, the Florida Dep a r tment of En vironmental Protection were bigger than first (Thompson 2013 a :3). Indeed, t both Silver and Weeki Wachee springs from private entertainment companies , and their conversion to State Parks, was motivated by the failure (or inability) to maintain aging facilities and en sure protection o f the springs from degradation. Rehabilitating and redeveloping these facilities necessitated the archaeological surveys conducted by the LSA. State agencies have resorted to mechanical and suction dredging to r emove benthic algae from sev eral springs (e.g., Chassahowitzka , Otter, and Weeki Wachee). And tens of millions of dollars have been spent constructing wastewater facilities and
276 purchasing land around springs to protect them from pollution. The City of Tallahassee, for instance, spent over a quarter of a billion dollars on wastewater treatment facilities to reduce nitrates in Wakulla Springs (Shockman 2015). Similarly, the proximity of Weeki Wachee Springs to U.S. Highway 19 and a massive parking lot required the installation of facilities to capture and treat stormwater runoff (Southwest Florida Water Management District [SWFWMD] 2009). Spri ngs are thus entangled with anthropogenic and planetary processes and events at multiple temporal and spatial scales. Springs are unique and complex hydrological and ecological entities, and so their extensive entanglement is perhaps not surprising. But I want to draw out two salient features of springs entanglement as they pertain to springs archaeo logy and conservation. The first is that springs entanglement itself is in many ways a barrier to conservation. Predicting the impacts of interventions on sprin gs ecology and hydrology is fraught with uncertainty because of their dynamic complexities that is, their entanglement with forces operating at divergent spatial and temporal scales. Changes taking place at springs (i.e., flow and pollution) have been ongo ing for decades, with a temporality scarcely perceptible on the scale of the human lifetime. Likewise, the spatial scale of springs hydrological ties is largely obscured from human perception. Springsheds the basins contributing water to a spring are not v isible on the surface and vary in size and extent over time. It is not at all apparent where, for example, nitrates from fertilizers on an agricultural field will end up after percolating through the soil and recharging the aquifer. Protecting springs wate r quality is largely a matter of managing land use practices within springsheds. However,
277 because these are moving targets, the process is far from straightforward. It is therefore difficult to determine what actions need to be taken on what portions of th e land surrounding a spring to best protect it from further pollution. It is likewise not obvious to the casual observer that because of the entanglement of springs actions in a given locality can have impacts in distant parts of the Floridan Aquifer, as w ith the example discussed above of groundwater pumping in Jacksonville impacting springs 100 km away . The spatial scale and invisibility of springs entanglement has economic implications as well. As noted in Chapter 2, springs are found primarily in the k arst regions of northern Florida. This area has a relatively low population density, in comparison to largely springs abundance is also located away from springs. Figure 6 8 shows the distribution of sprin gs in Florida, per Florida DEP data (2012), and median per capita income (PCI) by county in 2013. Count PCI ranged from $20,294 to $64,872, and counties can be divided into low, medium, and high income tiers in roughly $15,000 increments . These data reveal that 82.7 % of springs are located in low i ncome counties, 16.8% in medium income counties, and 0.5% in high income counties. Financial, and thus political, capital is largely concentrated in areas of the state devoid of springs. Recent research has demons trated that those living close to a spring are more vested in springs conservation (Alenicheva 2012. This is problematic for springs conservation, given that actions taking place in distant locales can negatively impact springs. Protecting springs requires people to change their behavior with regard to water use and the application of polluting substances, such as fertilizers, over a wide
278 swath that may be far removed from springs. As Jim Stevenson, retired Chair of the Florida Springs Task Force quoted abo ve, said in a recent interview: w e have spent substantial time and funding to educate Floridians about the plight of our springs and what citizens can do to overcome the they must be outraged by the loss of their springs... Our springs will survive or die by the level of public sentiment ( quoted in Swihart 2013). Education has not been enough, partially because of the entrapment of our entanglements. Floridians are so reliant on ground water, so entangled with it, that actions proposed to reduce withdrawals are difficult to implement. And the prospect of divesting ourselves of pumped groundwater completely and turning to other sources of freshwater is not considered possible , or not cons idered at all. The second salient feature of springs entanglement is the long term continuity in human activity at springs . The myriad ways that springs and humans are now co dependent are historically contingent and a result of the path dependencies instigated by human springs engagement in the past. The entanglement has increased through time, but a ncient human activities a t springs (e.g., extracting water, infilling marginal wetlands , aggrading shorelines , gathering for social events) are the same as those taking place today, scalar differences notwithstanding . Th e continuity of human springs entanglement over millennia und ercuts narratives of changelessness invoked by the springs eternal trope. Likewise, the p ath dependency and historical trajectories of spring s entanglement points to the futility of backwards looking conservation goals seeking a pristine or more natural st ate for springs. Referring to water usage in Marion County and flow reduction in Silver Springs, Ed Lowe, Chief S ci entist of the St. Johns River WMD said recently, ala Star Banner
279 2015) . Likewise, accept that spring s will never be returned to their former natural majesty . Nevertheless, this continues to be the target of springs conservation. Knight c ontinues, ethically, this should not be an option seriously considered by the concerned public or by those public officials who are in leadership or regulatory positions and who can potentially contribute to a reversal in the trajectory of depletion and p o llution that has en gulfed Florid springs . These c onservation efforts seek either to reverse time to a pristine state or to maintain the status quo (e.g., sustainable use) , but are unlikely to succeed. Indeed, the long term record of human landscape al teration at springs like Silver Glen discounts the very existence of a pristine or natural state to return to. We cannot escape our entanglement with springs by looking back. Rather , conservation efforts need to be future oriented, target ing activities and goals consonant with ongoing human use of springs. Again, the argument is not that springs will cease to exist without human intervention although their water quality and flow may be impaired but rather that they cannot continue to exist in the ways that people want them to without care. A central question of springs conservation, then, is what are springs for? Today, springs are for swimming, tubing, and camping with family and friends; contemplating, painting, photographing; studying, exca vating, working . Th e activities that people engage in to maintain springs (e.g., repairing aging facilities, remediating algal prol iferation) serve to maintain the aesthetic and ecological integrity of springs that underwrites all of the ways that people s eek to use springs . There are, however, competing interests. Take, for example, the case of Silver Springs. After becoming a state park in October 2013,
280 renovations and facility repairs continued under the management of the Florida DEP. As of late 2014, ov er 20 buildings had been demolished, and all zoological exhibits removed as part of returning the park to a more pristine state: [Park Manager Sally] Lieb walked the east side of the park on a recent day and stood by the remains of a 1 foot thick, 10 foot high wall. She pointed to an empty field that was once home to the park's cougars. The few feet of wall is all that remains. Lieb hopes to let the field return to its natural state. Then she pointed to a canal dividing where she stood and the five acre Ros s Allen Island, where captive alligators once were housed behind (Hiers 2014:2). However, the DEP also has to accommodate park visitors, some of whom are less enthusiastic about the ch anges. Hiers (2014:4) quoted one, among many, disappointed so future intervent ions at springs and the direction of their entanglement hinges on the question, what are springs for? Recent research has demonstrated that proximity is an important factor motivating public wi ll towards springs conservation, as those living less than five miles value the springs to a greater extent, more often engage in local volunteer actions, and have a greater willingness to increase personal knowledge and share their knowledge with others (Alenicheva 2012:14). As a corollary to this, di rect experience with springs is also an important factor determining public sentiment for springs conservation. In other words, i n order to people to value springs, and thus be motivated to take actions to protect them, springs must be experienced , people must
281 develop a relationship with these places . However, it is not enough for people to visit or admire springs. As Basso (1996 b :56 57) wrote: it is simply not the case, as some phenomenologists and growing numbers of nature writers would have us believe, t hat relationships to places are lived exclusively or predominantly in contemplative moments of social isolation. On the contrary, relationships to places are lived most often in the company of other people. Indeed, if places like Silver Glen Springs are an y guide, the value attached to springs is amplified and manifested in moments of social gathering. And these gatherings do not (and did not) take place in an historical vacuum, but are most affecting in the context of social memories of past gatherings, on ce materialized in the form of various pits, mounds, interments, and offerings. Continued recreation, then, is a key to mobilizing public sentiment for springs protection and conservation. Through recreation al gathering at springs, people can continue to d evelop the relationships with these places that are necessary for their ongoing cultural valuation. Likewise, the long term historical import of springs should be brought to the fore in conservation narratives and highlighted at recreational facilities. Th e goal here is not to provide a baseline for comparison or construct simplistic narratives of people living in harmony with springs, b ut to emphasize the generations of curre nt 2008:2). Springs can be made meaningful by accentuating the continuity of place, enlivening it with history, despite the gulf of experience that seemingly separates the ancient a nd modern. Conclusions and Future Directions In sum, the key claims argued in this dissertation are:
282 Spring flow began far earlier than thought in current archaeological interpretation, and did not directly precipitate riverine adaptation and shell moundin g in the St. Johns River valley . Springs were key to cultural developments in the Florida peninsula, but not (or not primarily) for subsistence reasons. Rather, some springs were made sacred through a series of social gatherings that drew on both the physi cality of springs and their historical significance. Springs facilitated movement and interaction across the peninsula, integrating far flung peoples. Springs continue to gather people today, and many modern activities are of the same kind, if not scale, as ancient ones. Springs are thus entangled hydrologically, historically, spatially, socially, and politically. Archaeological knowledge is relevant to springs conservation, as it d emonstrat es the long term continuity of human action at springs, reveals th e ways that springs were made meaningful in the past, and provid es a basis for future oriented conservation measures. There are several avenues of research that would further substantiate these claims and add to our knowledge of Florida springs archaeology . First, although archaeological survey and testing has taken place at several springs, as reported here, intensive testing of the kind necessary to make detailed reconstructions of diachronic changes at springs are relatively few. Silver Glen Spring is the exception here, as it has been extensively investigated by the St. Johns Archaeological Field School of the University of Florida. Several other springs with known archaeological deposits would benefit from similar sustained programs of excavation. Notably, eight archaeological sites have been documented proximate to Alexander Spring in the Ocala National Forest (Dunbar 2003; Willis 1995). Few of these si tes have been archaeologically tested, but they include both terrestrial and subaqueous deposits spanning the Paleoindian through St. Johns II periods. Controlled excavations along with l ithic
283 provenance determinations and dating of basal spring deposits a t Alexander and similar springs would yield data pertinent to the interpretations advanced here. Second, the bulk of archaeological research conducted at springs has taken place at large springs currently housed in state parks, U.S. Forest Service recreati on areas , and the like . We could learn much from studying smaller springs in the region, for example Juniper, Fern Hammock, Sweetwater, and Mormon Branch springs, all of which are proximate to Silver Glen and feed into Lake George. Likewise, l ittle is know n of the prehistory of the Wekiw a River, a tributary of the St. Johns, and nearby Seminol e State Forest. There is a concentration of over 30 springs t here , many of which have not been visited by archaeologists. Similarly, archaeological survey along the le ngthy spring runs of Alexander, Juniper, and Salt springs (among others) would allow for greater inference about the integration of springs into the larger landscape. Third, more detailed hydrological constructions of springs are likely possible. This woul d move beyond dating the onset of spring flow to examine fluctuating conditions in the spring over time. Plant macrofossils are generally well preserved in organic deposits like those documented at springs , and can be used to reconstruct the local plant co mmunities and surface wetness (Chambers et al. 2012). A more detailed reconstruction of surface moisture and water table depth can also be inferred from testate amoebae. These single celled organisms are highly sensitive to variations in moisture (Andrews 2012; Booth et al. 2010; Chambers et al. 2012; Escobar et al. 2008; Mitchell et al. 2008). The feasibility of extracting testate amoebae from spring deposits has not been evaluated , but a pilot study should be sufficient to determine this.
284 Finally, the hyp othesis discussed above that a reduction in grazer populations may negatively impact springs health points to the intriguing, if unconfirmed, possibility that the removal of gastropods from springs in ancient Florida had unintended consequences. The Florid a Rasp Elimia ( Elimia sp.), in particular is thought to be an important predator of filamentous algae in springs ecosystems (Liebowitz et al. 2014) and is frequently documented in spring side archaeological deposits. Data from archaeological collections at springs could be used to illustrate variability in Elimia populations. These can be correlated with changes in other material remains to provide new insight into variability in springs ecosystems in the past. If I may return once more to The Yearling , and Rawlings (1938:4) description of t excited Jody to watch the beginning of the ocean. There were other beginnings, true, but this one was his own With over 1,000 springs in Florida, there is no shortage of future study sites. This dissertation too is a beginning, and there is much yet to be learned about the history
285 Figure 6 1. Railroad and steamboats at Silver Springs in 1886. Photograph by George Barker, courtesy of the Library of Congress Prints and Photographs Division , Washington, D.C.
286 Figure 6 2. Interior of the health spa at White Springs, Hamilton County, Florida , ca. 1912. Courtesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/149898
287 Figure 6 3. The early twentieth century resort at DeLeon Spring. Cour tesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/158585
288 Figure 6 4. Weeki Wachee Spring in 1952. Courtesy of the State Archives of Florida, Florida M emory , http://floridamemory.com/items/show/149889
289 Figure 6 5 . Alexander Springs, circa 1947. Courtesy of the State Archives of Florida, Florida Memory, https://floridamemory.com/items/show/125603
290 Figure 6 6 . Represent ative shovel test pit profiles from Otter Springs. Clockwise from top left: STP s 1, 7, 14, 16. Note: photos are not to scale and were taken at an oblique angle
291 Figure 6 7 . Interpolated depth of modern infilling or subsurface disturbance in the vicinity of Weeki Wachee Springs. Sassaman 2013
292 Figure 6 8. County median per capita income and the distribution of springs in Florida. Springs data from the Florida Department of Environmental Protection (2012). Four offshore springs are not shown. Economic data from the U.S. Department of Commerce, Bureau of Economic Analysis (www.bea.gov)
293 APPENDIX A STRATIGRAPHIC DESCRIPTIONS
294 Table A 1. Stratigraphic units documented in 2009 test excavations and cores at Salt Springs (8MR2322) Stratigraphic Unit Maximum depth below surface (cm) Munsell Color Description Overburden 29 10YR3/3 10YR6/3 Dark brown and pale brown fine sand. Slightly finer near surface. Modern or historic human transported material. IA 1 67 10YR3/2 Very dark grey coarse sand coated with organic matter. Abundant whole and crushed Viviparus shells and few whole bivalve shells. Concreted shell and matrix common near top of stratum. IA 2 68 10YR2/2 Very dark brown loamy sand coated with organic matter. Abundant whole Viviparus shells and common lenses of whole and crushed bivalve shell. IB 80 10YR4/1 Dark grey sand with abundant whole and crushed Viviparus shell and few whole and crushed Pomacea and bivalve shells. II 96 10YR3/1 10YR2/1 Contorted lenses of very dark grey and black sand, coated with organic matter. Common vertebrate fauna, charcoal, charred and uncharred wood fragments, hickory nut, and other botanicals. No shell. III* 97 10YR4/3 10YR4/1 Brown sand mottled with dark grey sand. Common organic matter and botanicals, including hickory nut and wood fragments IV* 103 10YR7/1 Clean, light grey coarse sand with few small wood fragments V* 114 10YR4/3 10YR3/2 Stacked, alternating laminae of brown and very dark greyish brown loamy sand. Common organic matter VI* 117 10YR7/1 Clean, light grey coarse sand with large wood fragments at the top of the strata *Strata observed in percussion cores only.
295 Table A 2. Description of vibracore BC01 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 25 5YR2.5/1 Black mucky peat with common fine to medium vegetal fibers (decreases below 16 cm). Clear boundary. 31 10YR4/2 Dark grayish brown mucky medium sand. Clear boundary. 55 10YR2/1 Black mucky fine sand. Few medium vegetal fibers. Clear boundary. 59 10YR3/1 Very dark gray mucky medium sand. 66 10YR2/1 Black mucky medium sand. 72 10YR3/1 Very dark gray mucky medium sand. 100 10YR4/2 Dark grayish brown medium sand, mottled with very dark brown (10YR2/2) and grayish brown (10YR5/2) medium sand. 117 10YR2/1 Black mucky sand mottled with dark gray (10YR4/1) medium sand. Clear boundary. 153 10YR3/2 Very dark grayish brown mucky medium sand intercalated with dark brown (10YR3/3) and light brownish gray (10YR6/2) medium sand. 160 10YR2/2 Very dark brown mucky sand. 208+ 10YR3/3 Dark brown medium sand, grades with depth to brown (10YR5/3) medium sand. Table A 3. Description of vibracore BC02 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 17 10YR2/1 Black mucky peat, common medium vegetal fibers. Clear boundary 43 10YR2/1 Black sandy muck with common fine vegetal fibers. Gradual boundary. 75 10YR3/1 Very dark gray mucky sand with common fine to medium vegetal fibers. Clear boundary. 85 10YR4/1 Dark gray medium sand. Clear boundary. 94 10YR6/2 Light brownish gray medium to coarse sand. Clear boundary. 108 10YR3/1 Gray medium sand. Clear boundary 125+ 10YR6/2 Light brownish gray medium to coarse sand.
296 Table A 4. Description of vibracore BC03 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 54 10YR2/1 Black mucky peat with common tine to medium vegetal fibers. Gradual boundary. 88 10YR2/1 Black sandy muck with common fine to medium vegetal fibers. Clear boundary. 106+ 10YR3/2 Very dark grayish brown medium sand mottled with dark grayish brown (10YR4/2) medium sand. Table A 5. Description of vibracore EF01 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 11 10YR2.5/1 Black muck with common fine roots and vegetal fibers. Clear boundary. 19 10YR5/1 Gray medium sand mottled with black (10YR2.5/1) muck. Clear boundary. 60 10YR7/2 Light gray medium sand mottled with grayish brown (10YR5.2) medium sand. Common fine roots. Gradual boundary. 102 10YR8/1 White medium sand, slightly coarser than above. Common fine to medium roots. Gradual boundary. 126 10YR8/2 Very pale brown medium to fine sand. Few fine roots. Diffuse boundary. 143 10YR8/2 Very pale brown medium to fine sand, slightly coarser than above. Common fine roots. Diffuse boundary. 170 10YR8/2 Very pale brown medium sand. Lightens with depth to 10YR8/4. Clear boundary. 183 10YR8/2 Very pale brown medium to fine sand. Diffuse boundary. 205 10YR7/4 Very pale brown medium sand. Gradual boundary. 223 10YR8/4 Very pale brown medium sand. Diffuse boundary. 230+ 10YR7/6 Yellow medium sand.
297 Table A 6. Description of vibracore EF02 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 15 10YR2/1 Black mucky peat with common fine to medium vegetal fibers. Clear boundary. 29 10YR2/1 Black peat with abundant fine to medium vegetal fibers. Clear boundary. 35 10YR2/1 Black mucky peat with common fine to medium vegetal fibers. Clear boundary. 41 7.5YR2.5/1 Black sandy muck with abundant fine vegetal fibers. Clear boundary. 47 7.5YR2.5/1 Black sandy muck with abundant fine vegetal fibers. Mottled with gray to light gray (10yr5/1 10YR7/1) fine sand. Clear bounda ry. 52 10YR4/1 10YR5/1 Dark gray grading to gray medium sand. Clear boundary. 166 10YR6/2 10YR8/2 Light brownish gray medium to fine sand. Grades with depth to Very pale brown fine sand. Gradual boundary. 180 10YR7/4 Very pale brown fine sand with common yellowish brown (10YR5/6) redox concentrations. Clear boundary. 198 10YR7/2 Light gray medium sand, mottled with very pale brown (10YR7/4) fine sand. Few yellowish brown (10YR5/6) redox concentrations. 215+ 10YR7/4 Very pale brown mottled with light yellowish brown (10YR6/4) medium loamy sand. Dark gray (10YR4/1) striations may be clay lamellae.
298 Table A 7. Description of vibracore EF03 at Silver Glen Springs (8LA1 W) Depth below top of core (cm) Munsell Color Description 36 5YR2.5/1 Black peat. Compact, with many fine to medium roots and vegetal fibers. Clear boundary. 69 5YR2.5/1 Black mucky peat. Compact, with common fine roots and vegetal fibers. Clear boundary. 80 5YR3/2 Dark reddish brown mucky peat. Softer and spongier than above, with common fine roots and vegetal fibers. Clear boundary. 123 5YR3/1 Very dark gray muck, grading to black (10YR2.5/1) with depth. Soft, with few fine roots and common fine vegetal fibers. Coarse white (10YR8/1) sand stringers intervene. 131 N/A VOID shrinkage of below with drying. 136 10YR2.5/1 Black muck. Finer than above with few visible fibers. 142 10YR2.5/1 Black mucky medium sand, mottled with gray (10YR5/1) and very pale brown (10YR7/4) medium to coarse sand. Clear boundary. 157 10YR7/1 Light gray mottled with dark gray (10YR4/1) medium sand and OM staining. Clear boundary. 165 10YR3/2 Very dark grayish brown medium sand with splotchy OM staining. Clear boundary. 185 10YR6/2 Light brownish gray medium sand. Clear boundary. 187 10YR3/1 Very dark gray mucky medium sand with visible roots and leaves. Possible buried surface. 197+ 10YR4/1 Dark gray medium sand mottled with gray (10YR6/1) medium sand.
299 APPENDIX B SILVER SPRINGS LITHIC ARTIFACT ATTRIBUTES
300 Table B 1. Lithic debitage provenance and attribute data Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 21.01 SILICIFIED CORAL C BTF 1.5 0.2 0 Y 8MR93 21.01 OCALA MD BTF 4.5 13.2 0 N 8MR93 21.01 OCALA C BTF 3 3.3 0 N 8MR93 21.01 OCALA C BTF 2 1.6 0 N 8MR93 21.01 OCALA MD BTF 2.5 1 0 N 8MR93 21.01 OCALA P BTF 2.5 1.5 0 N 8MR93 21.01 OCALA P BTF 2 1.8 0 N 8MR93 21.01 OCALA P BTF 2.5 0.8 0 N 8MR93 21.01 OCALA C BTF 2 0.7 0 N 8MR93 21.01 OCALA P BTF 2 1 0 N 8MR93 21.01 OCALA C BTF 1.5 0.2 0 N 8MR93 21.01 OCALA C BTF 2 1.1 0 N 8MR93 21.01 OCALA C BTF 1.5 0.2 0 Y 8MR93 21.01 OCALA MD BTF 2 0.7 0 N 8MR93 21.01 OCALA P BTF 2 0.4 0 N 8MR93 21.01 OCALA C BTF 2 0.7 0 Y 8MR93 21.01 OCALA MD BTF 1.5 0.7 0 N 8MR93 21.01 OCALA P BTF 2 0.7 0 N 8MR93 21.01 OCALA P BTF 1.5 0.4 0 N 8MR93 21.01 OCALA P BTF 2 0.7 0 N 8MR93 21.01 OCALA C BTF 1.5 0.1 0 N 8MR93 21.01 OCALA C BTF 1.5 0.2 0 N 8MR93 21.01 OCALA P BTF 1.5 0.4 0 N 8MR93 21.01 OCALA C BTF 1 0.1 0 Y 8MR93 21.01 OCALA MD I 1.5 0.2 0 N 8MR93 21.01 OCALA MD BTF 1.5 0.1 0 N 8MR93 21.01 OCALA MD BTF 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.2 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA P I 1 0.1 0 N 8MR93 21.01 OCALA NO I 1 0.2 0 N
301 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 21.01 OCALA P I 1 0.1 0 N 8MR93 21.01 OCALA P I 1 0.1 0 N 8MR93 21.01 OCALA C PR 1 0.1 0 Y 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 Y 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 OCALA MD I 1 0.1 0 N 8MR93 21.01 SANTA FE MD I 1 0.1 0 N 8MR93 21.01 SUWANNEE LIMESTONE FORMATION MD I 1.5 0.2 0 N 8MR93 21.01 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 Y 8MR93 21.01 OCALA C BTF 1.5 0.3 0 Y 8MR93 21.01 SANTA FE MD BTF 2.5 1.6 0 Y 8MR93 21.01 SANTA FE MD I 2 0.9 0 N 8MR93 21.01 SUWANNEE LIMESTONE FORMATION MD I 1.5 1 5 N 8MR93 174.04 TAMPA LIMESTONE FORMATION C BTF 2 1.7 4 Y 8MR93 174.04 SANTA FE MD BTF 2.5 1.4 4 N 8MR93 174.04 SANTA FE MD BTF 2 0.5 0 I 8MR93 174.04 SANTA FE MD BTF 1 0.1 0 N 8MR93 174.04 SANTA FE MD I 1 0.1 0 N 8MR93 174.04 UPPER WITHLACOOCHEE RIVER P BTF 1.5 0.4 0 Y 8MR93 174.04 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.1 0 Y 8MR93 174.04 OCALA P BTF 3 2 0 Y
302 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 174.04 OCALA P BTF 2 1.6 0 N 8MR93 174.04 OCALA MD BTF 2.5 1.3 0 Y 8MR93 174.04 OCALA C BTF 2.5 1.2 0 N 8MR93 174.04 OCALA P BTF 1.5 0.2 0 Y 8MR93 174.04 OCALA P BTF 1.5 0.5 0 N 8MR93 174.04 OCALA MD BTF 2 0.1 0 Y 8MR93 174.04 OCALA C BTF 1.5 0.2 0 Y 8MR93 174.04 OCALA MD I 1 0.1 0 I 8MR1082 336.01 SILICIFIED CORAL MD BTF 1.5 0.2 0 Y 8MR1082 336.01 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.1 0 Y 8MR1082 336.01 OCALA P BTF 4 8.4 0 N 8MR1082 336.01 OCALA NO I 2 2 0 Y 8MR1082 336.01 OCALA C BTF 2 0.9 0 N 8MR1082 336.01 OCALA C BTF 1.5 0.6 0 N 8MR1082 336.01 OCALA P BTF 1.5 0.5 0 N 8MR1082 336.01 OCALA P I 1.5 0.3 0 Y 8MR1082 336.01 OCALA MD BTF 2 0.7 1 N 8MR1082 336.01 OCALA MD I 2 0.3 0 N 8MR1082 336.01 OCALA C BTF 1.5 0.2 0 N 8MR1082 336.01 OCALA P BTF 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1.5 0.3 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA C BTF 1.5 0.2 0 N 8MR1082 336.01 OCALA C BTF 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.2 0 Y 8MR1082 336.01 OCALA C BTF 1 0.1 4 N 8MR1082 336.01 OCALA MD BTF 1.5 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA C BTF 1 0.1 0 N 8MR1082 336.01 OCALA P BTF 1.5 0.1 0 N
303 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR1082 336.01 OCALA P BTF 1 0.1 0 N 8MR1082 336.01 OCALA P I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA P I 1 0.1 0 N 8MR1082 336.01 OCALA C BTF 1 0.2 0 N 8MR1082 336.01 OCALA C BTF 1 0.1 0 N 8MR1082 336.01 OCALA C PR 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR1082 336.01 OCALA MD I 1 0.1 0 N 8MR93 514.01 SILICIFIED CORAL C BTF 2 0.2 0 Y 8MR93 514.01 SILICIFIED CORAL MD BTF 1.5 0.1 0 Y 8MR93 514.01 SILICIFIED CORAL P BTF 1 0.1 0 Y 8MR93 514.01 TAMPA LIMESTONE FORMATION MD I 2 0.6 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 3 2.2 0 N 8MR93 514.01 SANTA FE C BTF 2 0.9 0 N 8MR93 514.01 UPPER WITHLACOOCHEE RIVER P BTF 3 1.7 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2.5 2.8 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2.5 1.4 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2 1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER C BTF 2 0.3 0 N 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2 0.9 1 Y
304 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 514.01 UPPER WITHLACOOCHEE RIVER P BTF 2 0.9 0 N 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2 0.5 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2.5 0.7 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.2 0 N 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.3 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 2 0.3 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1 0.1 0 N 8MR93 514.01 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.2 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD BTF 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER C NF 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER C PR 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 Y 8MR93 514.01 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 514.01 OCALA P CR 5 19.8 3 N 8MR93 514.01 OCALA P CR 4.5 15.6 2 N 8MR93 514.01 OCALA NO AS 1.5 1.6 0 N
305 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 514.01 OCALA P BTF 2.5 3.8 1 Y 8MR93 514.01 OCALA C BTF 2 1.9 0 Y 8MR93 514.01 OCALA MD BTF 3 2.8 0 Y 8MR93 514.01 OCALA NO AS 1.5 1.5 0 N 8MR93 514.01 OCALA MD I 2 1.5 0 N 8MR93 514.01 OCALA MD I 2 1.3 0 N 8MR93 514.01 OCALA C BTF 2 1.2 0 N 8MR93 514.01 OCALA P BTF 2 1 0 N 8MR93 514.01 OCALA P BTF 1.5 0.5 0 N 8MR93 514.01 OCALA MD BTF 2 0.4 0 Y 8MR93 514.01 OCALA P BTF 2 0.4 0 N 8MR93 514.01 OCALA MD I 1.5 0.4 0 N 8MR93 514.01 OCALA MD BTF 2.5 0.6 0 N 8MR93 514.01 OCALA C BTF 2 0.3 0 Y 8MR93 514.01 OCALA MD I 1.5 0.4 0 Y 8MR93 514.01 OCALA P BTF 1.5 0.4 0 Y 8MR93 514.01 OCALA MD I 1.5 0.1 0 Y 8MR93 514.01 OCALA P BTF 1.5 0.1 0 Y 8MR93 514.01 OCALA MD I 1.5 0.1 0 N 8MR93 514.01 OCALA C BTF 1.5 0.1 0 Y 8MR93 514.01 OCALA MD BTF 1.5 0.1 0 N 8MR93 514.01 OCALA MD I 1 0.1 0 N 8MR93 514.01 OCALA P BTF 1.5 0.1 0 N 8MR93 514.01 OCALA MD I 1 0.1 0 N 8MR93 514.01 OCALA MD I 1.5 0.1 0 N 8MR93 514.01 OCALA MD I 1.5 0.1 0 N 8MR93 514.01 OCALA MD I 1.5 0.1 0 N 8MR93 514.01 OCALA MD I 1 0.1 0 Y 8MR93 514.01 OCALA MD I 1 0.2 0 N 8MR93 514.01 OCALA P BTF 1 0.2 0 Y 8MR93 514.01 OCALA MD I 1 0.1 0 N 8MR93 185.02 SILICIFIED CORAL MD BTF 1.5 0.9 0 N
306 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 185.02 SUWANNEE LIMESTONE FORMATION P BTF 1.5 0.4 0 N 8MR93 185.02 OCALA MD I 1.5 0.4 0 N 8MR93 185.02 OCALA C BTF 3.5 2.9 0 Y 8MR93 185.02 OCALA C BTF 3.5 3.6 0 N 8MR93 185.02 OCALA C BTF 2 1.5 0 Y 8MR93 185.02 OCALA MD I 1.5 0.4 0 Y 8MR93 185.02 OCALA NO AS 2 1.9 5 N 8MR93 185.02 OCALA P BTF 1.5 0.4 0 N 8MR93 185.02 OCALA C BTF 1.5 0.5 0 N 8MR93 185.02 OCALA P BTF 1.5 0.3 0 N 8MR93 185.02 OCALA P BTF 1 0.1 0 Y 8MR93 185.02 OCALA MD I 1 0.1 0 Y 8MR93 185.02 OCALA C PR 1 0.1 0 Y 8MR93 225.02 SILICIFIED CORAL MD BTF 2 0.2 0 Y 8MR93 225.02 SILICIFIED CORAL MD I 1 0.05 0 Y 8MR93 225.02 OCALA C CR 7 63.7 4 N 8MR93 225.02 OCALA P BTF 4 11.2 0 N 8MR93 225.02 OCALA P BTF 3 5.1 2 N 8MR93 225.02 OCALA P BTF 3 5.9 0 N 8MR93 225.02 OCALA P BTF 3 3.2 0 N 8MR93 225.02 OCALA P BTF 3 2.8 0 Y 8MR93 225.02 OCALA C BTF 3 4.2 0 N 8MR93 225.02 OCALA MD I 2.5 1.7 0 N 8MR93 225.02 OCALA P BTF 2.5 1.1 0 Y 8MR93 225.02 OCALA MD I 2.5 0.9 0 N 8MR93 225.02 OCALA MD I 2.5 1.3 0 N 8MR93 225.02 OCALA MD I 2.5 1.3 0 N 8MR93 225.02 OCALA MD I 2.5 1.3 0 Y 8MR93 225.02 OCALA C I 2 0.7 0 N 8MR93 225.02 OCALA MD I 2 0.7 0 N 8MR93 225.02 OCALA C BTF 2 0.5 0 Y
307 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 225.02 OCALA MD I 2 0.7 0 N 8MR93 225.02 OCALA C BTF 2 0.8 0 N 8MR93 225.02 OCALA P BTF 2 0.3 0 N 8MR93 225.02 OCALA C BTF 2.5 0.4 0 N 8MR93 225.02 OCALA P BTF 2 0.4 0 N 8MR93 225.02 OCALA MD I 2 0.8 0 N 8MR93 225.02 OCALA P BTF 1.5 0.05 0 Y 8MR93 225.02 OCALA MD I 1.5 0.2 0 N 8MR93 225.02 OCALA P I 1.5 0.3 0 N 8MR93 225.02 OCALA MD BTF 2 0.5 0 N 8MR93 225.02 OCALA MD I 1.5 0.05 0 N 8MR93 225.02 OCALA P NF 1 0.05 0 N 8MR93 225.02 OCALA MD I 2 0.8 0 N 8MR93 225.02 OCALA MD I 2.5 1.4 0 N 8MR93 225.02 OCALA MD I 2 1.2 0 N 8MR93 225.02 OCALA MD I 1.5 0.3 3 N 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA C BTF 2 0.8 0 Y 8MR93 225.02 OCALA C I 1.5 0.3 0 N 8MR93 225.02 OCALA MD I 2 0.6 0 N 8MR93 225.02 OCALA MD I 1.5 0.1 0 Y 8MR93 225.02 OCALA MD I 1.5 0.4 0 N 8MR93 225.02 OCALA P BTF 1.5 0.2 0 Y 8MR93 225.02 OCALA P BTF 1.5 0.3 0 Y 8MR93 225.02 OCALA P BTF 1.5 0.2 0 Y 8MR93 225.02 OCALA NO I 1 0.1 0 Y 8MR93 225.02 OCALA NO TS 1 0.1 0 Y 8MR93 225.02 OCALA MD I 1.5 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 2 1.1 3 N
308 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 225.02 OCALA MD I 2 0.2 0 Y 8MR93 225.02 OCALA MD BTF 1.5 0.05 0 N 8MR93 225.02 OCALA MD BTF 1.5 0.3 0 Y 8MR93 225.02 OCALA MD I 1.5 0.2 0 N 8MR93 225.02 OCALA P BTF 1.5 0.1 0 Y 8MR93 225.02 OCALA MD I 1.5 0.3 0 N 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1.5 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA MD AS 1 0.2 3 N 8MR93 225.02 OCALA P I 1.5 0.2 0 N 8MR93 225.02 OCALA NO AS 1 0.3 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA C BTF 1 0.05 1 N 8MR93 225.02 OCALA P BTF 1 0.05 0 Y 8MR93 225.02 OCALA MD I 1 0.1 0 Y 8MR93 225.02 OCALA C BTF 1 0.05 0 N 8MR93 225.02 OCALA P I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1.5 0.1 0 N 8MR93 225.02 OCALA P I 1.5 0.05 0 N 8MR93 225.02 OCALA NO TS 1 0.05 0 Y 8MR93 225.02 OCALA C I 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA NO AS 1 0.1 0 Y 8MR93 225.02 OCALA MD I 1 0.05 0 Y 8MR93 225.02 OCALA MD I 1 0.1 0 N
309 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.2 0 N 8MR93 225.02 OCALA P BTF 1 0.05 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA NO I 1 0.1 0 N 8MR93 225.02 OCALA P I 1 0.05 0 N 8MR93 225.02 OCALA P I 1 0.1 0 Y 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.1 0 N 8MR93 225.02 OCALA MD I 1 0.05 0 N 8MR93 225.02 OCALA NO AS 1.5 0.3 5 N 8MR93 225.02 OCALA MD I 1.5 0.2 4 N 8MR93 225.02 OCALA C CR 1.5 0.4 5 N 8MR93 225.02 OCALA NO AS 1 0.2 3 N 8MR93 225.02 OCALA MD I 1 0.1 3 N 8MR93 225.02 OCALA MD I 1 0.1 3 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1.5 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 3.5 5.1 1 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP C BTF 3 4.4 2 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD BTF 7.5 56.3 2 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 2 0.3 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.1 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.5 0 Y
310 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1.5 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER C BTF 1 0.1 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER NO AS 0.5 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.05 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1 0.05 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 0.5 0.05 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 2 1.5 0 N 8MR93 225.02 SANTA FE C BTF 2.5 2.7 0 N 8MR93 225.02 SUWANNEE LIMESTONE FORMATION MD I 2.5 0.9 5 N 8MR93 225.02 HILLSBOROUGH RIVER MD BTF 1.5 0.1 0 Y 8MR93 225.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 225.02 SANTA FE MD I 1 0.1 0 Y 8MR93 225.02 SANTA FE MD I 1 0.1 0 Y 8MR93 225.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 Y 8MR93 225.02 SANTA FE MD I 1 0.1 0 Y
311 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 225.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER NO TS 1 0.1 0 Y 8MR93 225.02 OCALA P BTF 1.5 0.4 2 N 8MR93 225.02 OCALA MD I 1.5 0.4 0 N 8MR93 225.02 OCALA MD I 1.5 0.2 0 N 8MR93 225.02 OCALA MD I 1.5 0.2 0 Y 8MR93 225.02 OCALA P BTF 1 0.1 0 Y 8MR93 225.02 OCALA P I 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 1.5 0.4 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1.5 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.2 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.2 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 1.5 0.2 0 Y 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.2 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 225.02 UPPER WITHLACOOCHEE RIVER P BTF 0.5 0.2 0 Y 8MR93 225.02 INDETERMINATE P BTF 0.5 0.1 0 Y
312 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 330.02 SILICIFIED CORAL C BTF 2 0.5 0 N 8MR93 330.02 TAMPA LIMESTONE FORMATION MD BTF 3 2.2 0 Y 8MR93 330.02 SUWANNEE LIMESTONE FORMATION MD BTF 2 0.5 0 Y 8MR93 330.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 I 8MR93 330.02 UPPER WITHLACOOCHEE RIVER C BTF 2 0.9 0 Y 8MR93 330.02 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.1 0 N 8MR93 330.02 UPPER WITHLACOOCHEE RIVER P NF 1 0.1 0 N 8MR93 330.02 OCALA P BTF 3.5 7.2 0 N 8MR93 330.02 OCALA MD BTF 2.5 1.3 0 N 8MR93 330.02 OCALA MD BTF 2 0.6 0 Y 8MR93 330.02 OCALA MD BTF 2 0.5 0 N 8MR93 330.02 OCALA MD BTF 2 0.4 1 N 8MR93 330.02 OCALA C CR 2 1.1 0 N 8MR93 330.02 OCALA MD I 2 0.8 0 Y 8MR93 330.02 OCALA MD BTF 2 1.4 0 Y 8MR93 330.02 OCALA P BTF 2 1.1 0 N 8MR93 330.02 OCALA P BTF 2.5 1.5 0 N 8MR93 330.02 OCALA MD BTF 1.5 0.5 0 Y 8MR93 330.02 OCALA P BTF 1.5 0.3 0 Y 8MR93 330.02 OCALA P BTF 1.5 0.3 0 N 8MR93 330.02 OCALA MD I 1.5 0.3 0 N 8MR93 330.02 OCALA P BTF 1.5 0.2 0 N 8MR93 330.02 OCALA C BTF 1.5 0.3 0 Y 8MR93 330.02 OCALA MD BTF 1.5 0.3 0 N 8MR93 330.02 OCALA NO TS 1.5 0.4 0 Y 8MR93 330.02 OCALA NO AS 1.5 0.5 0 N 8MR93 330.02 OCALA C BTF 1.5 0.4 1 Y
313 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 330.02 OCALA P I 1.5 0.2 0 N 8MR93 330.02 OCALA MD BTF 2 0.3 0 Y 8MR93 330.02 OCALA P BTF 1.5 0.2 0 N 8MR93 330.02 OCALA NO TS 1 0.1 0 Y 8MR93 330.02 OCALA P BTF 1.5 0.1 0 Y 8MR93 330.02 OCALA P BTF 1 0.1 0 Y 8MR93 330.02 OCALA MD I 1.5 0.1 0 N 8MR93 330.02 OCALA MD I 1.5 0.1 0 Y 8MR93 330.02 OCALA P BTF 1.5 0.2 0 N 8MR93 330.02 OCALA P BTF 1.5 0.1 0 N 8MR93 330.02 OCALA P PR 1 0.1 0 Y 8MR93 330.02 OCALA P PR 1 0.1 0 N 8MR93 330.02 OCALA MD I 1 0.1 0 Y 8MR93 14401.08 GAINESVILLE MD BTF 2 0.6 0 Y 8MR93 14401.08 OCALA MD BTF 4 9.6 3 Y 8MR93 14401.08 OCALA MD BTF 3 1.9 0 Y 8MR93 14401.08 OCALA MD BTF 2 2 2 Y 8MR93 14401.08 OCALA P BTF 2 0.7 0 Y 8MR93 14401.08 OCALA MD I 2.5 0.7 0 Y 8MR93 14401.08 OCALA MD BTF 2 0.6 0 Y 8MR93 14401.08 OCALA MD BTF 2 1.3 0 I 8MR93 14401.08 OCALA MD I 1.5 0.5 0 Y 8MR93 14401.08 OCALA C BTF 1.5 0.2 0 Y 8MR93 14401.08 OCALA MD I 1.5 0.4 0 N 8MR93 14401.08 OCALA P BTF 2 0.3 0 Y 8MR93 14401.08 OCALA P BTF 1.5 0.4 0 Y 8MR93 14401.08 OCALA P BTF 2 0.4 0 N 8MR93 14401.08 OCALA P BTF 1.5 0.1 0 Y 8MR93 14401.08 OCALA MD I 1.5 0.3 0 N 8MR93 14401.08 OCALA C BTF 1 0.1 0 Y 8MR93 14401.08 OCALA MD BTF 1.5 1 0 Y
314 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 14401.08 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 1.5 0.8 2 Y 8MR93 14401.08 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP NO AS 1.5 1.1 0 Y 8MR93 14401.08 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP NO AS 1.5 0.6 0 Y 8MR93 14401.08 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.2 0 Y 8MR93 14401.08 UPPER WITHLACOOCHEE RIVER P PR 1 0.1 0 Y 8MR93 14401.08 SILICIFIED CORAL MD BTF 2 0.6 0 N 8MR93 14402.02 SUWANNEE LIMESTONE FORMATION C NF 1 0.1 0 N 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER P BTF 2 0.8 0 Y 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER C BTF 2.5 1.3 0 Y 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER MD BTF 3.5 2.7 1 N 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER MD I 2.5 2.4 0 N 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER MD BTF 2 0.4 0 Y 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.1 0 Y 8MR93 14402.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 14402.02 OCALA C BTF 4.5 7.5 0 N 8MR93 14402.02 OCALA C BTF 4 6.2 0 N 8MR93 14402.02 OCALA MD BTF 3 4.6 0 N 8MR93 14402.02 OCALA C BTF 3 4.4 2 Y 8MR93 14402.02 OCALA MD I 3.5 5 2 N
315 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 14402.02 OCALA MD I 2.5 2.5 3 N 8MR93 14402.02 OCALA NO TS 2.5 2 0 Y 8MR93 14402.02 OCALA MD BTF 2 0.7 0 N 8MR93 14402.02 OCALA MD BTF 2 0.9 0 N 8MR93 14402.02 OCALA C BTF 2.5 0.4 0 N 8MR93 14402.02 OCALA MD BTF 2 0.5 0 N 8MR93 14402.02 OCALA C BTF 2 0.6 0 N 8MR93 14402.02 OCALA MD BTF 1.5 0.5 2 Y 8MR93 14402.02 OCALA P BTF 2 1.1 2 N 8MR93 14402.02 OCALA P BTF 2 0.8 0 N 8MR93 14402.02 OCALA P BTF 1.5 0.2 0 N 8MR93 14402.02 OCALA C BTF 2 0.2 0 N 8MR93 14402.02 OCALA MD BTF 1.5 0.1 0 Y 8MR93 14402.02 OCALA P BTF 1.5 0.3 0 Y 8MR93 14402.02 OCALA P BTF 1.5 0.4 0 Y 8MR93 14402.02 OCALA P BTF 2 0.7 0 N 8MR93 14402.02 OCALA MD I 1.5 0.6 0 N 8MR93 14402.02 OCALA MD I 2 0.3 0 N 8MR93 14402.02 OCALA MD I 2 0.4 0 N 8MR93 14402.02 OCALA P BTF 1.5 0.2 0 Y 8MR93 14402.02 OCALA P BTF 1 0.1 0 N 8MR93 14402.02 OCALA P BTF 1.5 0.2 0 N 8MR93 14402.02 OCALA P BTF 1 0.2 0 Y 8MR93 14402.02 OCALA MD I 1.5 0.3 0 Y 8MR93 14402.02 OCALA NO I 1.5 1.4 5 N 8MR93 14402.02 OCALA MD BTF 1.5 0.1 0 Y 8MR93 14402.02 OCALA MD I 1.5 0.4 1 N 8MR93 14402.02 OCALA MD I 1.5 0.4 1 N 8MR93 14402.02 OCALA P BTF 1.5 0.3 0 N 8MR93 14402.02 OCALA MD BTF 1 0.1 0 Y 8MR93 14402.02 OCALA P BTF 2 0.1 0 Y 8MR93 14402.02 OCALA MD I 1 0.1 0 N
316 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 14402.02 OCALA MD I 1 0.1 0 N 8MR93 14402.02 OCALA P I 1 0.1 0 Y 8MR93 14402.02 OCALA MD I 1 0.1 0 N 8MR93 14402.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 SILICIFIED CORAL P CR 3 4.5 0 N 8MR93 7802.02 SILICIFIED CORAL NO AS 1.5 0.4 0 N 8MR93 7802.02 SILICIFIED CORAL MD I 1 0.1 5 N 8MR93 7802.02 OCALA C BTF 3.5 10.6 2 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 3 2 0 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.3 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.2 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.1 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.1 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.1 0 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 Y 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.1 0 Y
317 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER P I 1 0.1 0 N 8MR93 7802.02 UPPER WITHLACOOCHEE RIVER C PR 1 0.1 0 Y 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 2 0.5 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 2 0.6 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION P BTF 2 0.6 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION C CR 2 0.5 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 2 0.2 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION P BTF 1 0.1 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION C I 1 0.1 0 Y 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION P PR 1 0.1 0 Y 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 Y 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 7802.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 7802.02 OCALA C BTF 4.5 13.5 3 Y 8MR93 7802.02 OCALA C BTF 4.5 8 0 Y 8MR93 7802.02 OCALA C BTF 3.5 5.8 0 Y 8MR93 7802.02 OCALA MD I 3 6.1 1 Y 8MR93 7802.02 OCALA MD I 3 4.2 5 N 8MR93 7802.02 OCALA P BTF 3 3 0 Y 8MR93 7802.02 OCALA MD I 2 2 0 Y
318 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7802.02 OCALA C BTF 4 5.6 0 N 8MR93 7802.02 OCALA MD BTF 3.5 2.8 0 Y 8MR93 7802.02 OCALA MD BTF 3 1.7 0 N 8MR93 7802.02 OCALA P BTF 3 5.4 3 N 8MR93 7802.02 OCALA C BTF 3.5 3.8 2 N 8MR93 7802.02 OCALA P BTF 2.5 2.5 0 Y 8MR93 7802.02 OCALA C BTF 3 2.5 4 Y 8MR93 7802.02 OCALA C BTF 3 2.5 1 Y 8MR93 7802.02 OCALA MD BTF 2.5 1.4 0 Y 8MR93 7802.02 OCALA P BTF 2.5 1.2 0 Y 8MR93 7802.02 OCALA MD BTF 2.5 1.3 3 Y 8MR93 7802.02 OCALA MD BTF 2.5 1.7 0 N 8MR93 7802.02 OCALA P BTF 2 1.4 0 Y 8MR93 7802.02 OCALA MD I 2 1 0 N 8MR93 7802.02 OCALA MD I 2 1.4 0 Y 8MR93 7802.02 OCALA P BTF 2 1.3 0 N 8MR93 7802.02 OCALA P BTF 2 0.7 0 Y 8MR93 7802.02 OCALA C BTF 2 0.6 0 N 8MR93 7802.02 OCALA MD I 2 2.1 1 N 8MR93 7802.02 OCALA P BTF 2 0.5 0 N 8MR93 7802.02 OCALA C BTF 2 0.2 0 N 8MR93 7802.02 OCALA MD I 1.5 0.7 0 Y 8MR93 7802.02 OCALA P BTF 2 0.3 0 Y 8MR93 7802.02 OCALA P BTF 1.5 0.5 0 Y 8MR93 7802.02 OCALA MD I 2 0.8 0 N 8MR93 7802.02 OCALA MD I 2 1.2 5 N 8MR93 7802.02 OCALA MD BTF 2 0.3 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.3 0 N 8MR93 7802.02 OCALA C BTF 2 0.2 0 Y 8MR93 7802.02 OCALA P BTF 1.5 0.2 0 Y 8MR93 7802.02 OCALA MD I 1.5 0.9 0 N
319 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7802.02 OCALA C BTF 1.5 0.5 0 N 8MR93 7802.02 OCALA P BTF 1.5 0.2 0 Y 8MR93 7802.02 OCALA P BTF 1 0.2 0 Y 8MR93 7802.02 OCALA MD I 1.5 0.1 0 N 8MR93 7802.02 OCALA P I 1.5 0.4 0 N 8MR93 7802.02 OCALA MD I 1.5 0.4 0 Y 8MR93 7802.02 OCALA P BTF 1.5 0.4 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 Y 8MR93 7802.02 OCALA P BTF 1.5 0.1 0 Y 8MR93 7802.02 OCALA MD I 1.5 0.2 0 Y 8MR93 7802.02 OCALA MD I 1.5 0.2 0 Y 8MR93 7802.02 OCALA P I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 Y 8MR93 7802.02 OCALA MD I 1.5 0.1 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA P I 1 0.2 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 N 8MR93 7802.02 OCALA P BTF 1 0.1 0 Y 8MR93 7802.02 OCALA P BTF 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA P PR 1 0.1 0 N 8MR93 7802.02 OCALA C BTF 1 0.1 0 Y 8MR93 7802.02 OCALA NO TS 1 0.1 0 Y 8MR93 7802.02 OCALA C I 1.5 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA C I 1 0.1 0 N
320 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7802.02 OCALA P I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 1 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 Y 8MR93 7802.02 OCALA C I 1 0.1 0 N 8MR93 7802.02 OCALA NO I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.4 0 N 8MR93 7802.02 OCALA MD I 1.5 0.2 0 N 8MR93 7802.02 OCALA NO AS 1.5 0.4 0 Y 8MR93 7802.02 OCALA NO I 1.5 0.3 5 N 8MR93 7802.02 OCALA P I 1.5 0.1 0 N 8MR93 7802.02 OCALA P I 1.5 0.2 0 N 8MR93 7802.02 OCALA C I 1.5 0.1 0 Y 8MR93 7802.02 OCALA P I 1.5 0.1 0 N 8MR93 7802.02 OCALA MD I 1.5 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA P BTF 2 0.1 0 N 8MR93 7802.02 OCALA P I 1 0.2 0 N 8MR93 7802.02 OCALA P BTF 1.5 0.2 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA C BTF 1 0.2 0 Y 8MR93 7802.02 OCALA P I 1 0.1 0 N 8MR93 7802.02 OCALA P BTF 1 0.1 0 N 8MR93 7802.02 OCALA P NF 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA C PR 1 0.05 0 N 8MR93 7802.02 OCALA MD I 1 0.1 0 N 8MR93 7802.02 OCALA P BTF 1 0.1 0 Y 8MR93 7802.02 OCALA MD I 1 0.1 0 Y 8MR93 7802.02 OCALA P I 1 0.1 0 N 8MR93 7802.02 OCALA P BTF 1 0.05 0 N
321 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7802.02 OCALA NO I 1 0.05 0 N 8MR93 7802.02 OCALA MD I 1 0.05 0 N 8MR93 7802.02 OCALA P I 1 0.05 0 N 8MR93 7802.02 OCALA MD I 1 0.05 0 N 8MR93 7802.02 OCALA P I 1 0.1 0 N 8MR93 7402.03 SILICIFIED CORAL MD I 2 0.4 0 I 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP C BTF 1.5 0.2 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP C BTF 1.5 0.1 0 N 8MR93 7402.03 SUWANNEE LIMESTONE FORMATION MD BTF 3 2 0 N 8MR93 7402.03 SUWANNEE LIMESTONE FORMATION MD I 2.5 1.2 0 N 8MR93 7402.03 SUWANNEE LIMESTONE FORMATION NO AS 1.5 0.5 0 N 8MR93 7402.03 SUWANNEE LIMESTONE FORMATION MD I 1.5 0.1 0 N 8MR93 7402.03 SUWANNEE LIMESTONE FORMATION C I 1 0.1 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER C BTF 3 3.2 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER C BTF 5 4.6 0 Y 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER P BTF 3 1.2 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER P BTF 2.5 1.2 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER MD I 2 1.2 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.3 0 Y
322 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR93 7402.03 UPPER WITHLACOOCHEE RIVER P PR 1 0.1 0 N 8MR93 7402.03 OCALA MD CR 3 5.3 5 N 8MR93 7402.03 OCALA P BTF 3.5 4.8 0 N 8MR93 7402.03 OCALA C BTF 3 3.3 3 N 8MR93 7402.03 OCALA P BTF 3 1.8 0 N 8MR93 7402.03 OCALA P BTF 3 2.6 0 Y 8MR93 7402.03 OCALA P BTF 2.5 2 0 N 8MR93 7402.03 OCALA P BTF 3 1.9 0 N 8MR93 7402.03 OCALA C BTF 3 1.7 0 Y 8MR93 7402.03 OCALA P BTF 2.5 2.6 0 N 8MR93 7402.03 OCALA C BTF 2.5 1.2 0 N 8MR93 7402.03 OCALA P BTF 3 3.2 0 N 8MR93 7402.03 OCALA P BTF 2 0.8 0 N 8MR93 7402.03 OCALA MD BTF 2 0.9 0 N 8MR93 7402.03 OCALA MD I 2 0.7 0 N 8MR93 7402.03 OCALA MD BTF 2 0.5 0 N 8MR93 7402.03 SILICIFIED CORAL C BTF 1.5 0.2 0 Y 8MR93 7402.03 OCALA MD I 2 0.4 0 Y 8MR93 7402.03 OCALA P BTF 2 0.6 5 N 8MR93 7402.03 OCALA C BTF 2 0.4 0 N 8MR93 7402.03 OCALA MD BTF 2 0.6 0 N 8MR93 7402.03 OCALA P BTF 2 0.5 0 N 8MR93 7402.03 OCALA MD I 2 1.9 1 N 8MR93 7402.03 OCALA MD BTF 2 0.4 0 N 8MR93 7402.03 OCALA MD I 2 0.6 0 N 8MR93 7402.03 OCALA C BTF 2 0.4 0 N 8MR93 7402.03 OCALA C BTF 1.5 0.2 0 N 8MR93 7402.03 OCALA P BTF 2 0.2 0 N 8MR93 7402.03 OCALA MD I 2 0.4 0 N 8MR93 7402.03 OCALA MD I 1.5 0.1 0 N
323 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7402.03 OCALA MD I 1.5 0.3 0 N 8MR93 7402.03 OCALA MD I 1.5 0.5 0 N 8MR93 7402.03 OCALA MD I 1.5 0.2 0 N 8MR93 7402.03 OCALA P BTF 2 0.3 0 N 8MR93 7402.03 OCALA MD I 1.5 0.2 0 Y 8MR93 7402.03 OCALA P BTF 1.5 0.2 0 Y 8MR93 7402.03 OCALA P BTF 1.5 0.4 0 N 8MR93 7402.03 OCALA MD I 1.5 0.3 0 N 8MR93 7402.03 OCALA MD I 1.5 0.8 0 N 8MR93 7402.03 OCALA MD I 1.5 0.2 0 N 8MR93 7402.03 OCALA MD I 1.5 0.1 0 N 8MR93 7402.03 OCALA P BTF 1.5 0.2 0 N 8MR93 7402.03 OCALA P BTF 1.5 0.2 0 N 8MR93 7402.03 OCALA MD BTF 1.5 0.1 0 Y 8MR93 7402.03 OCALA MD I 1.5 0.2 0 N 8MR93 7402.03 OCALA P BTF 1 0.3 0 N 8MR93 7402.03 OCALA C BTF 1.5 0.2 0 Y 8MR93 7402.03 OCALA P I 1.5 0.2 0 Y 8MR93 7402.03 OCALA P I 1 0.1 0 N 8MR93 7402.03 OCALA P I 1 0.1 0 Y 8MR93 7402.03 OCALA P I 1.5 0.1 0 N 8MR93 7402.03 OCALA MD I 1.5 0.1 0 N 8MR93 7402.03 OCALA MD I 1.5 0.1 0 N 8MR93 7402.03 OCALA MD I 1.5 0.2 0 N 8MR93 7402.03 OCALA C I 1 0.1 0 N 8MR93 7402.03 OCALA C BTF 1 0.1 0 N 8MR93 7402.03 OCALA NO TS 1.5 0.1 0 Y 8MR93 7402.03 OCALA P PR 1 0.1 0 N 8MR93 7402.03 OCALA NO I 1 0.1 0 N 8MR93 7402.03 OCALA P I 1 0.1 0 N 8MR93 7402.03 OCALA MD I 1 0.1 0 N 8MR93 7402.03 OCALA P PR 1 0.1 0 N
324 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 7402.03 OCALA P BTF 1 0.1 0 N 8MR93 7402.03 OCALA P BTF 1 0.1 0 Y 8MR93 7402.03 OCALA P I 1 0.1 0 N 8MR93 7402.03 OCALA C PR 1 0.1 0 Y 8MR93 7402.03 OCALA MD I 1 0.1 0 N 8MR93 7402.03 OCALA MD I 1 0.1 0 N 8MR93 7402.03 OCALA P BTF 2 0.7 0 N 8MR93 7402.03 OCALA MD BTF 2 0.9 0 N 8MR93 414.02 OCALA P BTF 3.5 5.2 0 N 8MR93 414.02 OCALA P BTF 4 9.8 0 N 8MR93 414.02 OCALA NO TS 2 1 0 Y 8MR93 257.02 SANTA FE MD BTF 2 1 0 Y 8MR93 257.02 SANTA FE MD BTF 1.5 0.4 0 N 8MR93 257.02 SANTA FE MD BTF 2 0.8 0 N 8MR93 257.02 SANTA FE MD I 1.5 0.2 0 Y 8MR93 257.02 SANTA FE MD I 2 0.4 0 N 8MR93 257.02 SANTA FE MD I 1.5 0.4 0 N 8MR93 257.02 SANTA FE MD I 1.5 0.1 0 N 8MR93 257.02 OCALA P BTF 2 2 0 N 8MR93 257.02 OCALA C BTF 1.5 0.8 1 N 8MR93 257.02 OCALA C BTF 2 0.5 0 N 8MR93 257.02 OCALA MD BTF 1.5 0.3 0 N 8MR93 257.02 OCALA MD BTF 1.5 0.4 0 N 8MR93 257.02 OCALA MD I 2 0.5 0 N 8MR93 257.02 OCALA MD I 1.5 0.2 0 N 8MR93 257.02 OCALA C BTF 1.5 0.1 0 N 8MR93 257.02 OCALA NO I 2 0.6 0 N 8MR93 257.02 OCALA P BTF 1.5 0.7 0 N 8MR93 257.02 OCALA MD I 1.5 0.3 0 N 8MR93 257.02 OCALA C BTF 1.5 0.2 0 N 8MR93 257.02 OCALA MD I 1.5 0.1 0 N 8MR93 257.02 OCALA P BTF 1 0.1 4 N
325 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 257.02 OCALA MD I 1 0.1 5 N 8MR93 257.02 OCALA MD I 1.5 0.1 0 N 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 257.02 OCALA MD I 1.5 0.2 0 N 8MR93 257.02 OCALA NO I 1 0.2 0 N 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 257.02 OCALA P BTF 1 0.1 0 Y 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 257.02 OCALA MD I 1 0.1 0 N 8MR93 145.02 POSS. HILLSBOROUGH RIVER MD BTF 2 0.6 0 P 8MR93 145.02 SUWANNEE LIMESTONE FORMATION C BTF 1.5 0.1 0 Y 8MR93 145.02 SUWANNEE LIMESTONE FORMATION MD BTF 1.5 0.3 0 N 8MR93 145.02 OCALA C BTF 3 3.5 1 N 8MR93 145.02 OCALA MD BTF 2.5 1.2 0 N 8MR93 145.02 OCALA MD BTF 2.5 1.3 0 N 8MR93 145.02 OCALA MD I 1.5 0.3 0 Y 8MR93 145.02 OCALA P I 1.5 0.6 0 Y 8MR93 145.02 OCALA MD BTF 2 0.5 0 N 8MR93 145.02 OCALA C BTF 1.5 0.2 0 N 8MR93 145.02 OCALA NO TS 1 0.2 0 Y 8MR93 145.02 OCALA MD BTF 1.5 0.2 0 N 8MR1082 511.01 SANTA FE C BTF 5 12.4 4 N 8MR1082 511.01 SANTA FE NO AS 3.5 14.8 5 N 8MR1082 511.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 2 1.4 4 N 8MR1082 511.01 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.1 0 N
326 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR1082 511.01 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 N 8MR1082 511.01 OCALA MD BTF 4 5 0 N 8MR1082 511.01 OCALA P BTF 5 7.9 0 N 8MR1082 511.01 OCALA MD I 2.5 4.5 2 N 8MR1082 511.01 OCALA P BTF 3 3 0 N 8MR1082 511.01 OCALA P BTF 3 1.8 0 N 8MR1082 511.01 OCALA C BTF 2.5 1 0 N 8MR1082 511.01 OCALA P BTF 2 0.8 0 N 8MR1082 511.01 OCALA C BTF 2.5 1.4 0 N 8MR1082 511.01 OCALA MD BTF 2 0.4 0 N 8MR1082 511.01 OCALA MD I 2 0.8 0 Y 8MR1082 511.01 OCALA MD BTF 2 0.3 0 N 8MR1082 511.01 OCALA P BTF 2 0.5 0 N 8MR1082 511.01 OCALA P BTF 1.5 0.1 0 Y 8MR1082 511.01 OCALA MD I 1.5 0.1 0 N 8MR1082 511.01 OCALA MD BTF 2 0.3 0 N 8MR1082 511.01 OCALA C BTF 2 0.1 0 N 8MR1082 511.01 OCALA MD I 2 0.3 0 N 8MR1082 511.01 OCALA P BTF 1.5 0.2 0 N 8MR1082 511.01 OCALA P BTF 1.5 0.1 0 N 8MR1082 511.01 OCALA NO AS 1.5 0.3 0 N 8MR1082 511.01 OCALA C BTF 1 0.2 0 N 8MR1082 511.01 OCALA MD BTF 1.5 0.1 0 N 8MR1082 511.01 OCALA MD I 1 0.1 0 N 8MR1082 511.01 OCALA MD I 1.5 0.1 0 N 8MR1082 511.01 OCALA P I 1 0.1 0 N 8MR1082 511.01 OCALA P I 1 0.1 0 N 8MR1082 511.01 OCALA MD I 1 0.1 0 N 8MR1082 511.01 OCALA P BTF 1 0.1 0 N 8MR1082 511.01 OCALA P I 1 0.1 0 N 8MR1082 511.01 OCALA MD I 1 0.1 0 N
327 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR1082 511.01 OCALA MD I 1.5 0.1 0 N 8MR1082 511.01 OCALA MD I 1 0.1 0 N 8MR1082 510.03 OCALA NO AS 2 3.1 0 N 8MR1082 510.03 OCALA MD BTF 2 0.6 0 N 8MR1082 510.03 OCALA MD BTF 2 0.9 0 N 8MR1082 510.03 OCALA P BTF 1.5 0.4 0 Y 8MR1082 510.03 OCALA MD BTF 1.5 0.2 0 N 8MR1082 510.03 OCALA MD I 1.5 0.1 0 N 8MR1082 510.03 OCALA MD I 1.5 0.2 0 N 8MR1082 510.03 OCALA P BTF 1.5 0.6 5 N 8MR1082 510.03 OCALA MD I 1 0.1 0 N 8MR1082 510.03 OCALA C BTF 1.5 0.2 4 N 8MR1082 510.03 OCALA MD I 1 0.1 0 N 8MR1082 510.03 UPPER WITHLACOOCHEE RIVER MD BTF 2 1.5 4 N 8MR1082 510.03 UPPER WITHLACOOCHEE RIVER P BTF 2 1.1 4 N 8MR1082 510.03 UPPER WITHLACOOCHEE RIVER MD I 1 0.2 0 N 8MR1082 442.01 OCALA NO AS 1 0.3 0 N 8MR1082 442.01 OCALA NO TS 1 0.1 0 Y 8MR1082 442.01 OCALA MD I 1.5 0.2 0 N 8MR1082 442.01 OCALA MD I 1 0.1 0 N 8MR1082 442.01 OCALA MD I 1 0.1 0 Y 8MR1082 442.01 OCALA MD I 1 0.1 0 N 8MR1082 442.01 OCALA MD I 1 0.1 0 N 8MR1082 442.01 OCALA MD I 1.5 0.1 0 N 8MR1082 442.01 OCALA P BTF 1 0.1 0 N 8MR1082 442.01 OCALA NO AS 1.5 0.5 1 N 8MR1082 442.01 OCALA P BTF 1.5 0.3 0 N 8MR1082 442.01 OCALA MD I 2 1.2 5 N 8MR1082 442.01 OCALA MD BTF 2 1 0 N 8MR1082 442.01 OCALA C BTF 2 1.1 0 N
328 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR1082 442.01 OCALA P BTF 2 0.9 0 N 8MR1082 442.01 OCALA MD BTF 2 0.3 0 N 8MR1082 442.01 OCALA C BTF 4.5 4.3 0 N 8MR93 277.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 2.5 2.5 0 N 8MR93 277.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1.5 1.2 0 N 8MR93 277.02 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 1.5 0.2 0 N 8MR93 277.02 OCALA MD I 3.5 3.3 4 N 8MR93 277.02 OCALA C BTF 2.5 1.5 0 N 8MR93 277.02 OCALA C BTF 2.5 0.8 0 N 8MR93 277.02 OCALA P BTF 2.5 0.5 0 N 8MR93 277.02 OCALA MD BTF 2 0.6 0 N 8MR93 277.02 OCALA C BTF 2 0.4 0 N 8MR93 277.02 OCALA C BTF 1.5 2 0 N 8MR93 277.02 OCALA C BTF 1.5 0.3 0 N 8MR93 277.02 OCALA NO I 1 0.1 0 N 8MR93 277.02 OCALA MD I 1 0.1 0 N 8MR93 277.02 OCALA C NF 1.5 0.1 0 N 8MR93 13401.02 SILICIFIED CORAL MD BTF 2 0.2 0 Y 8MR93 13401.02 OCALA MD BTF 2.5 1.1 0 Y 8MR93 13401.02 OCALA P BTF 2.5 1.3 0 N 8MR93 13401.02 OCALA MD BTF 2 0.7 3 N 8MR93 13401.02 OCALA P BTF 2 0.4 0 Y 8MR93 13401.02 OCALA P BTF 1 0.3 0 N 8MR93 13401.02 OCALA MD I 2 0.8 0 N 8MR93 13401.02 OCALA MD I 1.5 0.3 0 N 8MR93 13401.02 OCALA C BTF 1.5 0.2 0 Y 8MR93 13401.02 OCALA C I 1.5 0.1 0 N
329 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 13401.02 OCALA P I 1 0.1 0 N 8MR93 13401.02 SUWANNEE LIMESTONE FORMATION P BTF 2.5 1.2 0 N 8MR93 13401.02 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 Y 8MR93 13401.02 INDETERMINATE MD I 1.5 0.1 0 N 8MR93 13401.02 INDETERMINATE MD I 1 0.1 0 N 8MR93 13401.02 INDETERMINATE C PR 1 0.1 0 Y 8MR93 13402.03 UPPER WITHLACOOCHEE RIVER MD I 1 0.05 0 N 8MR93 13402.03 UPPER WITHLACOOCHEE RIVER P BTF 1.5 0.2 0 N 8MR93 13402.03 SUWANNEE LIMESTONE FORMATION MD BTF 1.5 0.2 0 N 8MR93 13402.03 OCALA P BTF 4 8.5 0 N 8MR93 13402.03 OCALA P BTF 4 6.5 2 N 8MR93 13402.03 OCALA MD BTF 3 3.5 0 Y 8MR93 13402.03 OCALA P BTF 3 1.5 0 Y 8MR93 13402.03 OCALA P BTF 2.5 2.5 0 Y 8MR93 13402.03 OCALA C BTF 2.5 1.5 0 N 8MR93 13402.03 OCALA C BTF 2 0.8 0 Y 8MR93 13402.03 OCALA C BTF 2 0.4 0 Y 8MR93 13402.03 OCALA MD BTF 2.5 0.9 0 N 8MR93 13402.03 OCALA P CR 2 2.6 0 N 8MR93 13402.03 OCALA P BTF 2 2 1 N 8MR93 13402.03 OCALA C BTF 2 0.7 0 N 8MR93 13402.03 OCALA NO AS 2 1.2 0 N 8MR93 13402.03 OCALA C BTF 2 0.3 0 N 8MR93 13402.03 OCALA MD I 1.5 0.4 0 N 8MR93 13402.03 OCALA P BTF 1.5 0.2 0 Y 8MR93 13402.03 OCALA P BTF 1.5 0.2 0 Y 8MR93 13402.03 OCALA P I 1 0.05 0 N 8MR93 13402.03 OCALA MD I 1.5 0.05 0 N
330 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 13402.03 OCALA MD I 1.5 0.2 0 N 8MR93 13402.03 OCALA P I 1.5 0.4 0 N 8MR93 13402.03 OCALA P BTF 1 0.1 0 Y 8MR93 13402.03 OCALA MD I 1.5 0.3 0 Y 8MR93 13402.03 OCALA MD I 1.5 0.4 0 Y 8MR93 13402.03 OCALA P BTF 1 0.05 0 N 8MR93 13402.03 OCALA MD I 1.5 0.3 1 N 8MR93 13402.03 OCALA MD I 1.5 0.1 0 N 8MR93 13402.03 OCALA P BTF 2 0.5 0 N 8MR93 13402.03 OCALA P BTF 1.5 0.2 0 N 8MR93 13402.03 OCALA MD I 1.5 0.1 0 N 8MR93 13402.03 OCALA MD I 1.5 0.2 0 N 8MR93 13402.03 OCALA P I 1 0.1 0 N 8MR93 13402.03 OCALA MD I 1.5 0.3 0 N 8MR93 13402.03 OCALA C BTF 1 0.1 0 Y 8MR93 13402.03 OCALA NO AS 1 0.1 0 N 8MR93 239.01 OCALA P BTF 2 1 0 N 8MR93 239.01 OCALA P BTF 2 0.5 0 N 8MR93 239.01 OCALA MD BTF 3 1.9 0 Y 8MR93 239.01 OCALA C BTF 2 0.3 0 N 8MR93 239.01 OCALA P BTF 2 0.4 0 N 8MR93 239.01 OCALA MD BTF 1.5 0.3 0 N 8MR93 239.01 OCALA C I 1 0.1 0 Y 8MR93 239.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 Y 8MR93 239.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 2 1.2 0 Y 8MR93 239.01 SANTA FE MD BTF 1.5 0.4 0 I 8MR93 239.01 SANTA FE MD BTF 1.5 0.5 0 Y 8MR93 239.01 SANTA FE P BTF 1.5 0.2 0 N 8MR93 239.01 SANTA FE MD I 1.5 0.3 0 Y
331 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 239.01 OCALA MD I 1.5 0.4 0 N 8MR93 239.01 POSS. SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 239.01 POSS. OCALA LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 239.01 INDETERMINATE MD I 1.5 0.4 0 N 8MR93 239.01 UPPER WITHLACOOCHEE RIVER MD I 1 0.1 0 Y 8MR93 239.01 OCALA MD I 2.5 2.7 0 P 8MR93 239.01 OCALA MD I 1.5 0.5 0 N 8MR93 240.01 POSSIBLE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 2 0.2 0 Y 8MR93 240.01 OCALA MD I 2 0.6 0 I 8MR93 240.01 OCALA MD I 1.5 0.4 0 N 8MR93 240.01 OCALA MD I 1.5 0.5 2 Y 8MR93 240.01 OCALA C BTF 2 0.9 1 N 8MR93 240.01 OCALA C BTF 2.5 0.7 0 N 8MR93 241.03 INDETERMINATE C PR 1 0.05 0 N 8MR93 241.03 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP C BTF 2 0.5 0 N 8MR93 241.03 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P PR 1 0.05 0 N 8MR93 241.03 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP C PR 1 0.05 0 N 8MR93 241.03 SANTA FE MD BTF 1.5 0.1 0 N 8MR93 241.03 SANTA FE C PR 1 0.05 0 N 8MR93 241.03 OCALA C BTF 2 0.7 0 I 8MR93 241.03 OCALA C CR 2 1.6 1 I 8MR93 241.03 OCALA MD BTF 2.5 1.2 0 N
332 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 241.03 OCALA P BTF 2 1 0 N 8MR93 241.03 OCALA P BTF 2 0.6 0 N 8MR93 241.03 OCALA C BTF 2 0.4 0 N 8MR93 241.03 OCALA P BTF 2 0.6 0 N 8MR93 241.03 OCALA MD BTF 1.5 0.6 0 N 8MR93 241.03 OCALA P BTF 1.5 0.2 0 N 8MR93 241.03 OCALA P BTF 1.5 0.2 0 Y 8MR93 241.03 OCALA C BTF 1.5 0.4 0 N 8MR93 241.03 OCALA P BTF 1.5 0.5 0 N 8MR93 241.03 OCALA C BTF 1.5 0.3 0 N 8MR93 241.03 OCALA MD I 1.5 0.1 0 N 8MR93 241.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 UPPER WITHLACOOCHEE RIVER P BTF 2 0.6 0 N 8MR93 250.03 UPPER WITHLACOOCHEE RIVER MD I 2 1.3 0 Y 8MR93 250.03 UPPER WITHLACOOCHEE RIVER C PR 1 0.1 0 N 8MR93 250.03 OCALA P BTF 2 0.5 0 N 8MR93 250.03 OCALA MD BTF 1 0.1 0 Y 8MR93 250.03 OCALA C BTF 2.5 2.3 0 Y 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA P PR 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA C I 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA MD I 1 0.1 0 N 8MR93 250.03 OCALA C BTF 2.5 1.4 0 N 8MR93 250.03 OCALA C BTF 2.5 2.7 0 N 8MR93 250.03 OCALA P BTF 2.5 1.6 0 Y
333 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 250.03 OCALA MD BTF 2.5 1.4 0 N 8MR93 250.03 OCALA MD BTF 2 0.7 0 N 8MR93 250.03 OCALA P BTF 2 0.5 0 Y 8MR1082 41901.01 OCALA P BTF 3 1.4 0 Y 8MR1082 41901.01 OCALA MD BTF 4 3.3 0 Y 8MR1082 41901.01 OCALA MD BTF 2 1.2 0 N 8MR1082 41901.01 OCALA C BTF 1.5 0.3 0 N 8MR1082 41901.01 OCALA MD BTF 2 0.6 0 Y 8MR1082 41901.01 OCALA MD CR 2 1.6 0 N 8MR1082 41901.01 OCALA P BTF 2 0.4 0 Y 8MR1082 41901.01 OCALA NO AS 1 0.6 0 N 8MR1082 41901.01 OCALA P BTF 2 0.6 4 N 8MR1082 41901.01 OCALA P BTF 1.5 0.4 1 N 8MR1082 41901.01 OCALA MD I 1.5 0.1 0 N 8MR1082 41901.01 OCALA MD I 1.5 0.1 0 Y 8MR1082 41901.01 OCALA P BTF 1.5 0.2 0 N 8MR1082 41901.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 2 0.7 0 N 8MR1082 41901.01 UPPER WITHLACOOCHEE RIVER NO TS 1.5 0.7 0 Y 8MR1082 41901.01 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.4 0 Y 8MR1082 41901.01 UPPER WITHLACOOCHEE RIVER P BTF 2 1.2 0 Y 8MR1082 41901.01 POSS. HILLSBOROUGH RIVER MD BTF 3 1.6 0 N 8MR1082 41902.01 UPPER WITHLACOOCHEE RIVER MD BTF 1.5 0.2 0 N 8MR1082 41902.01 SANTA FE MD I 1.5 1.1 0 Y 8MR1082 41902.01 SANTA FE MD I 1.5 0.4 0 Y 8MR1082 41902.01 SANTA FE MD I 1.5 0.4 0 N 8MR1082 41902.01 SANTA FE MD I 1 0.1 0 N 8MR1082 41902.01 SANTA FE MD I 1 0.1 0 Y
334 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR1082 41902.01 SANTA FE MD I 1.5 0.1 0 N 8MR1082 41902.01 SANTA FE MD PR 1 0.1 0 Y 8MR1082 41902.01 SANTA FE C I 1 0.1 0 N 8MR93 329.01 EAST PANASOFFKEE P BTF 2.5 1 0 Y 8MR93 329.01 EAST PANASOFFKEE C BTF 2 1.6 0 Y 8MR93 329.01 EAST PANASOFFKEE P BTF 2 0.5 2 N 8MR93 329.01 SUWANNEE LIMESTONE FORMATION MD I 1.5 0.7 0 Y 8MR93 329.01 SUWANNEE LIMESTONE FORMATION MD BTF 2 0.8 0 Y 8MR93 329.01 SUWANNEE LIMESTONE FORMATION P BTF 1 0.1 1 Y 8MR93 329.01 OCALA C BTF 5 7 0 N 8MR93 329.01 OCALA C BTF 3.5 6.5 0 N 8MR93 329.01 OCALA P BTF 2 2.1 0 N 8MR93 329.01 OCALA P BTF 2 1.4 0 N 8MR93 329.01 OCALA MD BTF 2 0.9 0 N 8MR93 329.01 OCALA C BTF 1.5 0.7 2 N 8MR93 329.01 OCALA MD BTF 1.5 0.1 0 N 8MR93 329.01 OCALA P BTF 1.5 0.5 0 Y 8MR93 329.01 OCALA P BTF 1.5 0.1 0 N 8MR93 329.01 OCALA MD BTF 1 0.1 0 N 8MR93 329.01 OCALA MD BTF 1.5 0.2 0 N 8MR93 329.01 OCALA MD BTF 1 0.1 0 Y 8MR93 329.01 OCALA MD BTF 1 0.2 0 Y 8MR93 329.01 OCALA P BTF 1.5 0.2 0 N 8MR93 329.01 OCALA MD BTF 1.5 0.1 0 Y 8MR93 329.01 OCALA MD I 1 0.1 0 N 8MR93 329.01 OCALA C P 1 0.1 0 Y 8MR93 329.01 OCALA MD BTF 1.5 0.1 0 N 8MR93 329.01 OCALA C BTF 1.5 0.2 0 Y 8MR93 329.01 OCALA MD I 1 0.1 0 Y
335 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 329.01 OCALA P I 1 0.1 0 N 8MR93 329.01 OCALA P BTF 1 0.1 0 Y 8MR93 329.01 OCALA P BTF 1 0.1 0 N 8MR93 329.01 OCALA MD I 1 0.1 0 N 8MR93 329.01 OCALA MD I 1 0.1 4 N 8MR93 11601.02 OCALA C CR 5.5 29 5 N 8MR93 11601.02 OCALA MD I 2 0.4 0 N 8MR93 11601.02 OCALA MD BTF 2.5 1.5 0 N 8MR93 11601.02 OCALA P NF 2 0.2 0 N 8MR93 11601.02 OCALA P BTF 2.5 1.5 0 N 8MR93 11601.02 OCALA MD BTF 2.5 1.6 0 N 8MR93 11601.02 OCALA P BTF 1.5 0.5 0 N 8MR93 11601.02 OCALA P BTF 3 1.3 0 N 8MR93 11601.02 OCALA P BTF 2.5 1.1 0 N 8MR93 11601.02 OCALA P BTF 3.5 3.2 0 N 8MR93 11601.02 OCALA MD BTF 3 1.7 0 Y 8MR93 11601.02 OCALA P BTF 2.5 1.1 0 Y 8MR93 11601.02 OCALA P BTF 2 0.7 0 N 8MR93 11601.02 OCALA P BTF 2 0.3 0 N 8MR93 11601.02 OCALA MD I 1.5 0.3 0 N 8MR93 11601.02 OCALA P BTF 2 0.7 0 N 8MR93 11601.02 OCALA P BTF 2 0.9 0 N 8MR93 11601.02 OCALA C BTF 2 0.5 0 N 8MR93 11601.02 OCALA C BTF 2 0.7 0 N 8MR93 11601.02 OCALA MD I 1.5 0.4 0 N 8MR93 11601.02 OCALA P I 2 0.4 0 N 8MR93 11601.02 OCALA P BTF 1.5 0.3 0 N 8MR93 11601.02 OCALA C BTF 1.5 0.2 0 N 8MR93 11601.02 OCALA MD I 1.5 0.7 0 N 8MR93 11601.02 OCALA MD I 2 0.5 0 N 8MR93 11601.02 OCALA P BTF 1.5 0.4 1 N 8MR93 11601.02 OCALA P BTF 1.5 0.3 0 N
336 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 11601.02 OCALA MD I 1.5 0.3 0 Y 8MR93 11601.02 OCALA MD I 1.5 0.2 0 N 8MR93 11601.02 OCALA MD I 2 0.2 0 N 8MR93 11601.02 OCALA MD I 1.5 0.1 0 N 8MR93 11601.02 OCALA P BTF 1.5 0.1 0 N 8MR93 11601.02 OCALA MD I 1.5 0.3 0 N 8MR93 11601.02 OCALA P I 1.5 0.7 1 N 8MR93 11601.02 OCALA MD I 1.5 0.4 0 N 8MR93 11601.02 OCALA MD I 1.5 0.2 0 N 8MR93 11601.02 OCALA C BTF 1.5 0.3 0 N 8MR93 11601.02 OCALA MD I 1 0.1 0 N 8MR93 11601.02 OCALA MD I 1.5 0.1 0 N 8MR93 11601.02 OCALA MD I 1.5 0.05 0 N 8MR93 11601.02 OCALA MD I 1.5 0.1 0 N 8MR93 11601.02 OCALA P BTF 1.5 0.1 0 N 8MR93 11601.02 OCALA P BTF 1 0.1 0 N 8MR93 11601.02 OCALA C BTF 1.5 0.1 0 N 8MR93 11601.02 OCALA MD I 1 0.1 0 N 8MR93 11601.02 OCALA MD I 1 0.1 0 N 8MR93 11601.02 OCALA P BTF 1 0.05 0 N 8MR93 11601.02 OCALA MD I 1 0.2 0 Y 8MR93 11601.02 OCALA C BTF 1 0.05 0 N 8MR93 11601.02 OCALA P BTF 1 0.05 0 N 8MR93 11601.02 OCALA MD I 1 0.05 0 N 8MR93 11601.02 OCALA NO TS 2 1 0 Y 8MR93 11601.02 OCALA MD TS 2.5 1.5 0 Y 8MR93 11601.02 OCALA MD I 2 1.2 0 Y 8MR93 11601.02 OCALA NO TS 4 6.2 0 Y 8MR93 11601.02 OCALA MD BTF 2.5 1.6 0 N 8MR93 11601.02 OCALA P BTF 2 1 0 Y 8MR93 11601.02 OCALA NO TS 2.5 1.6 0 N 8MR93 11601.02 OCALA NO TS 2 0.9 1 Y
337 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 11601.02 OCALA MD BTF 2 0.7 0 Y 8MR93 11601.02 OCALA MD BTF 2 1 5 Y 8MR93 11601.02 OCALA MD I 2.5 1.7 0 Y 8MR93 11601.02 SUWANNEE LIMESTONE FORMATION MD I 2.5 1 0 Y 8MR93 11601.02 SUWANNEE LIMESTONE FORMATION MD I 2 0.4 0 Y 8MR93 11601.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.3 0 Y 8MR93 11601.02 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.2 0 N 8MR93 11601.02 UPPER WITHLACOOCHEE RIVER MD I 1 0.05 0 Y 8MR93 11601.02 SUWANNEE LIMESTONE FORMATION C BTF 3 1.6 0 N 8MR93 11601.02 SUWANNEE LIMESTONE FORMATION P BTF 2.5 0.9 0 N 8MR93 11601.02 SILICIFIED CORAL MD I 1 0.05 0 Y 8MR93 11602.01 OCALA NO TS 2 1.7 2 Y 8MR93 11602.01 OCALA P EBT 3 4.1 5 N 8MR93 11602.01 OCALA C BTF 4.5 10 0 N 8MR93 11602.01 OCALA MD BTF 2 0.9 0 Y 8MR93 11602.01 OCALA P BTF 1.5 0.3 0 N 8MR93 11602.01 OCALA P BTF 1.5 0.4 1 N 8MR93 11602.01 OCALA P BTF 1.5 0.3 0 N 8MR93 11602.01 OCALA MD I 1.5 0.3 0 N 8MR93 11602.01 OCALA P BTF 1 0.1 0 Y 8MR93 11602.01 OCALA MD I 1 0.1 0 N 8MR93 11602.01 OCALA MD I 1 0.05 0 N 8MR93 11602.01 OCALA MD I 1.5 0.5 0 Y 8MR93 11602.01 OCALA MD I 1.5 0.1 0 N 8MR93 11602.01 OCALA MD I 1 0.1 0 Y 8MR93 11602.01 OCALA MD I 1 0.1 0 N
338 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 11602.01 OCALA MD I 1.5 0.1 0 N 8MR93 11602.01 OCALA C BTF 1.5 0.2 0 N 8MR93 11602.01 OCALA MD I 1.5 0.1 0 N 8MR93 11602.01 OCALA P NF 1.5 0.1 0 N 8MR93 11602.01 OCALA MD I 1 0.1 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 2 0.5 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP P BTF 1.5 0.4 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 2.5 0.9 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1.5 0.3 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER C BTF 3 4.7 2 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER MD I 2 0.6 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER P BTF 1.5 0.4 0 N 8MR93 11602.01 SILICIFIED CORAL MD I 2 0.6 0 Y 8MR93 11602.01 SANTA FE P BTF 2.5 0.8 0 N 8MR93 11602.01 SANTA FE P BTF 2 0.3 0 N 8MR93 11602.01 SANTA FE C BTF 1.5 0.2 0 N 8MR93 11602.01 SANTA FE P I 1.5 0.2 0 N 8MR93 11602.01 TAMPA LIMESTONE FORMATION MD I 1.5 0.3 0 N 8MR93 11602.01 INDETERMINATE P BTF 1 0.1 0 Y 8MR93 11602.01 INDETERMINATE P BTF 1.5 0.4 0 N 8MR93 11602.01 INDETERMINATE P BTF 1 0.1 0 N 8MR93 11602.01 INDETERMINATE P I 1 0.1 0 N
339 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 11602.01 SANTA FE C BTF 1.5 0.3 0 N 8MR93 11602.01 UPPER WITHLACOOCHEE RIVER MD I 1 0.2 0 Y 8MR93 14301.05 OCALA MD BTF 3 3.4 4 Y 8MR93 14301.05 OCALA MD BTF 4 6.1 2 N 8MR93 14301.05 OCALA P BTF 2.5 1.6 0 Y 8MR93 14301.05 OCALA MD BTF 2.5 1.3 0 N 8MR93 14301.05 OCALA P BTF 2.5 0.8 0 N 8MR93 14301.05 OCALA C BTF 2.5 0.5 0 Y 8MR93 14301.05 OCALA MD BTF 2 0.5 0 Y 8MR93 14301.05 OCALA MD I 2 0.9 0 Y 8MR93 14301.05 OCALA MD BTF 2.5 1.8 0 N 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 1.5 0.3 0 Y 8MR93 14301.05 OCALA MD BTF 2.5 1.3 0 N 8MR93 14301.05 OCALA C BTF 2 0.2 0 N 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD BTF 1.5 0.4 0 N 8MR93 14301.05 OCALA C BTF 2 0.3 0 Y 8MR93 14301.05 OCALA P BTF 2 0.7 0 Y 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP NO I 2 0.8 2 N 8MR93 14301.05 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 14301.05 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 14301.05 POSSIBLE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 Y 8MR93 14301.05 OCALA P I 1 0.1 0 N
340 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade ( cm ) Weight (g) Dorsal Cortex Thermal alteration 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER C BTF 1.5 0.1 0 N 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1.5 0.3 0 Y 8MR93 14301.05 SUWANNEE LIMESTONE FORMATION MD I 1.5 0.1 0 Y 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.2 0 N 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER C NF 1 0.1 0 N 8MR93 14301.05 SUWANNEE LIMESTONE FORMATION MD I 1 0.1 0 N 8MR93 14301.05 INDETERMINATE P I 1 0.1 0 N 8MR93 14301.05 UPPER WITHLACOOCHEE RIVER, GREEN SWAMP MD I 1 0.1 0 N 8MR93 14302.01 UPPER WITHLACOOCHEE RIVER P BTF 3 4.6 0 Y 8MR93 14302.01 UPPER WITHLACOOCHEE RIVER P BTF 2.5 1.6 0 Y 8MR93 14302.01 UPPER WITHLACOOCHEE RIVER MD I 1.5 0.4 0 N 8MR93 14302.01 SANTA FE C BTF 3 2.2 0 N 8MR93 14302.01 SANTA FE MD I 2 1.7 0 N 8MR93 14302.01 SANTA FE MD I 1.5 0.3 0 N 8MR93 14302.01 OCALA NO AS 4 11.5 4 N 8MR93 14302.01 OCALA C BTF 5 24.5 0 N 8MR93 14302.01 OCALA C BTF 5.5 15.3 3 N 8MR93 14302.01 OCALA NO AS 2 3.9 4 N 8MR93 14302.01 OCALA NO AS 2 2.2 3 N 8MR93 14302.01 OCALA NO AS 1.5 1.3 4 N 8MR93 14302.01 OCALA NO AS 1.5 0.9 4 N
341 Table B 1 . Continued Site number Catalog number Quarry cluster Morphological type Technological type Size grade (cm) Weight (g) Dorsal Cortex Thermal alteration 8MR93 14302.01 OCALA C BTF 2 1.5 4 N 8MR93 14302.01 OCALA NO AS 1.5 0.8 0 N 8MR93 14302.01 OCALA P BTF 2.5 2.7 0 N 8MR93 14302.01 OCALA P BTF 2.5 1.8 0 N 8MR93 14302.01 OCALA MD I 2.5 1.5 0 N 8MR93 14302.01 OCALA MD BTF 2.5 0.9 0 N 8MR93 14302.01 OCALA P R BTF 2 1.5 0 N 8MR93 14302.01 OCALA MD BTF 2 0.9 0 N 8MR93 14302.01 OCALA P BTF 2.5 1.4 0 N 8MR93 14302.01 OCALA C BTF 2 0.8 0 Y 8MR93 14302.01 OCALA P BTF 2 1.2 0 N 8MR93 14302.01 OCALA MD I 2 0.7 0 N 8MR93 14302.01 OCALA MD I 1.5 0.3 0 N 8MR93 14302.01 OCALA C BTF 1.5 0.4 0 N 8MR93 14302.01 OCALA MD I 1.5 0.4 4 N 8MR93 14302.01 OCALA P BTF 1.5 0.3 0 N 8MR93 14302.01 OCALA P I 1.5 0.3 0 Y 8MR93 14302.01 OCALA C BTF 1.5 0.1 0 Y 8MR93 14302.01 OCALA P BTF 1 0.2 0 N 8MR93 14302.01 OCALA MD I 1 0.1 0 N 8MR93 7401.01 OCALA C CR 7 169.7 2 N
342 Table B 2 . Key to headings and abbreviations for lithic debitage attribute data. Heading Entry Description Site number Site inventory number assigned by the Florida Master Site Files Catalog number Inventory number assigned in the Laboratory of Southeastern Archaeology Quarry Cluster Provenance, as determined by microscopic examination Morphological type Debitage type, based on morphology (Sullivan and Rozen 1985) C Complete flake MD Medial or distal flake fragment NO Non orientable debitage P Proximal flake fragment Technological type Debitage type, based on technology AS Angular shatter BTF Biface thinning flake CR Core reduction I Indeterminate NF Notching flake PR Pressure flake TS Thermal shatter Size grade Debitage size, determined with nested squares in 0.5 cm increments Weight Weight in grams Dorsal Cortex Proportion of dorsal surface comprised of cortex 0 0% 1 1 25% 2 26 50% 3 51 75% 4 76 99% 5 100% Thermal alteration Presence or absence of evidence for thermal alteration
343 Table B 3 . Lithic tool provenance and attribute data. Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR93 18.02 HAFTEDBIFACE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP AREA 2.3 BOTH ORBITOIDS AND MILIOLIDS (LARGE), SECONDARY POROSITIY AND CRYSTAL LINED VOIDS 8MR93 19.02 MODIFIEDFLAKE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP AREA 0.8 BOTH ORBITOIDS AND MILIOLIDS, SECONDARY POROSITY 8MR93 21.02 BIFACE OCALA 12 8MR93 21.03 HAFTEDBIFACE OCALA 20.8 SCATTERED INFREQUENT SMALL ORBITOIDS 8MR93 16.02 HAFTEDBIFACE OCALA 2.1 STEM 8MR93 10.02 HAFTEDBIFACE OCALA 24.9 THIS BIFACE FORM RESEMBLES "NOVICE" FLINTKNAPPER BIFACES FROM WEST WILLIAMS SITE (AUSTIN ET AL. 2002) IN HILLSBOROUGH COUNTY 8MR93 11.05 HAFTEDBIFACE OCALA 2 STEM 8MR93 2.02 BIFACE OCALA 6.8 ABUDANT SMALL ORBITOIDS 8MR93 51502.02 MODIFIEDFLAKE OCALA 28.7 MODERATE TO FREQUENT SMALL TO LARGE ORBITOIDS AND SECONDARY POROSITY 8MR93 284.01 BIFACE OCALA 18.2 70CMBS 8MR93 29.03 MODIFIEDFLAKE SUWANNEE LIMESTONE FORMATION 56.5 8MR93 3101.01 CORE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP AREA 130.4 ORBITOIDS AND MILIOLIDS 8MR93 444.04 HAFTEDBIFACE OCALA 7.3 STEM (LOOKS LIKE A MID SECTION OF A BIFACE), SMALL SCATTERED ORBITOIDS, SECONDARY POROSITY 8MR93 174.02 HAFTEDBIFACE OCALA 5.5 8MR93 34.03 MODIFIEDFLAKE INDETERMINATE 3.2 "PINHEAD" CORAL 8MR93 43.05 BIFACE POSSIBLE SANTA FE 11.5 ABUNDANT MILIOLIDS, NO SAND, SOMEWHAT POORLY SILICIFIED 8MR1082 336.02 HAFTEDBIFACE OCALA 18.3
344 Table B 3 . Continued Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR1082 336.03 HAFTEDBIFACE OCALA 1.9 STEM 8MR93 45.01 MODIFIEDFLAKE OCALA 5.3 SCATTERED INFREQUENT SMALL ORBITOIDS 8MR93 47.04 BIFACE OCALA 9 OVERSHOT FLAKE, NOT BIFACE FRAGMENT 8MR93 514.03 HAFTEDBIFACE OCALA 0.7 STEM 8MR93 514.04 HAFTEDBIFACE OCALA 23.9 8MR93 51.02 BIFACE POSSIBLE OCALA 5.9 TAMPA/ST. MARKS FORMATION, OCCURRS AS RESIDUUM IN OCALA QC 8MR1082 416.02 UNIFACE/SCRAPER OCALA 104.8 8MR1082 417.02 BIFACE GAINESVILLE 35.1 LARGE AND ABUNDANT ORBITOIDS, SIMILAR TO SAMPLES FROM SW20TH AVE IN GAINESVILLE 8MR1082 418.02 HAFTEDBIFACE GAINESVILLE 9.9 LARGE AND ABUNDANT ORBITOIDS, CRYSTAL LINED VOIDS 8MR1082 418.03 BIFACE OCALA 0.7 COMMON ORBITOIDS, SECONDARY POROSITY, CRYSTAL LINED VOIDS 8MR1082 41902.02 BIFACE INDETERMINATE 8.7 IMPACT FRACTURED 8MR1082 420.02 MODIFIEDFLAKE GAINESVILLE 14.5 ABUNDANT, MODERATE TO LARGE SIZED ORBITOIDS, CRYSTAL LINED VOIDS 8MR1082 420.03 MODIFIEDFLAKE OCALA 214.8 FLAKE BLANK/QUARRY BLANK 8MR93 55.02 MODIFIEDFLAKE OCALA 59.8 8MR1082 426.02 MODIFIEDFLAKE OCALA 5.4 8MR93 469.01 HAFTEDBIFACE OCALA 12.2 SMALL WIDELY SCATTERED ORBITOIDS 8MR93 10801.01 HAFTEDBIFACE SANTA FE 1.2 STEM, SOMEWHAT LARGE ORBITOIDS, SECONDARY POROSITY, SAND INCLUSIONS
345 Table B 3 . Continued Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR93 109.12 BIFACE UPPER WITHLACOOCHEE RIVER 2.6 109.11 OR 109.12? BAG SAYS 109.11, TAG SAYS 109.11, EXCEL SAYS 109.12; LARGE FOSSILS AND PRESENCE OF SAND SUGGEST UPPER WITHLACOOCHEE ALTHOUGH SANTA FE QC CANNOT BE RULED OUT 8MR93 110.03 HAMMERSTONE OCALA LIMESTONE FORAMATION 27 LIMESTONE, NOT CHERT, HAMMERSTONE USE DUBIOUS 8MR93 185.01 CORE OCALA 96.6 UNIDIRECTIONAL 8MR93 305.07 MODIFIEDFLAKE OCALA 1.4 305.06 OR 305.07? BAG AND FS DIFFER. COMMON SMALL AND SCATTERED ORBITOIDS 8MR93 306.03 BIFACE OCALA 0.9 ORBITOIDS PRESENT 8MR93 226.03 CORE OCALA 85.1 THERMALLY ALTERED, PROBABLY EARLY REDUCTION DEBITAGE INSTEAD OF A CORE 8MR93 225.01 TESTEDCOBBLE 888 OVERWEIGHT 8MR93 225.04 HAFTEDBIFACE OCALA 37.9 MODERATE TO FREQUENT SMALL ORBITOIDS 8MR93 64.02 HAFTEDBIFACE OCALA 2.4 SMALL SCATTERED ORBITOIDS AND SECONDARY POROSITY 8MR93 330.01 BIFACE OCALA 1.5 PINELLAS 8MR93 14402.01 HAFTEDBIFACE OCALA 19.4 8MR93 208.02 BIFACE POSSIBLE GAINESVILLE 4.6 LIPPED FLAKE, NOT BIFACE. FREQUENT ORBITOIDS, SOME LARGE, PROBABLY GAINESVILLE QC BUT COULD BE OCALA 8MR93 208.03 MODIFIEDFLAKE OCALA 4.5 DOESN'T APPEAR TO BE MODIFIED 8MR93 205.01 BIFACE OCALA 0.7 8MR93 411.01 CORE OCALA 151.9 8MR93 82.01 HAFTEDBIFACE OCALA 5.8 100CMBS, FREQUENT SMALL ORBITOIDS AND SECONDARY POROSITY
346 Table B 3 . Continued Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR93 75.01 BIFACE OCALA 19.8 MODERATE TO FREQUENT SMALL ORBITOIDS 8MR93 7802.01 HAFTEDBIFACE OCALA 11.3 8MR93 30702.01 HAMMERSTONE 134.6 8MR93 30702.02 HAFTEDBIFACE HILLSBOROUGH RIVER 7 8MR93 30702.03 BIFACE OCALA 54 THREMALLY FRACTURED PREFORM FRAGMENT. FRACTURED DURING THERMAL ALTERATION? COMMON SMALL AND SCATTERED ORBITOIDS 8MR93 81.01 BIFACE POSSIBLE SUWANNEE LIMESTONE 0.3 8MR1082 351.01 CORE OCALA LIMESTONE FORMATION, PROBABLE OCALA QUARRY CLUSTER 37.6 COMMON TO FREQUENT SMALL ORBITOIDS 8MR93 7402.01 HAFTEDBIFACE OCALA 27.3 35 45CMBS 8MR93 7402.02 CORE OCALA 14.5 8MR93 223.01 MODIFIEDFLAKE OCALA 31.4 8MR93 414.01 CORE OCALA 78.1 8MR93 11601.03 BIFACE OCALA 14.2 FEW, SCATTERED ORBITOIDS 8MR93 122.03 HAFTEDBIFACE OCALA 1.3 HAFT SNAP BENDING FRACTURE PERPENDICULAR, MODERATE MEDIUM TO SMALL ORBITOIDS 8MR93 122.04 HAFTEDBIFACE OCALA 6.9 SCATTERED INFREQUENT SMALL ORBITOIDS 8MR93 257.01 HAFTEDBIFACE OCALA 21.4 8MR1082 352.01 HAFTEDBIFACE OCALA 2.3 STEM, SMALL SCATTERED ORBITOIDS 8MR1082 43901.02 MODIFIEDFLAKE OCALA 28.1 BIFACE, SMALL COMMON TO FREQUENT SCATTERED ORBITOIDS, SECONDARY POROSITY 8MR93 145.01 HAMMERSTONE OCALA 206.2 HEAVILY PATINATED 8MR1082 511.02 BIFACE SANTA FE 25.1 AT73CMBS 8MR1082 510.02 CORE OCALA 465.7 8MR1082 442.02 HAFTEDBIFACE OCALA 18.7 AT57CMBS
347 Table B 3 . Continued Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR93 303.04 MODIFIEDFLAKE OCALA 1.6 SMALL SCATTERED ORBITOIDS 8MR93 302.01 BIFACE OCALA 1 FREQUENT SMALL ORBITOIDS 8MR93 277.01 CORE OCALA 60.2 8MR93 310.01 BIFACE OCALA 0.6 LIPPED FLAKE, NOT BIFACE. 8MR93 313.02 MODIFIEDFLAKE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP AREA 8.4 ORBITOIDS AND MILIOLIDS, SAND PRESENT 8MR93 313.03 CORE OCALA 44.5 8MR93 13402.02 BIFACE OCALA 34.2 8MR93 13402.04 BIFACE OCALA 10.7 8MR93 13102.02 HAFTEDBIFACE OCALA 2.6 TIP, COMMON SMALL ORBITOIDS, SECONDARY POROSITY, SOMEWHAT SPOTTY SILICIFICATION 8MR93 241.01 CORE/HAMMERSTONE OCALA 85.3 EXHAUSTEDCOREUSEDASHAMMER 8MR93 241.02 CORE OCALA 156.01 POSSIBLE, FREQUENT BUT SMALL SIZED ORBITOIDS 8MR93 250.01 BIFACE OCALA 47.4 DISTAL PREFORM FRAGMENT, PROBABLE LATERAL SNAP DURING THINNING 8MR93 250.02 HAFTEDBIFACE OCALA 1.1 STEM 8MR93 460.03 HAFTEDBIFACE OCALA 3.3 SMALL SCATTERED ORBITOIDS 8MR93 7401.03 CORE OCALA 162.6 8MR93 7401.04 HAMMERSTONE OCALA 412.9 POSSIBLE CORE NUCLEUS NO CLEAR EVIDENCE OF BATTERING AND USE AS HAMMERSTONE 8MR93 7301.01 CORE OCALA 278.3 8MR93 3501.03 BIFACE OCALA 1.6 BIFACE SHOULDER/EDGE FRAGMENT, SMALL SCATTERED ORBITOIDS AND SECONDARY POROSITY 8MR93 3501.04 MODIFIEDFLAKE POSSIBLE BROOKSVILLE 0.3 ABUNDANT SMALL MILIOLIDS, ECHINOID SHELL FRAGMENT, NO SAND
348 Table B 3 . Continued Site number Catalog number Form Quarry cluster Weight (g) Notes 8MR93 3502.03 MODIFIEDFLAKE UPPER WITHLACOOCHEE RIVER, GREEN SWAMP AREA 8.9 ORBITOIDS AND MILIOLIDS, SAND PRESENT
349 LIST OF REFERENCES Abshire, A. E., Alden L. Potter, Allen R. Taylor, Clyde H. Neil, Walter H. Anderson, John I. Rutledge and Stevenson B. Johnson (editors) 1935 Some Further Papers on Aboriginal Man in the Neighborhood of the Ocala National Forest . Civilian Conservation Corps, Company 1420, Ocala Camp, Florida. Alenicheva, Diana thesis, Int erdisciplinary Ecology P rogram, School of Natural Resources and Environment , University of Florida, Gainesville. Allen, Greg 2013 Now Endangered, Florida's Silver Springs Once Lured Tourists . Electronic document, http://www.npr.org/2013/04/13/177105692/before disney floridas silver springs lured tourists , accessed October 9, 2015. Alley, Richard B., Paul A. M ayewski, Todd Sowers, Minze Stuiver, Kendrick C. Taylor, and Peter U. Clark 1997 Holocene Climatic Instability: A Prominent, Widespread Event 8200 yr ago. Geology 25:483 486. Alt, David, and H. K. Brooks. 1965 Age of the Florida Marine Terraces. The Jou rnal of Geology 73:406 411. Anderson, David G., and Kenneth E. Sassaman 2012 Recent Developments in Southeastern Archaeology: From Colonization to Complexity . The Society for American Archaeology, Washington, D.C. d Kenneth E. Sassaman 1996 Environmental and Chronological Considerations. In The Paleoindian and Early Archaic Southeast , edited by David G. Anderson and Kenneth E. Sassaman, pp. 3 15. The University of Alabama Press, Tuscaloosa. Anderson, Elyse 2015 D raft Report: Silver Glen Run, Locus C & D (8LA1 W). Manuscript on file, Laboratory of Southeastern Archaeology, University of Florida, Gainesville. Andrews, Travis DeWitt 2012 Testate Amoebae as Hydrological Proxies in the Florida Everglades. Unpublished Master's thesis, Earth and Environmental Sciences, Lehigh University, Bethlehem, Pennsylvania.
350 Archaeological Consultants Inc., and Janus Research 2001 Phase III Mitigative Excavations at Lake Monroe Outlet Midden (8VO53), Volusia County, Florida . Report Submitted to U.S. Department of Transportation Federal Highway Administration and Florida Department of Transportation District Five, Sarasota. Copies available from Florida Master Site Files, Tallahassee. Arthur, Jonathan D., P.G. 1149, A lan E. Baker, James R. Cichon, Alex R. Wood and Andrew Rudin 2005 Florida Aquifer Vulnerability Assessment (FAVA): Contamination Potential of Florida's Principal Aquifer Systems . Division of Resource Assessment and Management, Florida Geological Survey, T allahassee. Arz, Helge W., Frank Lamy, and JÃ¼rgen PÃ¤tzold 2006 A Pronounced Dry Event Recorded around 4.2 ka in Brine Sediments from the Northern Red Sea. Quaternary Research 66:432 441. Ashley, Keith H. and Nancy Marie White (editors) 2012 Late Prehis toric Florida: Archaeology at the Edge of the Mississippian World . University Press of Florida, Gainesville. Aten, Lawrence E. 1999 Middle Archaic Ceremonialism at Tick Island, Florida: Ripley Bullen's 1961 Excavation at the Harris Creek Site. The Florida Anthropologist 52:131 200. Aucott, Walter R. 1988 Areal Variation in Recharge to and Discharge from the Floridan Aquifer System in Florida . Water Resources Investigations Report 88 4057. U.S. Geological Survey, Reston, Virginia. Austin, Robert J . 1997 The Economics of Lithic Resource Use in South Central Florida. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. Austin, Robert J. , and Richard W. Estabrook 2000 Chert Distribution and Exploitation in Peninsular Florida. The Florida Anthropologist 53:116 130 Bair, Cinnamon 2011 Polk Chronicles: Kissengen Spring was Spot for Fun. Electronic document, http://www.theledger.com/article/20110823/COLUMNISTS/108235000 , accessed August 14, 2015.
351 Balsillie, James H. and Joseph F. Donoghue 2004 High Resolution Sea Level History for the Gulf of Mexico since the Last Glacial Maximum . Report of Investigations 103. Florida Geological Survey, Tallahassee. 2011 Northern Gulf of Mexico Sea Level History for the Past 20,000 Years. In Gulf of Mexico: Origin, Waters, and Biota , Vol. 3: Geology, edited by Noreen A. Buster and Charles W. Holmes, pp. 53 69. Texas A&M University Press, College Station. Barber, D.C., A. Dyke, C. Hillaire Marcel, A.E. Jennings, J.T. Andrews, M.W. Kerwin, G. Bilodeau, R. McNeely, J. Southon, M. D. Morehead, and J. M. Gagnon 1999 Forcing of the Cold Event of 8,200 years ago by Catastrophic Drainage of Laurentide Lakes. Nature 400:344 348. Barrett, John C. 1999 The Mythical Landscapes of the British Iron Age. In Archaeologies of Landscape: Contemporary Perspectives , edited by Wendy Ashmore and A. Bernard Knapp, pp. 253 265. Blackwell, Malden, Massachusetts. 2001 Agency, the Duality of Structure, and the Problem of the Archaeological Record. In Archaeological Theory Today, edited by Ian Hodder, pp. 141 64. Blackwell, Malden, Massachusetts. Bartram, John 1769 A Description of East Florida, with a Journal, kept by John Bartram of Philadelphia, Botanist to His Majesty for the Floridas . William Stork, London. Bartram, William 1996 Travels and Other Writings . Library of America, New York. Basso, Keith H. 1996 a Wisdom Sits in Places: Landscape and Language among the Western Apache . University of New Mexico Press, Albuquerque. 1996b Wisdom Sits in Places: Notes on a Western Apache Landscape. In Senses of Place , edited by Steven Feld and Keith H. Basso, pp. 53 90. School for A dvanced Research Press, Santa Fe, New Mexico. Beasley, Virgil R., III 2009 Monumentality during the Mid Holocene in the Upper and Middle St. Johns River Basins, Florida. Unpublished Ph.D. dissertation, Department of Anthropology, Northwestern University, Evanston, Illinois.
352 Beck, Charlotte, Amanda K. Taylor, George T. Jones, Cynthia M. Fadem, Caitlyn R. Cook, and Sara A. Millward 2002 Rocks are Heavy: Transport Costs and Paleoarchaic Quarry B ehavior in the Great Basin. Journal of Anthropological Archa eo logy 21: 481 507. Beck, Robin A., Douglas J. Bolender, James A. Brown and Timothy K. Earle 2007 Eventful Archaeology: The Place of Space in Structural Transformation. Current Anthropology 48:833 860. Belcourt, Andrew N., Martin F. Dickinson, and Lucy B . Wayne 2009 Cultural Resources Survey and Assessment, Silver Springs Retention Pond, Marion County Florida . SouthArc Inc. Submitted to Florida Master Site Files, Tallahassee. Bell, Catherine 1992 Ritual Theory, Ritual Practice . Oxford University Press, New York. Bell, Tyler, and Gary Lock 2000 Topographic and Cultural Influences on Walking the Ridgeway in L ater Prehistoric T imes. In Beyond the Map: Archaeology and Spatial Technologies , edited by Gary Lock, pp. 85 100. IOS Press, Amsterdam. Belleville , Bill 2011 Salvaging the Real Florida: Lost and Found in the State of Dreams . University Press of Florida, Gainesville. Belaineh, Getachew, Joseph Stewart, Peter Sucsy, Louis H. Motz, Kijin Park, Shahrokh Rouhani, and Michael Cullum 2012 Groundwater Hy drology. In St. Johns River Water Supply Impact Study , edited by Edgar F. Lowe, Lawrence E. Battoe, Hal Wilkening, Michael Cullum, and Tom Bartol, pp. 4 i 4 100. Technical Publication SJ2012 1. St. Johns River Water Management District, Palatka, Florida. Bender, Barbara 2001 Landscapes On the Move. Journal of Social Archaeology 1:75 89. Bennett Michael W. 2004 Hydrogeologic Investigation of the Floridan Aquifer System: Big Cypress Preserve Collier County, Florida ida Water Management District, West Palm Beach, Florida. Blakeslee, Donald J. 2010 Holy Ground, Healing Water: Cultural Landscapes at Waconda Lake, Kansas . Texas A&M University Press, College Station .
353 Bland, Myles C.P., and Marsha A. Chance 2000 An Intensive Cultural Resource Assessment Survey of the Fanning Springs State Recreation Area, Levy County, Florida . Report of Investigations 191. Environmental Services, Inc., Jacksonville, Florida. Blessing, Meggan E. 2011 Zooarchaeological Assemblage. I n Archaeological Investigations at Salt Springs (8MR2322), Marion County, Florida , edited by Jason M. O'Donoughue, Kenneth E. Sassaman, Meggan E. Blessing, Johanna B. Talcott, and Julie C. Byrd , pp. 65 85 . Technical Report 1 1. Laboratory of Southeastern Ar chaeology, Department of Anthropology, University of Florida. Bleich er, Niels, and Carsten Schubert 2015 Why Are They Still T here? A Model of Accumulation and Decay of Organic Prehistoric Cultural Deposits. Journal of Archaeological Science 61 : 277 286. Blum, Michael D., and TorbjÃ¶rn E. TÃ¶rnqvist 2000 Fluvial Responses to Climate and Sea Level Change: A Review and Look Forward. Sedimentology 47(s1):2 48. Bonn, Mark A. 2004 Visitor Profiles, Economic Impacts and Recreational Aesthetic Values Associated with Eight Priority Florida Springs Located in the St. Johns River Water Management District . Special Publication SJ2004 SP35 . St. Johns River Water Management District, Palatka, Florida. Bonn, Mark A., and Frederick W. Bell 2003 Economic Impact of Selec ted Florida Springs on Surrounding Local Areas . Report prepared for the Florida Department of Environmental Protection, Division of State Lands, Florida Springs Task Force, Tallahassee. Booth, Robert K., Stephen T. Jackson, Steven L. Forman, John E. Kutzbach, E.A. Bettis, Joseph Kreigs, and David K. Wright 2005 A Severe Centennial Scale Drought in Midcontinental North America 4200 years ago and Apparent Global Linkages. The Holocene 15:321 328. Bradley, Richard 2000 An Archaeology of Natural Places . Routledge, London. 2005 Ritual and Domestic Life in Prehistoric Europe . Routledge, London. Brady, James E., and Wendy Ashmore 1999 Mountains, Caves, Water: Ideational Landscapes of the Ancient Maya. In Archaeologies of Landscape: Contemporary Perspec tives , edited by Wendy Ashmore and A. Bernard Knapp, pp. 124 145. Blackwell, Malden, Massachusetts.
354 Brenneman, Walter L., and Mary G. Brenneman 1995 Crossing the Circle at the Holy Wells of Ireland . University Press of Virginia, Charlottesville . Brenner , M., L.W. Keenan, S.J. Miller, and C.L. Schelske 1999 Spatial and Temporal Patterns of Sediment and Nutrient Accumulation in Shallow Lakes of the Upper St. Johns River Basin, Florida. Wetlands Ecology and Mana gement 6:221 240. Brenner, Mark, Claire L. S chelske, and Lawrence W. Keenan 2001 Historical Rates of Sediment and Nutrient Accumulation in Marshes of the Upper St. Johns River Basin, Florida, USA. Journal of Paleolimnology 26:241 257. Broe c ker Wallace S. 2006 Was the Younger Dryas Triggered by a Flood? Science 312:1146 1148. Broe c ker Wallace S., George H. Denton, R. Lawrence Edwards, Hai Cheng, Richard B. Alley, and Aaron E. Putnam 2010 Putting the Younger Dryas Cold Event into Context. Quaternary Science Reviews 29:1078 1081. Berson, Thomas R. 2011 Silver Springs: The Florida Interior in the American I magination . Unpublished Ph.D. dissertation, Department of History, University of Florida , Gainesville. Booth, R.K., M. Lamentowicz and D.J. Charman 2010 Preparation and Analysis of Testate Amoebae in Peatland Paleoenvironmental Studies. Mires and Peat 7(2):1 7. Brooks, H.K. 1981 Guide to the Physiographic Divisions of Florida . Institute of Food and Agricultural Sciences, University of Florida, Gain esville. Brown, Bill 2001 Thing Theory. Critical Inquiry 28:1 22. Brown, Canter, Jr. 1991 . University of Central Florida Press, Orlando. Brown, D.P. 1984 Impact of Development on Availability and Quality of Ground Water in Eastern Nassau County, Florida, and Southeastern Camden County, Georgia . Water Resources Investigations Report 83 4190. U.S. Geological Survey, Reston, Virginia.
355 Brown, James A., and Robert K. Vierra 1983 What Happened in the Middle Archaic? Introduc tion to an Ecological Approach to Koster Site Archaeology. In Archaic Hunters and Gatherers in the American Midwest , edited by James L. Phillips and James A. Brown, pp. 165 195. Academic Press, New York . Brown, Kenneth M., Dennis Varza and Terry D. Richar dson 1989 Life Histories and Population Dynamics of Two Subtropical Snails (Prosobranchia:Viviparidae). Journal of the North American Benthological Society 8:222 228. Brown, Mark T., Kelly Chinners Reiss, Matthew J. Cohen, Jason M. Evans , Patrick W. Inglett, Kanika Sharma Inglett, K. Ramesh Reddy , Thoma s K. Frazer, Charles A. Jacoby, Edward J. Phlips , Robert L. Knight & Sky K. Notestein , Richard G. Hamann , and Kathleen A. McKee 2008. Summary and Synthesis of Available Literature on the Effects of Nutrients on Springs Organisms and Systems . University of Florida Water Institute . Report submitted to the Florida Department of Environmental Protection, Tallahassee. BrÃ¼ck, Joanna 1999 Ritual and Rationality: Some Problems of Interpretation in European Archaeology. European Journal of Archaeology 2(3):313 344. Burt, Al 2003 Healing Waters: Springs Represent a Disappearing Part of Florida. Tallahassee Democrat 27 April:E1, E5. Tallahassee, Florida. Bullen, Ripley P. 1953 Excavations at Manatee Spring s, Florida. The Florida Anthropologist 6:53 67. 1975 A Guide to the Identification of Florida Projectile Points . Revised ed. Kendall Books, Gainesville. Bush, P.W. and R.H. Johnston 1988 Ground Water Hydraulics, Regional Flow, and Ground Water Developm ent of the Floridan Aquifer System in Florida and in Parts of Georgia, South Carolina, and Alabama . Professional Paper 1403 C. U.S. Geological Survey, Washington, D.C. Butt, Peter, Alaa Aly , and David Toth 2006 Silver Springs Spring Vent Documentation an d Geochemical Characterization . Special Publication SJ2008 SP6 . St. Johns River Water Management District, Palatka, Florida.
356 Byrd, Julia C. 2011 Archaic Bone Tools in the St. Johns River Basin, Florida: Microwear and Manufacture Traces. Unpublished Florida State University, Tallahassee. Carmichael, David, Jane Hubert, and Brian Reeves 1994 Introduction. In Sacred Sites, Sacred Places , edited by David L. Carmichael, Jane Hubert, Brian Reeve s, and Audchild Schanche, pp. 1 8 . Routledge, New York. Carr, Archie 1996 A Naturalist in Florida: A Celebration of Eden . Yale University Press, New Haven. Carr, Susan Latham 2009 Soiling Glen Springs. Electronic document, http://www.ocala.com/article/20090803/ARTICLES/908031008 , accessed October 2, 2015. Casey, Edward S. 1996 How to Get from Space to Place in a Fairly Short Stretch of Time: Phenomenological Prolegomena. In Senses of Place , edi ted by Steven Feld and Keith H. Basso, pp. 13 52. School for Advanced Research Press, Santa Fe, New Mexico. Chambers, Frank M., Robert K. Booth, Francois De Vleeschouwer, Mariusz Lamentowicz, Gael Le Roux, Dmitri Mauquoy, Jonathan E. Nichols and Bas van G eel 2012 Development and Refinement of Proxy Climate Indicators from Peats. Quaternary International 268:21 33. Chapman, John 2000 Fragmentation in Archaeology: People, Places and Broken Objects in the Prehistory of South Eastern Europe . Routledge, New York. Chapman, John, and Bisserka Gaydarska 2007 Parts and Wholes: Fragmentation in Prehistoric Context . Oxbow Books, Oxford. Charlton, Ro 2008 Fundamentals of Fluvial Geomorphology . Routledge, New York Claassen, Cheryl 2010 Feasting with Shellfish in the Southern Ohio Valley: Archaic Sacred Sites and Rituals . Univ ersity of Tennessee Press, Knocville.
357 Clausen, Carl J., H.K. Brooks, and Al B. Wesolowsky 1975 The Early Man Site at Warm Mineral Springs, Florida. Journal of Field Archaeology 2:3:191 213 Clausen, C.J., A.D. Cohen, Cesare Emiliani, J.A. Holman, and J.J. Stipp 1979 Little Salt Spring, Florida: A Unique Underwater Site. Science 203(4381):609 614. Connoly, James, and Mark Lake 2006 Geographical Information Systems in Archaeolo gy . Cambridge University Press, New York. Cooke, C.W. 1939 Scenery of Florida, Interpreted by a Geologist . Bulletin 17. Florida Geological Survey, Tallahassee. 1945 The Geology of Florida . Bulletin 29. Florida Geological Survey, Tallahassee. Connor, Jody N., and Thomas V. Belanger 1981 Ground Water Seepage in Lake Washington and the Upper St. Johns River Basin, Florida. Journal of the American Water Resources Association 17:799 805. Copeland, Rick (compiler) 2003 Florida Springs Classification Syst em and Spring Glossary . Special Publication 52. Florida Geological Survey, Tallahassee. Cordell, Ann S. 2004 Paste Variability and Possible Manufacturing Origins of Late Archaic Fiber Tempered Pottery from Selected Sites in Peninsular Florida. In Early P ottery: Technology, Function, Style, and Interaction in the Lower Southeast , edited by Rebecca Saunders and Christopher T. Hays, pp. 63 104. University of Alabama Press, Tuscaloosa. Delcourt, Paul A., and Hazel R. Delcourt 2004 Prehistoric Native America ns and Ecological Change: Human Ecosystems in Eastern North America since the Pleistocene . Cambridge University Press, New York. DeMarrais, Elizabeth, Christ Gosden, and Colin Renfrew (editors) 2004 Rethinking Materiality: The Engagement of Mind with Mat erial World . McDonald Institute for Archaeological Research , Cambridge. DeMort, Carole L. 1991 The St. Johns River System. In The Rivers of Florida , edited by Robert J. Livingston, pp. 97 120. Springer Verlag, New York.
358 Denevan, William M. 1992 The Pristine Myth: The Landscape of the Americas in 1492. Annals of the Association of American Geographers 82:369 385. Denson, Robin L., Gary D. Ellis , and Russell Dorsey 1995 Archaeological Survey of De Leon Springs State Recreation Area . Ellis Archaeo logy. Submitted to Florida Department of Environmental Protection. Copies available from Florida Master Site Files, Tallahassee. Dickinson, Martin F., and Lucy B. Wayne 2002 Cultural Resources Survey and Assessment, Silver Springs Boat Landing & Train Depot, Marion County, Florida . SouthArc, Inc. Submitted to Smart Parks of Florida, Inc. Copies available from Florida Master Site Files, Tallahassee. 2003 Cultural Resources Survey and Assessment : Fanning Springs State Recreation Area Proposed Cabin Site s, Levy County, Florida . SouthArc, Inc. Submitted to Florida Master Site Files, Tallahassee. Donar, C., E.F. Stoermer, and M. Brenner 2009 The Holocene Paleolimnology of Lake Apopka, Florida. Nova Hedwigia 135:57 70. Donders, Timme H. 2014 Middle Holoc ene Humidity Increase in Florida: Climate or Sea Level? Quaternary Science Reviews 103:170 174. Donders, Timme H., Hugo Jan de Boer, Walter Finsinger, Eric C. Grimm, Stefan C. Dekker, Gert Jan Reichart, and Friederike Wagner Cremer 2011 Impact of the Atl antic Warm Pool on Precipitation and Temperature in Florida during North Atlantic Cold Spells. Climate Dynamics 36:109 118. Donders, Timme H., Friederike Wagner, David L. Dilcher, and Henk Visscher 2005 Mid to Late Holocene El NiÃ±o Southern Oscillation D ynamics Reflected in the Subtropical Terrestrial Realm. Proceedings of the National Academy of Sciences of the United States of America 102:10904 10908. Donoghue, Joseph F. 2011 Sea Level History of the Northern Gulf of Mexico Coast and Sea Level Rise Sc enarios for the Near Future. Climatic Change 107:17 33. Doran, Glen H. (editor) 2002 Windover: Multidisciplinary Investigations of an Early Archaic Florida Cemetery . University of Florida Press, Gainesville. Douglas, Marjory Stoneman 1967 Florida: The Long Frontier . Harper & Row, New York.
359 Dunbar, James S. 1990 Site File s , Tallahassee. 1991 Resource Orientation of Clovis and Suwannee Age Paleoindian Sites in Florida. In Clov is: Origins and Adaptations , edited by R. Bonnichsen and K. L. Turnmire, pp. 185 213. Center for the Study of the First Americans, Oregon State University, Corvallis. 2002 Chronostratigraphy and Paleoclimate of Late Pleistocene Florida and the Implicatio ns of Changing Paleoindian Land Use. Unpublished Master s thesis, Department of Anthropolog y, Florida State University, Tallahassee. 2003 Alexander Springs Inspection of August 4, 2003 . Bureau of Archaeological Research, CARL Archaeological Survey. Submi tted to Florida Division of Historical Resources. Copies available from Florida Master Site Files, Tallahassee. 2006 Paleoindian Archaeology . In First Floridians and Last Mastodons: The Page Ladson Site in the Aucilla River , edited by S. David Webb, pp. 403 435. Springer, New York. Dunbar, James S., and Glen Doran 2010 Paleoindian Sites Revisited Known Sites and New Perspectives . Florida State University. Submitted to Florida Master Site Files, Tallahassee. Dunbar, James S., Michael K. Faught, and S. D avid Webb 1988 Page/Ladson (8JE 591): An Underwater Paleo Indian Site in N orthwestern Florida. The Florida Anthropologist 41: 442 452. Dunbar, James S., and Ben I. Waller 1983 A Distribution Analysis of the Clovis/Suwannee Paleo Indian Sites of Florida A Geographic Approach. The Florida Anthropologist 36:18 30. Dwyer, Peter D. 1996 The Invention of Nature. In Redefining Nature: Ecology, Culture and Domestication , edited by Roy Ellen and Katsuyoshi Fukui, pp. 157 186. Berg, Oxford. Dye, David H. 1996 Riverine Adaptation in the Midsouth. In Of Caves and Shell Mounds , edited by Kenneth C. Carstens and Patty Jo Watson, pp. 140 158. University of Alabama Press, Tuscaloosa.
360 Earl, Steven 2009 Ichetucknee: Sacred Waters . University Press of Florida, Gaines ville. Edwards, W.E. 1954 The Helen Blazes Site of Central Eastern Florida: A Study in Method u tilizing the Disciplines of Archaeology, Geology, and Pedology. Unpublished Ph.D. dissertation, Department of Geology, Columbia University, New York. Emerson, Thomas E., Dale L. McElrath, and Andrew C. Fortier (editors) 2009 Archaic Societies: Diversity and Complexity across the Midcontinent . State University of New York Press, Albany. Endonino, Jon C. 2007 A Reevaluation of the Gainesville, Ocala, and Lake Panasoffkee Quarry Clusters. The Florida Anthropologist 60:19 38. 2009 The Thornhill Lake Archaeological Research Project: 2005 2008. The Florida Anthropologist 61:149 165. 2010 Thornhill Lake: Hunter Gatherers, Monuments, and Memory. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. Enfield, David B., Alberto M. Mestas NuÃ±ez, and Paul J. Trimble 2001 The Atlantic Multidecadal Oscillation and its Relation to Rainfall and River Flows in the Continental U.S. Geophysical Research Letters 28:2077 2080. Escobar, Jaime, Mark Brenner, Thomas J. Whitmore, William F. Kenney and Jason H. Curtis 2008 Ecology of Testate Amoebae (Thecamoebians) in Subtropical Florida Lakes. Journal of Paleolimnology 40:715 731. Estabrook, Richard W. and Laura M. Weant 1993 Cultural Resource Assessment Survey of the Gemini Springs Project Site Volusia County, Florida . Janus Research . Submitted to Pinnacle Companies, Orlando. Copies available from Florida Master Site Files, Tallah assee. Faught, Michael K. 2009 Silver Spring Paleoindian Research Project, Report of Field Operations, 15 17 December, 2003 . Archaeological Research Cooperative, Inc. Submitted to Florida Master Site Files, Tallahassee. Faught, Michael K., and Joseph F. Donoghue 1997 Marine Inundated Archaeological Sites and Paleofluvial Systems: Examples from a Karst Controlled Continental Shelf Setting in Apalachee Bay, Northeastern Gulf of Mexico. Geoarchaeology 12:417 458.
361 Faught, M ichael K., and J ames C. Waggoner Jr. 2012 The Early Archaic to Middle Archaic transition in Florida: An A rgument for D iscontinuity. The Florida Anthropologist 65 :153 175. Faure, Hugues, Robert C Walter, and Douglas R Grant 2002 The Coastal Oasis: Ice Age Springs on Emerged Continental Shelves. Global and Planetary Change 33:47 56. Ferguson, G.E., C.W. Lingham, S.K. Love, and R.O. Vernon 1947 Springs of Florida . Bulletin 31. Florida Geological Survey, Tallahassee. Ferring, C. Reid 1986 Rates of Fluvial Sedimentation: Implications for Archaeological Variability. Geoarchaeology 1:259 274. Filley, T.R., K.H. Freeman, T.S. Bianchi, M. Baskaran, L.A. Colarusso, and P.G. Hatcher 2001 An Isotopic Biogeochemical Assessment of Shifts in Organic Matter Input to Holocene Sediments from Mud Lak e, Florida. Organic Geochemistry 32:1153 1167. Fleury, Perrine, Michel Bakalowicz, and Ghislain de Marsily 2007 Submarine Springs and Coastal Karst Aquifers: A Review. Journal of Hydrogeology 339:79 92. F lorea, Lee J., and H. L. Vacher 2006 Springflow Hydrographs: Eogenetic vs. Telogenetic Karst. Ground Water 44: 352 361. 2007 Eogenetic Karst Hydrology: Insights from the 2004 Hurricanes, Peninsular Florida. Ground Water 45: 439 446. Florida Department of Community Affairs and Florida Department of Environmental Protection 2002 Protecting Florida's Springs: Land Use Planning Strategies and Best Management Practices . Florida Department of Community Affairs and Florida Department of Environmental Protection, Tallahassee. Florida Departm ent of Environmental Protection (FDEP) 2012 FDEP GeoData Directory. Electronic document, http://www.dep.state.fl.us/gis/datadir.htm/ , accessed July 1, 2015. 2014 About Florida Springs. Electron ic document, http://www.dep.state.fl.us/springs/about.htm , accessed October 8, 2015.
362 2015 DEP Springs Funding Projects . Electronic document, http://fdep.maps.arcgis.com/apps/MapJournal/index.html?appid=002cea9e3e454 943aa67ae0934c2b3f1 , accessed October 8, 2015. Florida Springs Initiative 2007 Program Summary and Recommendations . Flori da Department of Environmental Protection, Tallahassee. Florida Springs Institute Members 2015 Guest C olumn: Jacksonville has S erious Groundwater I ssues . Electronic document, http://jacksonville.com/business/columnists/2015 09 04/story/guest column jacksonville has serious groundwater issues , acces sed October 7, 2015. Florida Springs Task Force 2000 . Submitted to the Florida Department of Environmental Protection. Copies available from the Florida Department of Environmental Protection, Ta llahassee. Florida Trend 2015 Florida Springs to Receive $82 Million Toward Restoration P rojects . Electronic document, htt p://www.floridatrend.com/article/19068/gov scott distributes 82 million for florida springs restoration projects , accessed October 8, 2015. Fogelson, Raymond D. 1989 The Ethnohistory of Events and Nonevents. Ethnohistory 36(2):133 147. Fowler, Catherine S., and Don D. Fowler (editors) 2008 The Great Basin: People and Place in Ancient Times . School for Advanced Research Press, Santa Fe, New Mexico. Frazier, Charles 2006 Thirteen Moons . Random House, New York. Gainesville Sun and Ocala Star Banner 2013 Fragile Springs. Electronic document, http://fragilesprings.com/ , accessed October 9, 2015. Gaiser, E.E., B.E. Taylor, and M.J. Brooks 2001 Establishment of Wetlands on the Southeastern Atlantic Coastal Plain: Paleolimnological Evidence of a mid Holocene Hydrologic Threshold from a South Carolina Pond. Journal of Paleolimnology 26:373 391. Gibson, Jon L. 2001 The Ancient Mounds of Poverty Po int: Place of Rings . University Press of Florida, Gainesville.
363 Gillespie, Susan D. 2007 Comment on "Eventful Archaeology: The Place of Space in Structural Transformation" by R. Beck, D. Bolender, J. Brown, and T. Earle. Current Anthropology 48:846 847. Gilmore, Zackary I. 2011 Silver Glen Run, Locus B (8LA1 West). In St. Johns Archaeological Field School 2007 10: Silver Glen Run (8LA1) , by Kenneth E. Sassaman, Zackary I. Gilmore, and Asa R. Randall , pp. 171 314. Technical Report 12. Laboratory of South eastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2014 Gathering Places: Histories of Material and Social Interaction at Late Archaic Shell Mounds in Florida. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. 2015 Subterranean Histories: Pit Events and Place Making in Late Archaic Florida. In The Archaeology of Events: Cultural Change and Continuity in the Ancient Southeast , edited by Zackary I. Gilmore and Jason M. 140. University of Alabama Press, Tuscaloosa. Gilmore, Zackary I., and Jason M. O'Donoughue 2015 Introduction: The Enigma of the Event. In The Archaeology of Events: Cultural Change and Continuity in the Pre Columbian Southeast , edit ed by Zackary I. Gilmore and J ason 22. The University of Alabama Press, Tuscaloosa. Gleason, Patrick J., and Peter Stone 1994 Age, Origin, and Landscape Evolution of the Everglades Peatland. In Everglades: The Ecosystem and its Rest oration , edited by Steven Davis and John Ogden, pp. 149 197. St. Lucie Press, Boca Raton, Florida. Goggin, John M. 1947 A Preliminary Definition of Archaeological Areas and Periods in Florida. American Antiquity 13:114 127. 1952 Space and Time Perspect ive in Northern St. Johns Archeology, Florida . University Press of Florida, Gainesville. Gosden, Chris 2005 What Do Objects W ant? Journal of Archaeological Method and Theory 12 : 193 211. Gribben, Arthur 1992 Holy Wells and Sacred Water Sources in Britain and Ireland: An Annotated Bibliography . Garland Pub lishing, New York .
364 Grimm, E.C., G.L. Jacobson Jr, W.A. Watts, B.C.S. Hansen and K.A. Maasch 1993 A 50,000 Year Record of Climate Oscillations from Florida and its Temporal Correlation. Science 61 :9. Grimm E.C., S. Lozano Garcia, H. Behling, and V Markgraf 2001 Holocene Vegetation and Climate Variability in the Americas. In Interhemispheric Climate Linkages , edited by V Markgraf, pp. 352 370. Academic Press, San Diego. Grimm, E.C., W.A. Watts, G.L. Jacobson, B. Hansen, H.R. Almquist and A.C. Dieffenbacher Krall 2006 Evidence for Warm Wet Heinrich Events in Florida. Quaternary Science Reviews 25:2197 2211. Hays, Christopher T., and Richard A. Weinstein 2004 Early Pottery at Poverty Point: Orig ins and Functions. In Early Pottery: Technology, Function, Style, and Interaction in the Lower Southeast , edited by Rebecca Saunders and Christopher T. Hays, pp. 150 168. University of Alabama Press, Tuscaloosa. Heffernan, James B., Dina M. Liebowitz, Tho mas K. Frazer, Jason M. Evans, and Matthew J. Cohen 2010 Algal Blooms and the Nitrogen Enrichment Hypothesis in Florida Springs: Evidence, Alternatives, and Adaptive Management. Ecological Applications 20:816 829. Heidegger, Martin 1971 Poetry, Language , Thought . Harper, London. Hemmings, E. Thomas 1975 The Silver Springs Site, Prehistory in the Silver Springs Valley, Florida . The Florida Anthropologist 28:141 158. Hiers, Fred rogress . Electronic document, http://www.ocala.com/article/20141002/ARTICLES/141009969 , accessed October 7, 2015. Hine, Albert C. 2013 Geologic History of Florida . University Press of Flor ida, Gainesville. Hodder, Ian 2011a Human Thing Entanglement: Towards an Integrated Archaeological Perspective. Journal of the Royal Anthropological Institute 17:154 177.
365 2011b Wheels of Time: Some A spects o f Entanglement Theory and the Secondary Products R evolution. Journal of World Prehistory 24: 175 187. 2012 Entangled: An Archaeology of the Relationships between Humans and Things. Wiley Blackwell, Malden, Massachusetts. Horton, B.P., W.R. Peltier, S.J. Culver, R. Drummond, S.E. Engelhart, A.C . Kemp, and K.H. Thomson 2009 Holocene Sea Level Changes along the North Carolina Coastline and their Implications for Glacial Isostatic Adjustment Models. Quaternary Science Reviews 28:1725 1736. Howey, Meghan C.L. 2007 Using Multi Criteria Cost Surfac e Analysis to Explore Past Regional Landscapes: A Case Study of Ritual Activity and Social I nteraction in Michigan, AD 1200 1600. Journal of Archaeological Science 34 :1830 1846. Huang, Yongsong, Bryan Shuman, Yi Wang, Thompson Webb III, Eric C. Grimm, and George L. Jacobson Jr. 2006 Climatic and Environmental Controls on the Variation of C 3 and C 4 Plant Abundances in Central Florida for the Past 62,000 Years. Palaeogeography, Palaeoclimatology, Palaeoecology 237:428 435. Huber, Red 2012 Looking Back at Winter Park's Famous Sinkhole. Electronic document, http://www.orlandosentinel.com/os fla360 looking back at winter pa rks famous sinkhole 20121113 story.html , accessed July 31, 2015. Hubert, Jane 1994 Sacred Beliefs and Beliefs of Sacredness. In Sacred Sites, Sacred Places , edited by David L. Carmichael, Jane Hubert, Brian Reeve s, and Audchild Schanche, pp. 9 19 . Routle dge, New York. Hudson, Charles M. 1976 The Southeastern Indians . University of Tennessee Press, Knoxville. Hughes, J. D., H. L. Vacher, and Ward E. Sanford 2009 Temporal Response of Hydraulic Head, Temperature, and Chloride Concentrations to Sea Level Changes, Floridan Aquifer System, USA. Hydrogeology Journal 17:793 815. Huntsinger, Lynn, and MarÃa FernÃ¡ndez GimÃ©nez 2000 Spiritual Pilgrims at Mount Shasta, Cali fornia. The Geographical Review , 90:536 558.
366 Ingold, Tim 2011 Being Alive: Essays on Movement, Knowledge and Description . Routledge, New York . 2012 Toward an Ecology of Materials . Annual Review of Anthropology 41: 427 442. 2013 Making: Anthropology, Archaeology, Art and Architecture . Routledge, New York . Jahn, Otto L. and Ripley P. Bullen 1978 The Tick Island Site, St. Johns River, Florida . Publication 10. Florida Anthropological Society, Gainesville. Jones, Gregg W., Sam B. Upchurch, and Kyle M. Chapman 1996 Origin of Nitrate in Ground Water Discharging from Rainbow Springs, Marion County, Florida . Southwest Florida Water Management District, Brooksville, F lorida. Johnson, J ay K. 1981 Yellow Creek Archaeological Project, Volume 2 . Publications in Anthropology 2. Tennessee Valley Authority, Chattanooga. Joyce, Rosemary A., and Susan D. Gillespie (editors) 2015 Things in Motion: Object Itineraries in Anthropological Practice . School for Advanced Research Press, Santa Fe, New Mexico. Karst Envir onmental Services, Inc. 2008 Submerged Springs Site Documentation: August and September 2007 . Special Publication SJ2008 SP7. St. Johns River Water Management District, Palatka, Florida. Katz, B.G. 2004. Sources of Nitrate Contamination and Age of Wate r in Large Karstic Springs of Florida. Environmental Geology 46:689 706. Kelly, Martin H., and James A. Gore 2008 Florida River Flow Patterns and the Atlantic Multidecadal Oscillation. River Research Applications 24:598 616. Kidder, Tristram R. 2002 Mapping P overty Point. American Antiquity ( 67 ):89 101. 2010 Hunter Gatherer Ritual and Complexity: New E vidence from Poverty Point, Louisiana. In Ancient Complexities: New Perspectives in Precolumbian North America , edited by Susan M. Alt, pp. 32 51. Uni versity of Utah Press , Salt Lake City.
367 Kindinger, Jack L., Davis, Jeffrey B., and Flocks, James G. 1994 High Resolution Single Channel Seismic Reflection Surveys of Orange Lake and Other Selected Sites of North Central Florid a. Open File Report 94 616. U.S. Geological Survey, St. Petersburg, Florida. 1999 Geology and Evolution of Lakes in North Central Florida. Environmental Geology 38:301 321. 2000 Subsurface Characterization of Selected Water Bodies in the St. Johns Rive r Water Management District, Northeast Florida . Open File Report 2000 180. U.S. Geological Survey, St. Petersburg, Florida. Knight, Robert L. orever . Electronic document, http://www.gainesville.com/article/20080928/OPINION03/809280253 , accessed October 9, 2015. 2012 The River of Denial. Electronic document, http://www.gainesville.com/article/20121216/OPINION03/121219918 , accessed October 20, 2015. 2013 Restoring Florida's S prings . Electronic document, http://www.gainesville.com/ article/20131020/opinion03/131019617 , accessed October 20, 2015. 2014 Paying the Springs Bill. Electronic document, http://www.gainesville.com/article/20140824/OPINION03/1408 29810 , accessed October 20, 2015. 2015 Silenced Springs: Moving from Tragedy to Hope . Howard T. Odum Florida Springs Institute, Gainesville. Knight, Vernon James 1989 Symbolism of Mississippian Mounds. In Powhatan's Mantle: Indians in the Colonial Southeast , edited by P. H. Wood, Gregory A. Waselkov and W. T. Hatley, pp. 279 291. University of Nebraska Press, Lincoln. 2002 Hydrogeology and Simulated Effects of Ground Water Withdrawals fro m the Floridan Aquifer System in Lake County and in the Ocala National Forest and Vicinity, North Central Florida . Water Resources Investigations Report 02 4207. U.S. Geological Survey, Tallahassee, Florida.
368 Knox, J. C. 1995 Fluvial Systems Since 20,000 years BP. In Global Continental Palaeohydrology , edited by K.J. Gregory and L. Starkel, pp. 87 108. Wiley, Chichester. Kroening, Sharon E. 2004 Streamflow and Water Quality Characteristics at Selected Sites of the St. Johns R iver in Central Florida, 1933 to 2002 . Scientific Investigations Report 2004 5177. U.S. Geological Survey, Altamonte Springs, Florida. Kurek, Joshua, Les C. Cwynar, and Ray W. Spear 2004 The 8200 cal yr BP Cooling Event in Eastern North America and the U tility of Midge Analysis for Holocene Temperature Reconstructions. Quaternary Science Reviews 23:627 639. Land, L.A., and C.K. Paull 2000 Submarine Karst Belt Rimming the Continental Slope in the Straits of Florida. Geo Marine Letters 20:123 132. Lane, Ed 1986 Karst in Florida . Special Publication 29. Florida Geological Survey, Tallahassee. 2001 The Spring Creek Submarine Springs Group, Wakulla County, Florida . Special Publication 47. Florida geological Survey, Tallahassee. Lanier, Sidney 1876 Florida: Its Scenery, Climate, and History: With an Account of Charleston, Savannah, Augusta, and Aiken, a Chapter for Consumptives, Various Papers on Fruit Culture, and a Complete Hand Book and Guide . J B. Lippincott & Co mpany , Philadelphia . Le Conte, Jo hn 1861 On the Optic al Phenomena Presented by the " Silver Spring," in Marion County, Florida. The American Journal of Science and Arts 31:1 11. Liebowitz, Dina. M., Matthew J. Cohen, James B. Heffernan, Lawrence V Korhnak, and Thomas K Frazer 2014 Envir onmentally Mediated Consumer Control of Algal Proliferation in Florida Springs. Freshwater Biology 59: 2009 2023. Livingston, Robert J. 1991 The Oklawaha River System. In The Rivers of Florida , edited by Robert J. Livingston, pp. 85 95. Springer Verlag, New York.
369 Logan, Patrick 1980 The Holy Wells of Ireland . Colin Smythe Limited, Gerrards Cross, Buckinghamshire . Lucas, Gavin 2008 Time and the Archaeological Event. Cambridge Archaeological Journal 18:59 65. Lucero, Lisa J., and Andrew Kinkella 2015 Pilgrimage to the Edge of the Watery Underworld: an Ancient Maya Water Temple at Cara Blanca, Belize. Cambridge Archaeological Journal 25: 163 185. McCall, George A. 1974 Letters from the Frontiers . University Press of Florida, Gainesville. McCar thy, Helen 2004 of Nirvana. In Beyond Primitivism: Indigenous Religious Traditions and Modernity , edited by Jacob K. Olupona, pp. 172 178. Routledge, New York. McGee, Ray M. and Ryan J. Wheeler 1994 Stratigraphic Excavations at Groves' Orange Midden, Lake Monroe, Volusia County, Florida: Methodology and Results. The Florida Anthropologist 47:333 349. McLeod, Michael 1986 That Sinking Feeling in Florida, You Never Know When Your Law n or Your Neighorhood Will Be Swallowed up by Nature's Vacuum. Electronic document, http://articles.orlandosentinel.com/1986 01 19/news/0190220042 _1_mae rose sinkhole rose owens , accessed July 31, 2015. Magny, Michel, Boris VanniÃ¨re, Gianni Zanchetta, Eric Fouache, Gilles Touchais, Lera Petrika, CÃ©line Coussot, Anne VÃ©ronique Walter Simonnet, and Fabien Arnaud 2009 Possible Complexity of the Clima tic Event Around 4300 3800 cal. BP in the Central and Western Mediterranean. The Holocene 19:823 833. Marella, Richard L. 2008 Water Use in Florida, 2005 and Trends 1950 2005 . Fact Sheet 2008 3080 . U.S. Geological Survey, Reston, Virginia. 2014 Water Withdrawals, Use, and Trends in Florida, 2010 . Scientific Investigations Report 2014 5088. U.S. Geological Survey, Reston, Virginia.
370 Marquardt, William H., and Patty Jo Watson (editors) 2005 Archaeology of the Middle Green River Region, Kentucky . Unive rsity Press of Florida, Gainesville. Marrinan, Rochelle A., H. Stephen Hale and William M. Stanton 1990 Test Excavations at Silver Glen Springs, Florida (8MR123) . Miscellaneous Report Series 2. Department of Anthropology, Florida State University, Tallah assee. Marshall, Curtis H. , Roger A. Pielke , Sr., Louis T. Steyaert, and Debra A. Willard 2004 The Impact of Anthropogenic Land Cover Change on the Florida Peninsula Sea Breezes and Warm Season Sensible Weather. Monthly Weather Review 132 : 28 52. Martin, Jonathan B., and Randolph W. Dean 2001 Exchange of Water Between Conduits and Matrix in the Floridan Aquifer. Chemical Geology 179: 145 165. Martin, Richard A. 1966 Eternal Spring: Man's 10,000 Years of History at Florida's Silver Springs . Great Outdoo rs Publishing Co mpany, St. Petersburg, Florida. Mauss, Marcel 1990  The Gift: The Form and Reason for Exchange in Archaic Societies . W.W. Norton, New York. Mayewski, Paul A., Eelco E. Rohling, J. Curt Stager, WibjÃ¶rn KarlÃ©n, Kirk A. Maasch, L. Dav id Meeker, Eric A. Meyerson, Francoise Gasse, Shirley van Kreveld, Karin Holmgren, Julia Lee Thorp, Gunhild Rosqvist, Frank Rack, Michael Staubwasser, Ralph R. Schneider, and Eric J. Steig 2004 Holocene Climate Variability. Quaternary Research 62:243 255. Meskell, Lynn (editor) 2005 Archaeologies of Materiality . Blackwell, Malden, Massachusetts. Milanich, Jerald T. 1994 Archaeology of Precolumbian Florida . University Press of Florida, Gainesville. Miller, Daniel 2005 Materiality . Duke University Press, Durham, North Carolina. Miller, James A. 1986 Hydrogeologic Framework of the Floridan Aquifer System in Florida and in Parts of Georgia, Alabama, and South Carolina. Professional Paper 1403 B. U.S. Geological Survey, Washington, D.C.
371 1997 Hydrog eology of Florida. In The Geology of Florida , edited by Anthony F. Randazzo and Douglas S. Jones, pp. 69 88. University Press of Florida, Gainesville. Miller, James J. 1992 Effects of Environmental Change on Late Archaic People of Northeast Florida. The Florida Anthropologist 45:100 106. 1998 An Environmental History of Northeast Florida . University Press of Florida, Gainesville. Mitchell, Edward A. D., Daniel J. Charman and Barry G. Warner 2008 Testate Amoebae Analysis in Ecological and Paleoecologic al Studies of Wetlands: Past, Present and Future. Biodiversity and Conservation 17(9):2115 2137. Moore, Clarence B. 1894a Certain Sand Mounds of the St. Johns River, Florida. Part I. Journal of the Academy of Natural Sciences of Philadelphia 10:4 128. 1 894b Certain Sand Mounds of the St. Johns River, Florida. Part II. Journal of the Academy of Natural Sciences of Philadelphia 10:129 246. 1895 Certain Sand Mounds of the Ocklawaha River, Florida. Journal of the Academy of Natural Sciences of Philadelphi a 10:518 543. Moore, Paul J., Jon B. Martin, and Elizabeth J. Screaton 2009 Geochemical and Statistical Evidence of Recharge, Mixing, and Controls on Spring Discharge in an Eogenetic Karst Aquifer. Journal of Hydrology 376:443 455. Moran, John 2013 Reclaiming Our S prings . Electronic document, http://www.gainesville.com/article/20131229/OPINION03/131229706/ 1/opinion03?Title=John Moran Reclaiming our springs , accessed October 9, 2015. Morphy, Howard 1995 Landscape and the Reproduction of the Ancestral Past. In Anthropology of Landscape: Perspe ctives on Place and Space , edited by Eric Hirsch and Michael O'Hanlon, pp. 184 209. Clarendon Press, Oxford. Morrissey, Sheila K., Jordan F. Clark, Michael Bennett, Emily Richardson, and Martin Stute 2010 Groundwater Reorganization in the Floridan Aquife r Following Holocene Sea Level Rise. Nature Geoscience 3:683 687.
372 Motz, Louis H., and Ahmet Dogan 2004 North Central Florida Active Water Table Regional Groundwater Flow Model . Special Publication SJ2005 SP16. St. Johns River Water Management District, P alatka, Florida. Munch, Douglas A. David J. Toth, Ching tzu Huang, Jeffery B. Davis, Carlos M. Fortich, William L. Osburn , Edward J. Phlips, Erin L. Quinlan, Michael S. Allen, Melissa J. Woods, Patrick Cooney, Robert L. Knight, Ronald A. Clarke, and Scott L. Knight 2006 Fifty Year Retrospective Study of the Ecology of Silver Springs, Florida . Special Publication SJ2007 SP4. St. Johns River Water Management District, Palatka, Florida. Nash, Ronald J. 1997 Archetypal Landscapes and the Interpretation of M eaning. Cambridge Archaeological Journal 7:57 69. Neeley, Michelle 2015 . Electronic document, http://www.wuft.org/news/2015/10/01/art exhibit featuring floridas springs opens at thomas center/ , accessed October 9, 2015. Neill Wilfred T. 1958 A Stratified Early Site at Silver Springs, Florida. The Florida Anthropologist 11:33 52. 1964 Association of Suwannee Points and Extinct Animals. The Florida Anthropologist . 17:17 32. Norman, Robert 2010 Images of America: Ocala National Forest . Arcadia Publishing, Charleston. Ocala Star Banner 2015 Editorial: Science Enough? Electro nic document, http://www.ocala.com/article/20150913/OPINION01/150919948 , accessed October 2, 2015. 2010 Shell Springs Eternal. Paper presented at the 67th An nual Meeting of the Southeastern Archaeological Conference, Lexington, Kentucky. 2013 Conspicuous in their Absence: Shell Sites and Freshwater Springs in the St. Johns River Valley . Paper presented at the 70th Annual Meeting of the Southeastern Archaeolo gical Conference, Tampa, Florida.
373 2015 Beyond the Event Horizon: Moments of Consequence (?) in the St. Johns River Valley. In The Archaeology of Events: Cultural Change and Continuity in the Pre Columbian Southeast , edited by Zackary I. Gilmore and Jason M. 61. University of Alabama Press, Tuscaloosa O'Donoughue, Jason M., and Kenneth E. Sassaman 2013 Phase I Archaeological Survey of Weeki Wachee Springs State Park, Hernando County, Florida . Technical Report 18. Laboratory of Southea stern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2014 Cultural Resources Assessment Survey within Otter Spring Park, Gilchrist County, Florida . Technical Report 1 9 . Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. O'Donoughue, Jason M., Kenneth E. Sassaman, Meggan E. Blessing, Johanna B. Talcott, and Julie C. Byrd 2011 Archaeological Investigations at Salt Springs (8MR2322), Marion County, Florida . Technical Report 11. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Odum, Howard T. 1957a Primary Production Measurements in Eleven Florida Springs and a Marine Turtle Grass Community . Limnology and Oceanography 2 :85 97. 1957b Trophic Structure and Productivity of Silver Springs, Florida. Ecological Monographs 27:55 112. Olsen, B j Ã¸ rnar 2003 Material Culture After Text: Re membering T hings. Norwegian Archaeological Review 36: 87 104. 2010 In Defense of Things . Alta Mira, Walnut Creek, Calif ornia . Olsen, B j Ã¸ rnar, Michael Shanks, Timothy Webmoor, and Christopher Witmore 2012 Archaeology: The Discipline of Things . University of California Press, Berkeley. Olwig, Kenneth 1993 Sexual Cosmology: Nation and Landscape at the Conceptual Interstices of Nature and Culture; or What Does Landscape Really Mean? In Landscape: Politics and Perspective , edited by Barbara Bender, pp. 307 343. Berg, Oxford. Ortmann, Anthony L., and Tristram R. Kidder 2013 Building Mound A at Poverty Point, Louisiana: Monumental Public Architecture, Ritual Practice, and Implications for Hunter Gatherer Complexity. Geoarchaeology 28 :66 86.
374 Otvos, Ervin G. 2004 Holocene Gulf Levels: Recognition Issues and an Updated Sea Level Curve. Journal of Coastal Research 20:680 699. Pandion Systems, Inc. 2003 Carrying Capacity Study of Silver Glen Spring and Run . DEP Contract Number SL 982. Submitted to the Florida Department of Environmental Protection. Copies availab le from Florida Department of Environmental Protection, Tallahassee. Peek, Harry M. 1951 Cessation of Flow of Kissengen Spring in Polk County, Florida . Report of Investigations 7. Florida Geological Survey, Tallahassee. Phelps, G.G. 2004 Chemistry of G round Water in the Silver Springs Basin, Florida, with an Emphasis on Nitrate . Scientific Investigations Report 2004 5144. U.S. Geological Survey, Reston, Virginia. Phelps, G.G., and K.P. Rohrer 1987 Hydrogeology in the Area of a Freshwater Lens in the F loridan Aquifer System, Northeast Seminole County, Florida . Water Resources Investigation Report 86 4097. U.S. Geological Survey, Washington, D.C. Piatek, Bruce J. 1994 The Tomoka Mound Complex in Northeast Florida. Southeastern Archaeology 13:109 118. Pielke , R. A. , Sr., R.L. Walko, L. T. Steyaert, P. L. Vidale, G.E. Liston, W. A. Lyons, and T.N. Chase 1999 The Influence of Anthropogenic Landscape Changes on Weather in South Florida. Mon thly Wea ther Rev iew 127 :1663 1673. Pirkle, William A. 1971 The Offset Course of the St. Johns River, Florida. Southeastern Geology 13:39 59. Pittman, Craig 2012 a Florida's Vanishing Springs. Electronic document, htt p://www.tampabay.com/news/environment/water/floridas vanishing springs/1262988 , accessed August 20, 2013. 2012b Scientists Puzzled by Silver Glen's Mystery A lgae . Electronic document, http://www.tampabay.com/news/environment/water/scientists puzzled by silver glens mystery algae/1262974 , accessed April 23, 2013.
375 Plummer, L. Niel 1993 Stable Isotope Enrichment in Paleowaters of the Sout heast Atlantic Coastal Plain, United States. Science 262:2016 2020. Pope, Alexander 1848 An Essay on Man in Four Epistles . Merriam, Moore & Co., Troy, New York. Porter, Kevin M. 2009 Possible Dugout Canoe (8MR3554) at Silver Glen Springs Recreational Area, Marion County, Florida. Florida Bureau of Archaeological Research. Submitted to Florida Master Site Files. Copies available from Florida Master Site Files, Tallahassee. Potter, Ald en L. 1935 The Remains at Silver Glenn Springs. In Some Further Papers on Aboriginal Man in the Neighborhood of the Ocala National Forest , edited by A. E. Abshire, Alden L. Potter, Allen R. Taylor, Clyde H. Neil, Walter H. Anderson, John I. Rutledge and S tevenson B. Johnson, pp. 13 14. Civilian Conservation Corps, Company 1420, Ocala Camp, Florida. Prufer, Keith M., and James E. Brady (editors) 2005 Stone Houses and Earth Lords: Maya Religion in Cave Context . University Press of Colorado, Boulder. Purdy, Barbara A. 1991 The Art and Archaeology of Florida's Wetlands . CRC Press, Boca Raton. Quillen, Amanda K., Evelyn E. Gaiser, and Eric C. Grimm 2013 Diatom Based Paleolimnological Reconstruction of Regional Climate and Local Land Use Change from a Protected Sinkhole Lake in Southern Florida, USA. Journal of Paleolimnology 49:15 30. Quinn, Rhonda L., Bryan D. Tucker, and John Krigbaum 2008 Diet and M obility in Middle Archaic Florida: Stable Isotopic and Faunal Evidence from the Harris Creek Archaeo logical Site (8VO 24), Tick Island. Journal of Archaeological Science 35 : 2346 2356. Randall, Asa R. 2007 St. Johns Archaeological Field School 2005: Hontoon Island State Park. Technical Report 7 . Laboratory of Southeastern Archaeology, Department of Anthr opology, University of Florida, Gainesville.
376 2010 Remapping Histories: Preceramic Archaic Community Construction along the Middle St. Johns River, Florida. Unpublished Ph.D. dissertation, Department of Anthropology, University of Florida, Gainesville. 2 011 Remapping Archaic Social Histories along the St. Johns River, Florida. In Hunter Gatherer Archaeology as Historical Process , edited by K enneth E. Sassaman and D onald H. Holly Jr., pp. 120 142. University of Arizona Press, Tucson. 2013 The Chronology and History of Mount Taylor Period (ca. 7400 4600 cal B.P.) Shell Sites on the Middle St. Johns River, Florida. Southeastern Archaeology 32:193 217. 2014a Freshwater Shellfishing 9,000 Years Ago in Northeast Florida. Paper presented at the Paper presented at the 71st Annual Meeting of the Southeastern Archaeological Conference, Columbia, South Carolina. 2014b LiDAR aided Reconnaissance and Reconstruction of Lost Landscapes: An Example of Freshwater Shell Mounds (ca. 7500 500 cal b.p.) in Northeast Florida. Journal of Field Archaeology 39(2):162 179. 2015 Constructing Histories: Archaic Freshwater Shell Mounds and Social Landscapes of the St. Johns River, Florida . University Press of Florida, Gainesville. Randall, Asa R., Meggan E. Bless ing, and Jon C. Endonino 2011 Cultural Resource Assessment Survey of Silver Glen Springs Recreational Area in the Ocala National Forest, Florida . Technical Report 13. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florid a, Gainesville. Randall, Asa R. and Kenneth E. Sassaman 2005 St. Johns Archaeological Field School 2003 2004: Hontoon Island State Park. Technical Report 6 . Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2010 (E)mergent Complexities during t he Archaic in Northeast Florida . In Ancient Comple xities: New Perspectives in Prec olumbian North America , e dited by S usan M . Alt, pp. 8 31. University of Utah Press, Salt Lake City. 2012 2012 Field Scho ol Summaries: St. Johns Archaeological Field School: Silver Glen Springs Run. The Florida Anthropologist 65:248 250.
377 Randall, Asa R., Kenneth E. Sassaman, Zackary I. Gilmore, Meggan E. Bles sing, and Jason M. O'Donoughue. 2014 New Histories of Pre Columbian Florida , edited by N eill J. Wallis and A sa R. Randall, pp. 18 37. University Press of Florida, Gainesville. Randall, Asa R., and Bryan Tucker 2012 A Mount Taylor Period Radiocarbon Assay from the Bluffton Burial Mound (8VO23). The Florida Anthropologist 65:219 225. Randazzo, Anthony F. 1997 The Sedimentary Platform of Florida: Mesozoic to Cenozoic. In The Geology of Florida , edited by Anthony F. Randazzo and Douglas S. Jones, pp. 39 58. University Press of Florida, Gainesville. Rattue, James 1995 The Living Stream: Holy Wells in Historical Context . Boydell Press, Woodbridge, Suffolk. Rawlings, Marjorie Kinnan 1938 The Yearling . Collier MacMillan, New Y ork. Reese, Ronald S. and Emily Richardson 2008 Synthesis of the Hydrogeologic Framework of the Floridan Aquifer System and Delineation of a Major Avon Park Permeable Zone in Central and Southern Florida . Scientific Investigations Report 2007 5207. U.S. Geological Survey, Washington, D.C. Rink, W. Jack, James S. Dunbar, and Kevin E. Burdette 2012 The Wakulla Springs Lodge Site (8WA329): 2008 Excavations and New OSL Dating Evidence. The Florida Anthropologist 65:5 22. Rink, W.J., J.S. Dunbar, Glen H. Doran, Charles Frederick, and Brittney Gregory 2012 Geoarchaeological Investigations and OSL Dating Evidence in an Archaic and Paleoindian Context at the Helen Blazes Site (8BR27), Brevard County, Florida. The Florida Anthropologist 65:85 105. Robison, Jim 1987 A Sinkhole Chronology. Electronic document, http://articles.orlandosentinel.com/1987 12 27/news/0170120189_1_sinkhole winter park mae r ose , accessed July 21, 2015.
378 Rosenau, Jack C., Glen L. Faulkner, Charles W. Hendry, Jr., and Robert W. Hull 1977 Springs of Florida . Bulletin 31 (Revised). Bureau of Geology, Tallahassee, Florida. Rosenswig, Robert M. 2015 A Mosaic of Adaptation: The Archaic Period. Journal of Archaeological Research 23:115 162. Rupert, Frank R. 1988 The Geology of Wakulla Springs . Open File Report 22. Florida Geological Survey, Tallahassee. Russo, Michael 2004 Measuring She ll Rings for Social Inequality. In Signs of Power: The Rise of Cultural Complexity in the Southeast , edited by Jon L. Gibson and Philip J. Carr, pp. 26 70. University of Alabama Press, Tuscaloosa. Sahlins, Marshall 1985 Islands of History . The University of Chicago Press, Chicago. 1991 The Return of the Event, Again: With Reflections on the Beginnings of the Great Fijian War of 1843 to 1855 between the Kingdoms of Bau and Rewa. In Clio in Oceania: Toward a Historical Anthropology , edited by Aletta Biers ack, pp. 37 99. Smithsonian Institution Press, Washington, D.C. St. Johns River Water Management District (SJRWMD) 2015 a GIS Download Library. Electronic document, http://floridaswa ter.com/gisdevelopment/docs/themes.html , accessed July 1, 2015. 2015b Springs of the District. Electronic document, http://floridaswater.com/springs/springslist.html , accessed September 17, 2015. Samek, Kelly Journal of Land Use & Environmental Law 19:569 595. Santos Granero, Fernando 1998 Writing History into the Landscape: Space, Myth, and Ritual in Contemporar y Amazonia. American Ethnologist 25(2):128 148. Sassaman, Kenneth E. 1993 Early Pottery in the Southeast: Tradition and Innovation in Cooking Technology . University of Alabama Press, Tuscaloosa.
379 2003a St. Johns Archaeological Field School 2000 2001: Bl ue Spring and Hontoon Island State Parks . Technical Report 4. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2003b Crescent Lake Archaeological Survey, 2002: Putnam and Flagler Counties, Florida . T echnical Report 5. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. 2003c New AMS Dates on Orange Fiber Tempered Pottery from the Middle St. Johns Valley and Their Implications for Culture History in Northeast Florida. The Florida Anthropologist 56:5 13. 2004 Common Origins and Divergent Histories in the Early Pottery Traditions of the American Southeast. In Early Pottery: Technology, Function, Style, and Int eraction in the Lower Southeast , edited by Rebecca Saunders and Christopher T. Hays, pp. 23 39. University of Alabama Press, Tuscaloosa. 2010 The Eastern Archaic, Historicized . AltaMira Press, Lanham, Maryland. 2012 Futurologists Look Back . Archaeologies 8:250 268. 2013 Drowning out the Past: How Humans Historicize Water as Water Historicizes Them. In Big Histories, Human Lives: Tackling Problems of Scale in Archaeology , edited by John Robb and Timothy R. Pauketat, pp. 171 192. School for A dvanced Research Press, Santa Fe, New Mexico. Sassaman, Kenneth E., Zackary I. Gilmore, and Asa R. Randall 2011 St. Johns Archaeological Field School 2007 10: Silver Glen Run (8LA1) . Technical Report 12. Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida, Gainesville. Sassamann, Kenneth E., and Asa R. Randall 2012 Shell Mounds of the Middle St. Johns Basin, Northeast Florida. In The Origins of New World Monumentality , edited by Richard L. Burger and Robert M. Rosen swig, pp. 53 77. University Press of Florida, Gainesville. Saunders, Joe W., Rolfe D. Mandel, C. Garth Sampson, Charles M. Allen, E. Thurman Allen, Daniel A. Bush, James K. Feathers, Kristen J. Gremillion, C. T. Hallmark, H. Edwin Jackson, Jay K. Johnson, Reca Jones, Roger T. Saucier, Gary L. Stringer, and Malcolm F. Vidrine 2005 Watson Brake, a Middle Archaic Mound Complex in Northeast Louisiana. American Antiquity 70:631 668 Saunders, Rebecca, and Michael Russo 2011 Coastal Shell Middens in Florida: A View from the Archaic period. Quaternary International 239: 38 50.
380 Scherer, Reed P. 1988 Freshwater Diatom Assemblages and Ecology/Paleoecology of the Okefenokee Swamp/Marsh Complex, Southern Georgia, U.S.A. Diatom Research 3:129 157. Schiffer, Donna M . 1998 Hydrology of Central Florida Lakes A Primer . Circular 1137. U.S. Geological Survey, Reston, Virginia. Schmidt, Walter 1997 Geomorphology and Physiography of Florida. In The Geology of Florida , edited by Anthony F. Randazzo and Douglas S. Jones, pp. 1 12. University Press of Florida, Gainesville. Schofield, Kate 2013 Fla. Museum Opens New Exhibits Saturday H ighlightin Cultural H istory . Electronic document, https://www.flmnh.ufl.edu/pressroom/2013/03/19/fla museum opens new exhibits saturday highlighting states springs cultural history/ , accessed October 9, 2015. Scott, Thomas M. 198 3 The Hawthorn Formation of Northeast Florida, Part I: The Geology of the Hawthorn Formation of Northeast Florida . Report of Investigations 94. Florida Geological Survey, Tallahassee. 1988 The Lithostratigraphy of the Hawthorn Group (Miocene) of Florida . Bulletin 59. Florida Geological Survey, Tallahassee 1997 Miocene to Holocene History of Florida. In The Geology of Florida , edited by Anthony F. Randazzo and Douglas S. Jones, pp. 57 69. University Press of Florida, Gainesville. 2011 Geology of the Fl orida Platform Pre Mesozoic to Recent. In Gulf of Mexico: Origin, Waters, and Biota , Vol. 3: Geology, edited by Noreen A. Buster and Charles W. Holmes, pp. 17 31. Texas A&M University Press, College Station. Scott, Thomas M., Guy H. Means, Ryan C. Means, and Rebecca P. Morgan 2002 First Magnitude Springs of Florida . Open File Report 85. Florida Geological Survey, Tallahassee. Scott, Thomas M., Guy H. Means, Rebecca P. Meegan, Ryan C. Means, Sam B. Upchurch, R. E. Copeland, J. Jones, Tina Roberts, and Ala n Willet 2004 Springs of Florida . Bulletin 66. Florida Geological Survey, Tallahassee.
381 Screaton, E., J. B. Martin, B. Ginn, and L. Smith 2004 Conduit Properties and Karstification in the Unconfined Floridan Aquifer. Ground Water 42: 338 346. Sears, Wil liam H. 1960 The Bluffton Burial Mound. The Florida Anthropologist 13:55 60. Seinfeld, Daniel M. 2013 An Assessment of Submerged Archaeological Deposits and Evidence for Looting at the Silver Glen Springs Run Site (8MR3605/8LA1) . Report on File, Florida Bureau of Archaeological Research, Tallahassee. Sewell, William H., Jr. 2005 Logics of History: Social Theory and Social Transformation . University of Chicago Press, Chicago. Shockman, Elizabeth 2015 Florida's Natural Springs are C hanging And D isappearing . Electronic document, http://www.pri.org/stories/2015 09 05/floridas natural springs are changing and disappearing , accessed October 7, 2015. Louis H. Motz, and Qing Sun 2004 Comparison of Estimated Areas Contributing Recharge to Selected Springs in North Central Florida by Using Multiple Ground Water Flo w Models . Open File Report 03 448. U.S. Geological Survey, Tallahassee, Florida. Siddall, M., E.J. Rohling, A. Almogi Labin, C. Hemleben, D. Meischner, I. Schmelzer, and D.A. Smeed 2003 Sea Level Fluctuations during the Last Glacial Cycle. Nature 423:853 858. Simms, Alexander R., Kurt Lambeck, Anthony Purcell, John B. Anderson, and Antonio B. Rodriguez 2007 Sea Level History of the Gulf of Mexico since the Last Glacial Maximum with Implications for the Melting History of the Laurentide Ice Sheet. Quater nary Science Reviews , 26:920 940. Skibo, James M., and Michael Brian Schiffer 2008 People and Things: A Behavioral Approach to Material Culture . Springer, New York. Smith, Bruce D. 1986 Archaeology of the Southeastern United States: from Dalton To DeSoto 10,500 500 BP . In Advances in World Archaeology , Volume 5 , edited by F red Wendorf and A ngela E. Close, pp. 1 92. Academic Press. New York.
382 Smith, D.E., S. Harrison, C.R. Firth, and J.T. Jordan 2011 The Early Holocene Sea Level Rise. Quaternary Sci ence Reviews 30:1846 1860. Smith, Douglas L., and Kenneth M. Lord 1997 Tectonic Evolution and Geophysics of the Florida Basement. In The Geology of Florida , edited by Anthony F. Randazzo and Douglas S. Jones, pp. 13 26. University Press of Florida, Gainesville. Smock, Leonard A., Anne B. Wright, and Arthur C. Benke 2005 Atlantic Coastal Rivers of the Southeastern United States. In Rivers of North America , edited by Arthur C. Benke and Colbert F. Cushing, pp. 73 124. Elsevier Academic Press, Burling ton, Massachusetts. Sommer, Eleanor K. 2012 Fountains of Life: A Look at Florida Springs from Sacred Waters to Green Slime. University of Florida College of Journalism and Communications. Electronic document, http://stateofwater.org/ecosystems/springs/ , accessed October 7, 2015. Southeastern Archaeological Research, Inc. 2008 Cultural Resource Assessment Survey of State Road 40 from One Mile West of State Road 326 to State Road 15 (US 17) PD&E Study, Marion, Lake, and Volusia Counties, Florida . Submitted to Inwood Consulting Engineers, Project No. 2018 05177. Copies Available from SEARCH, Inc., Jonesville, FL. Southwest Florida Water Management District (SWFWMD) 2009 Weeki Wachee Springs Resto ration Project . Electronic document, http://www.swfwmd.state.fl.us/springs/weeki wachee/restoration project.php , accessed October 6, 2015. Spechler, Rick M. 1994 S altwater Intrusion Quality of Water in Floridan Aquifer System, Northeastern Florida . Water Resources Investigation Report 92 4174. U.S. Geological Survey, Tallahassee. Spicuzza, Mary 2005 Restoration of Sea Wall Under W ay . Electronic document, http://www.sptimes.com/2005/09/24/news_pf/Hernando/Restoration_of_sea_wa. shtml , accessed October 6, 2015.
383 Spivey, S. Margaret, Spivey, S. Margaret, Tristam. R. Kidder, Anthony L. Ortmann, and Lee Arco 2015 Pilgrimage to Poverty Point? In The Archaeology of Events: Cultural Change and Continuity in the Pre Columbian Southeast, edited by Zackary I. Gilmore and Jason M. O'Donoughue, pp. 141 159. University of Alab ama Press, Tuscaloosa. Springs Eternal Project 2013 Why We Love this Spring Silver Glen Springs. Electronic document, http://springseternalproject.org/springs/silver glen/why we love this spring silver glen/ , accessed September 15, 2015. Stanton, William M. 1995 Archaic Subsistence in the Middle St. Johns River Valley: Silver Glen Springs and the Mt. Taylor Period. Unpu blished Master s thesis, Department of Anthropology, Florida State University, Tallahassee. Steponaitis, Vincas P. 1986 Prehistoric Archaeology in the S outheastern United States, 1970 1985. Annual Review of Anthropology 15:363 404. Stevenson, R. Jan, Ag nieszka Pinowska, Andrea Albertin, James O. Sickman 2007 Ecological Condition of Algae and Nutrients in Florida Springs . DEP Grant Number S0291 . Report submitted to the Florida Department of Environmental Protection , Tallahassee. Sterling, Maurice, and C .A. Padera 1998 The Upper St. Johns River Project The Environmental Transformation of a Public Flood Control Project . Professional Paper SJ98 PP1. St. Johns River Water Management District, Palatka, Florida. Stewart J oseph , Peter Sucsy , and John Hendrickson 2006 Meteorological and Subsurface Factors Affecting Estuarine Conditions within Lake George in the St. Johns River, Florida. Proceedings of the Seventh International Conference on Hydroscience and Engineering, Philadelphia, PA. Strang, Veron ica 2005 Common Senses : Water, Sensory Experience and the Generation of Meaning. Journal of Material Culture 10 :92 120. 2008 The Social Construction of Water. In Handbook of Landscape Archaeology , edited by Bruno David and Julian Thomas, pp. 123 130. Left Coast Press, Walnut Creek, California. Strathern, Marilyn 1988 The Gender of the Gift: Problems with Women and Problems with Society in Melanesia . University of California Press, Berkeley
384 Stringfield, V.T., and H.H. Cooper 1951 Geologic and Hydrol ogic Features of an Artesian Submarine Spring East of Florida . Report of Investigations 7. Florida Geological Survey, Tallahassee. Strong, William A. 2004. Temporal Water Chemistry Trends within Individual Springs and within a Population of Florida Spring Gainesville. Sullivan, Alan P. , III , and Kenneth C. Rozen 1985 Debitage Analysis and Archaeological I nterpr etation. American Antiquity 50 :755 779. Swarenski, P.W., C.D Reich, R.M Spechler, J.L Kindi nger, and W.S Moore 2001 Using Multiple Geochemical Tracers to Characterize the Hydrogeology of the Submarine Spring off Crescent Beach, Florida. Chemical Geology 179:187 202. Swihart, Tom 2013 Four Questions for Jim Stevenson . Electronic document, http://www.wateryfoundation.com/?p=8887 , accessed October 2, 2015. TaÃ§on, Paul S. 1999 Identifying Ancient Sacred Landscapes in Australia: From Physical to Social. In Archaeologies of Landscape: Contemporary Perspectives , edited by Wendy Ashmore and A. Bernard Knapp, pp. 33 57. Blackwell, Malden, Massachusetts. Talcott, Johanna B. 2011 Paleoethnobotanical Assemblage. In Archaeological Investigations at Salt Springs (8MR2322), Marion County, Flor ida , edited by Jason M. O'Donoughue, Kenneth E. Sassaman, Meggan E. Blessing, Johanna B. Talcott, and Julie C. Byrd , pp. 85 104 . Technical Report 12, Laboratory of Southeastern Archaeology, Department of Anthropology, University of Florida. Tampa Bay Time s http://www.tampabay.com/specials/2012/reports/florida springs/ , accessed October 9 , 2015. Tarasov, Lev, and W.R. Pelt ier 2005 Arctic Freshwater Forcing of the Younger Dryas Cold Reversal. Nature 435:662 665.
385 Tesar, Louis D., and Calvin B. Jones 2004 Wakulla Springs Lodge Site (8WA329) in Edward Ball Wakulla Springs State Park, Wakulla County, Florida . Florida Burea u of Archaeological Research, Tallahassee. Theodoratus, Dorothea J., and Frank LaPena. 1994 Wintu Sacred Geography of Northern California. In Sacred Sites, Sacred Places , edited by David L. Carmichael, Jane Hubert, Brian Reeves, and Audchild Schanche, pp. 20 31. Routledge, New York. Thomas, Julian 1999 An Economy of Substances in Earlier Neolithic Britain. In Material Symbols: Culture and Economy in Prehistory , edited by John E. Robb, pp. 70 89. Center for Archaeological Investigations, Occasional Pap er 26. Southern Illinois University, Carbondale. Thompson 1964 Archaeological Site Form, 8MR83. Report on file, Florida Master Site Files, Tallahassee. Thompson, Bill 2013 a Silver Springs' Unveiling as State Park S et Tuesday . Electronic document, http://www.ocala.com/article/20130930/ARTICLES/130939978 , accessed October 6, 2015. 2013b Treasure Trove of Artifacts F ound at Silver Springs . Electronic document, http://www.ocala.com/article/20130812/ARTICLES/130819934 , accessed October 9, 2015. Thulman, David K. 2009 Freshwater Availability as the Constraining Factor in the Middle Paleoindia n Occupation of North Central Florida. Geoarchaeology 24:243 276. 2012 Paleoindian Occupations along the St. Johns River, Florida. The Florida Anthropologist 65:79 83. Tippett, Krista 2015 T The Inner Landscape of Beauty. Electronic document, http://www.onbeing.org/program/john odonohue the inner landscape of beauty/transcript/7801 , accessed September 7, 2015. Tolbert, Margaret Ross 2010 AQUIFERious . Fidelity Press, Addison, T exas.
386 TÃ¶rnqvist, TorbjÃ¶rn E., Juan L. GonzÃ¡lez, Lee A. Newsom, Klaas van der Borg, Arie F.M. de Jong, and Charles W. Kurnik 2004 Dec iphering Holocene Sea Level History on the US Gulf Coast: A High Resolution Record from the Mississippi Delta. Geological Society of America Bulletin 116:1026 1039. Toscano, Marguerite A., and Joyce Lundberg 1999 Submerged Late Pleistocene Reefs on the T ectonically Stable SE Florida Margin: High Precision Geochronology, Stratigraphy, Resolution of Substage 5a Sea Level Elevation, and Orbital Forcing. Quaternary Science Reviews 18:753 767. Toscano, M.A., and I.G. Macintyre 2003 Corrected Western Atlantic Sea Level Curve for the Last 11,000 Years Based on Calibrated 14 C dates from Acropora palmata Framework and Intertidal Mangrove Peat. Coral Reefs 22:257 270. Toth, David J., and Brian G. Katz 2006 Mixing of Shallow and Deep Gro undwater as Indicated by the Chemistry and Age of Karstic Springs. Hydrogeology Journal 14:827 847. Tucker, Bryan D. 2009 Isotopic Investigations of Archaic Period Subsistence and Settlement in the St. Johns River Drainage, Florida. Unpublished Ph.D. dis sertation, Department of Anthropology, University of Florida, Gainesville. Turner, Ernest 1989 The Souls of my Dead Brothers. In Conflict in the Archaeology of Living Traditions , edited by Robert Layton, pp. 189 194. Unwin Hyman, Boston. Upchurch, Sam B ., Richard N. Strom, and Mark G. Nuckels. 1982 Methods of Provenance Determination of Florida C herts. Department of Geology, University of So uth Florida , Tampa. Submitted to the Florida Division of Archives, History, and Records Management, Bureau of Historic Sites and Properties, Tallahassee. U.S. Environmental Protection Agency 2002 A Lexicon of Cave and Karst Terminology with Special Reference to Environmental Karst Hydrology . U.S. Environmental Protection Agency, Washington, D.C. Van Soelen, E. E., G. R. Brooks, R. A. Larson, JS Sinninghe DamstÃ©, and G. J. Reichart 2012 Mid to Late Holocene Coastal Environmental Changes in S outhwest Florida, USA. The Holocene 22: 929 938.
387 Waters, Michael R., and Thomas W. Stafford 2007 Redefining the Age of Clovis: Implications for the P eopling of the Americas. Science 315:1122 1126. Watkins, Tom 2013 In Florida, A Spring Cleanup Yields Cornucopia of H istory . Electronic document, http://www.cnn.com/2013/11/16/us/florida spring artifacts/index.html , accessed October 9, 2015. Watts, William A. 1969 A Pollen Diagram from Mud Lake, Marion County, North Central Florida. Geological Society of America Bulletin 80:631 642. 1971 Postglacial and Interglacial Vegetation History of Southern Georgia and Central Florida. Ecology 52:676 690. 1975 A Late Quaternary Record of Vegetation from Lake Annie, South Central Florida. Geology 3:344 346. 1980 The Late Quaternary Vegetation History of the Southeastern United States. Annual Review of Ecology and Systematics 11:387 409. Watts, William A., Eric C. Grimm and T. C. Hussey 1996 Mid Holocene Forest History of Florida and the Coastal Plain of Georgia and South Carolina. In Archaeology of the Mid Holocene Southeast , edited by Kenneth E. Sassaman and David G. Anderson, pp. 28 40. University Press of Florida, Gainesville. Watts, William A. and Barbara C.S. Hansen 1988 Environments of Florida in the Lat e Wisconsin and Holocene. In Wet Site Archaeology , edited by Barbara A. Purdy, pp. 307 323. The Telford Press, Caldwell, New Jersey. 1994 Pre Holocene and Holocene Pollen Records of Vegetation History from the Florida Peninsula and their Climatic Implica tions. Palaeogeography, Palaeoclimatology, Palaeoecology 109:163 176. Watts, William A., Barbara C.S. Hansen and Eric C. Grimm 1992 Camel Lake: A 40 000 yr Record of Vegetational and Forest History from Northwest Florida. Ecology : 731056 1066. Watts William A., and M. Stuiver 1980 Late Wisconsin Climate of Northern Florida and the Origin of Species Rich Deciduous Forest. Science 210:325 327
388 Webb, S. David (editor) 2006 First Floridians and Last Mastodons : The Page Ladson S ite in the Aucilla River . Springer, New York. Weber, Kenneth A. and Robert. G. Perry 2006 Groundwater Abstraction Impacts on Spring Flow and Base F low in the Hillsborough River Basin, Florida, USA. Hydrogeology Journal 14:1252 1264. Webmoor, Timothy, and Christopher Witmore 2008 Things are Us! A Commentary on Human/Things Relations Under the Banner rchaeology. Norwegian Archaeological Review 41 :53 70. Weiner, Annette B. 1992 Inalienable Possessions: The Paradox of Keeping While Giving . University of California P ress, Berkeley. Wharton, Barry, George Ballo and Mitchell Hope 1981 The Republic Groves Site, Hardee County, Florida. The Florida Anthropologist 34:59 80. Wheeler, Ryan J. 2001 8MR3173: FBAR Canoe and Log Boat Recording Form. Manuscript on file, Florid a Master Site Files, Tallahassee. Wheeler, Ryan J., James J. Miller, Ray M. McGee, Donna Ruhl, B renda Swann, and Melissa Memory 2003 Archaic Period C an oes from Newnans Lake, Florida. American Antiquity ( 68 ):533 551. Wheeler, Ryan J., Christine L. Newman and Ray M. McGee 2000 A New Look at the Mount Taylor and Bluffton Sites, Volusia County, with an Outline of the Mount Taylor Culture. The Florida Anthropologist 53:133 157. White, Devin A., and Sarah L. Surface Evans (editors) 2012 Least Cost Analysis of Social Landscapes: Archaeological Case Studies . University of Utah Press, Salt Lake City . White, William A. 1970 The Geomorphology of the Florida Peninsula . Geological Bulletin 51. Florida Bureau of Geology, Tallahassee. White, William B. 2002 Karst Hydrology: Recent Developments and Open Questions. Engineering Geology 65(2):85 105.
389 Willard, Debra A., and Christopher E. Bernhardt 2011 Impacts of Past Climate and Sea Level Change on Everglades Wetlands: Placing a Century of Anthropogenic Change int o a Late Holocene Context. Climatic Change 107:59 80. Willard, Debra A., Christopher E. Bernhardt, Gregg R. Brooks, Thomas M. Cronin, Terence Edgar, and Rebekka Larson 2007 Deglacial Climate Variability in Central Florida, USA. Palaeogeography, Palaeocli matology, Palaeoecology 251:366 382. Williams, Lester J., and Eve L. Kuniansky 2015 Revised Hydrogeologic Framework of the Floridan Aquifer System in Florida and Parts of Georgia, Alabama, and South Carolina . Professional Paper 1807. U.S. Geological Surv ey, Reston, Virginia. Williams, Stanley A. 2006 Simulation of the Effects of Groundwater Withdrawals from the Floridan Aquifer System in Volusia County and Vicinity . Technical Publication SJ2006 4. St. Johns River Water Management District. Palatka, Flor ida. Willis, Ray mond F. 1972 8MR123: Silver Glen Springs. Archaeological Site Form. Manuscript on file, Florida Master Site Files, Tallahassee. 1995 Alexander Springs Beach Reconstruction, Seminole Ranger District, Ocala National Forest, Lake County . FY 1995 Heritage Resources Status Report 10 . U.S. Forest Service, Tallahassee. Winnemem Wintu Tribe 2015 Mt. Shasta: Desecration of Panther Spring, Our Genesis Place. Electronic document, http://w ww.winnememwintu.us/mount shasta/ , accessed September 7, 2015. Woodfill, Brent K.S., Jon Spenard, and Megan Parker 2015 Caves, Hills, and Caches: The Importance of Karst Landscapes for the Prehispanic and Contemporary Maya . Geological Society of America Special Papers 516, in press , d oi:10.1130/2015.2516(16) . Wyman, Jeffries 1875 Fresh Water Shell Mounds of the St. Johns River, Florida . Memoir 4. Peabody Academy of Science, Salem, Massachusetts. 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 Mixed Siliciclastic Carbonate, Deltaic Coastline: Suwannee River, Florida, U.S.A. Journal of Sedimentary Research 75:621 635.
390 Zarikian, Carlos A.Alvarez, Peter K. Swart, John A. Gifford, and Patricia L. Blackwelder 2005 Holocene Paleohydrology of Little Salt Spring, Florida, Based on Ostracod Assemblages and Stable Isotopes. Palaeogeography, Palaeoclimatology, Palaeoecology 225:134 156.
391 BIOGRAPHICAL SKETCH began his college education at the University of Florida (UF), where he received a B.A. in a nthropology in 2004. The following year h e enrolled in graduate school at the University of Tennessee in Knoxville, where he received an M.A. i n anthropology . In 2008, Jason returned to UF to pursue a Ph.D. in anthropology , studying the long term s on was a warded a Ph.D. in 2015, and resides in Gainesville, Florida with his wife, Amanda, and two children.